Good audio transformer? Buy used Dynaco on eBay. One is for pair EL84 used in the Dynco ST-35 amp (Z-565), other is for EL34 used in Dynaco ST-70 amp (PA-060). You can use many low power tubes for the EL84 transformer 8K primary impedance and the EL34 transformer. is I believe 6K. 6K is good for 6L6 and similar tubes. Unloading a 6L6 from 4.5K listed tube PP impedance makes for lower distortion at 6K impedance. You can buy new clones from Triode Electronics.
I read an article on audio transformer tests and the Dynaco were by far best on the list of 40+ tested transformers. The site had input and output waveforms. Audio transformers vary a lot and most were poor performers.
I read an article on audio transformer tests and the Dynaco were by far best on the list of 40+ tested transformers. The site had input and output waveforms. Audio transformers vary a lot and most were poor performers.
Tubesaregreat,
Optimationman is looking for Single Ended output transformers.
You suggested the output transformers from Dyna ST -35 and ST-70.
They are Push Pull transformers, and will not work for Single Ended circuits.
At risk of ruining the Dyna transformers, I did something to them.
See my Post # 16 in this thread.
I was able to pull the end bells off the transformers, and see that my particular ones, had Wooden Shims that held the laminations in place inside the coil assembly.
Removing the laminations from some transformers has been known to scratch and ruin the coil assembly.
There are lots of good single ended output transformers out there; damaging a set of push pull transformers is not good; I was very lucky that I did not.
Optimationman is looking for Single Ended output transformers.
You suggested the output transformers from Dyna ST -35 and ST-70.
They are Push Pull transformers, and will not work for Single Ended circuits.
At risk of ruining the Dyna transformers, I did something to them.
See my Post # 16 in this thread.
I was able to pull the end bells off the transformers, and see that my particular ones, had Wooden Shims that held the laminations in place inside the coil assembly.
Removing the laminations from some transformers has been known to scratch and ruin the coil assembly.
There are lots of good single ended output transformers out there; damaging a set of push pull transformers is not good; I was very lucky that I did not.
Both I and Bofufet are interested in what items affect the sound of output transformers. Materials obviously have an important role shaping the characteristics. I've read that also the physical makeup contributes. Such as input impedance, the number of secondary windings, potted, and size, that is 7 vs 30 watt. I'm speculating that each manufacturer has a house sound (correct me). I'm mostly interested in the major sound affecting categories. What can be specified when ordering a set of SE transformers and their contribution to the expected enhancement/detraction to that house sound.
Decware’s new UFO transformers are supposed to be stellar. Based on the Peerless of old, with much interleaving.I’m still looking for more info on what the core is made of, but I hear they go out to 5 times the 20kH rating. Some might say you can’t hear above 20kH, but those frequencies give harmonics to the sounds you can hear. If anybody has more info on the UFO output transformers or the peerless, in regards to their build, please post it!
I don't have any experience with Decware transformers, but I've yet to see an output transformer that's flat much past 20kHz without feedback. That's not to say they don't exist but I'd be very surprised if any OPT exhibited that much bandwidth without assistance. I've worked with a number of Peerless transformers and the advantage is their good behavior with feedback and the relative lack of extreme high-frequency phase shifts that lead to instability. If I use a Peerless or UTC in a no-feedback amp, it's pretty much guaranteed to start rolling off after 15kHz. I did test some Finemet PP transformers and they were flat to 20kHz without feedback. They were not at all happy with feedback, however, and it was very difficult to achieve good stability. Most of the classic OPTs after 1950 were designed with feedback in mind. As far as I know, the goal wasn't excessive open-loop response but rather a controlled response under feedback conditions.
Generalizations:
If you can apply lots of global negative feedback in a particular specific amplifier,
And if that works and is stable for a number of different output transformer models and manufacturers . . .
Then they all "sound" similar. The difference is how soon they saturate at low frequencies and output power levels.
And, when you reach saturation, that also "sounds" similar, but remember, it happens at different low frequency versus output power according to the model of the transformer.
Global Negative Feedback does not fix saturation, it makes it worse.
$0.03
If you can apply lots of global negative feedback in a particular specific amplifier,
And if that works and is stable for a number of different output transformer models and manufacturers . . .
Then they all "sound" similar. The difference is how soon they saturate at low frequencies and output power levels.
And, when you reach saturation, that also "sounds" similar, but remember, it happens at different low frequency versus output power according to the model of the transformer.
Global Negative Feedback does not fix saturation, it makes it worse.
$0.03
I'm answering for a friend (ChatGPT):
Q: As to tube audio output transformers what are the details which make them excellent ?
A:
The quality of tube audio output transformers is crucial for the performance of tube amplifiers, and several key details contribute to making them excellent. Here are the primary factors:
1. Core Material:
Q: As to tube audio output transformers what are the details which make them excellent ?
A:
The quality of tube audio output transformers is crucial for the performance of tube amplifiers, and several key details contribute to making them excellent. Here are the primary factors:
1. Core Material:
- High-Quality Core Materials: High-permeability materials like silicon steel, nickel, or amorphous cores reduce losses and improve efficiency. The choice of core material affects the transformer's saturation, frequency response, and distortion levels.
- Core Size and Design: Adequate core size prevents saturation at low frequencies and allows for sufficient headroom to handle the amplifier's power output without distortion.
- Precision Winding: Close attention to winding geometry, such as interleaving and bifilar winding, minimizes leakage inductance and maintains consistent impedance across frequencies.
- Layered or Sectional Winding: Helps in reducing parasitic capacitance and leakage inductance, improving high-frequency response and stability.
- Tight Coupling: Ensures efficient transfer of energy between the primary and secondary windings, maintaining the fidelity of the audio signal.
- Proper Impedance Ratios: Accurately matching the impedance of the amplifier's output stage with the speaker load is critical for maximum power transfer and sound quality. Transformers must have the correct turns ratio to ensure this match.
- Wide Bandwidth: High-quality transformers should provide a flat frequency response over the audio spectrum (typically 20 Hz to 20 kHz or beyond) to avoid coloration of the sound.
- Low Frequency Performance: Core design and size should allow the transformer to handle low frequencies without distortion or saturation.
- High Frequency Performance: Proper winding techniques and low parasitic capacitance help maintain high-frequency response without roll-off.
- Minimized Harmonic Distortion: Careful design and high-quality materials reduce harmonic distortion, preserving the amplifier's sound quality.
- Low Core Saturation: Ensuring the core does not saturate at high power levels prevents distortion and maintains sound fidelity.
- Adequate Power Rating: The transformer should be rated to handle the amplifier's maximum output power without overheating or saturating.
- Low Mechanical Vibration: Reducing mechanical vibration and noise ensures that the transformer does not introduce unwanted hum or noise into the audio signal.
- Solid Construction: Sturdy construction with high-quality insulation and secure mounting reduces the risk of mechanical failure and hum.
- Magnetic Shielding: Proper shielding reduces hum and interference from external magnetic fields.
- Orientation and Placement: Correct placement and orientation relative to other components in the amplifier can minimize hum and noise.
- Temperature Handling: High-quality transformers are designed to operate within safe temperature ranges, ensuring reliability and longevity even under continuous high power operation.
- Rigorous Testing: High-end transformers undergo extensive testing for frequency response, distortion, power handling, and overall consistency to ensure they meet the required specifications.
Last edited:
FrankieS,
I Agree!
We can add one more required characteristic of output transformers . . . Low Insertion Loss.
For the following below example, the insertion loss will be calculated at 1kHz where the primary inductance and distributed capacitance are less of a factor. We need to connect a non-inductive load resistor across the secondary.
Here is an example of an output transformer that has a little more insertion loss than we might want it to have, 1 dB.
A 5k primary impedance according to turns ratio to the secondary, but the primary also has 300 Ohms DCR.
An 8 Ohm secondary impedance, according to turns ratio versus the primary; and 0.48 Ohms DCR.
The primary and resistively loaded secondary are essentially at the frequency where inductive reactance and distributed capacitance are not an issue. The input to output phase tends to be near 0 degrees at 1 kHz. That means the "turns impedance" and DCR at 0 degrees makes the impedance simple, it is purely resistive.
There is no need to do a Trig calculation of the total impedance. Example, it is simply 5000 Ohms and 300 Ohms in series, 5300 Ohms.
I have never measured an output transformer that has 0 degrees input to output at less than 500Hz, or more than 2kHz, pick the frequency that your transformer input to output is 0 degrees for this discussion.
The turns are in series with the DCRs, when the input to output phase is 0 degrees.
20 Log (5000/5300) = 0.5dB insertion loss of the primary. It is just simple voltage divider, when phase is 0 degrees.
20 Log (8/8.48) = 0.5dB insertion loss of the secondary. It is just simple voltage divider, when phase is 0 degrees.
We have 0.5dB loss in the primary, and 0.5 dB loss in the secondary. The total insertion loss is at least 1.0 dB (there may be other losses at 1kHz that. have nothing to do with DCR).
Now you know how to measure the minimum insertion loss:
First,
Measure the turns ratio, drive the primary with a low impedance, leave the output open (no load across it); change the frequency until you get 0 degrees from input to output. Then measure the voltage ratio at that frequency.
Voltage Ratio Squared = impedance ratio
25:1 for example. 25 squared is 625. 8 Ohms x 25 = 5000 Ohms.
Calculate the primary and secondary impedance, according to the voltage ratio.
Now, calculate the insertion loss of the primary, and of the secondary, that is caused by the DCRs. Use the example I gave above, just put the proper values in for the transformer you have.
Is 1.0 db insertion loss significant?
Put 12.6 Watts from the vacuum tube(s) into the primary, and all you get out is 10.0 Watts.
You decide if 1.0 dB insertion loss is significant for you.
When the old RCA, GE, Tungsol, etc. listed the power out from the tube, the specification is just that, the power from the tube.
They did not allow for the insertion loss of your transformer, you forgot to tell them that, all the way back in 1936 what your output transformer insertion loss is.
And people wonder why they get less power out than they read in the tube data sheet!
Remember, the actual primary impedance and actual secondary impedance will be higher than just according to the turns ration.
The DCRs increase the impedances of both primary and secondary.
If you forget that fact, you thought you had a 5k primary, but the two DCR losses make it a 5500 Ohm primary; do your tubes "care" about that?
If that is not important to you, then remember, the output transformer insertion loss also reduces the Open Loop Damping Factor of your amplifier.
Did the above factors ever make you wonder why you got the lower performance from your amplifier?
$0.03
I Agree!
We can add one more required characteristic of output transformers . . . Low Insertion Loss.
For the following below example, the insertion loss will be calculated at 1kHz where the primary inductance and distributed capacitance are less of a factor. We need to connect a non-inductive load resistor across the secondary.
Here is an example of an output transformer that has a little more insertion loss than we might want it to have, 1 dB.
A 5k primary impedance according to turns ratio to the secondary, but the primary also has 300 Ohms DCR.
An 8 Ohm secondary impedance, according to turns ratio versus the primary; and 0.48 Ohms DCR.
The primary and resistively loaded secondary are essentially at the frequency where inductive reactance and distributed capacitance are not an issue. The input to output phase tends to be near 0 degrees at 1 kHz. That means the "turns impedance" and DCR at 0 degrees makes the impedance simple, it is purely resistive.
There is no need to do a Trig calculation of the total impedance. Example, it is simply 5000 Ohms and 300 Ohms in series, 5300 Ohms.
I have never measured an output transformer that has 0 degrees input to output at less than 500Hz, or more than 2kHz, pick the frequency that your transformer input to output is 0 degrees for this discussion.
The turns are in series with the DCRs, when the input to output phase is 0 degrees.
20 Log (5000/5300) = 0.5dB insertion loss of the primary. It is just simple voltage divider, when phase is 0 degrees.
20 Log (8/8.48) = 0.5dB insertion loss of the secondary. It is just simple voltage divider, when phase is 0 degrees.
We have 0.5dB loss in the primary, and 0.5 dB loss in the secondary. The total insertion loss is at least 1.0 dB (there may be other losses at 1kHz that. have nothing to do with DCR).
Now you know how to measure the minimum insertion loss:
First,
Measure the turns ratio, drive the primary with a low impedance, leave the output open (no load across it); change the frequency until you get 0 degrees from input to output. Then measure the voltage ratio at that frequency.
Voltage Ratio Squared = impedance ratio
25:1 for example. 25 squared is 625. 8 Ohms x 25 = 5000 Ohms.
Calculate the primary and secondary impedance, according to the voltage ratio.
Now, calculate the insertion loss of the primary, and of the secondary, that is caused by the DCRs. Use the example I gave above, just put the proper values in for the transformer you have.
Is 1.0 db insertion loss significant?
Put 12.6 Watts from the vacuum tube(s) into the primary, and all you get out is 10.0 Watts.
You decide if 1.0 dB insertion loss is significant for you.
When the old RCA, GE, Tungsol, etc. listed the power out from the tube, the specification is just that, the power from the tube.
They did not allow for the insertion loss of your transformer, you forgot to tell them that, all the way back in 1936 what your output transformer insertion loss is.
And people wonder why they get less power out than they read in the tube data sheet!
Remember, the actual primary impedance and actual secondary impedance will be higher than just according to the turns ration.
The DCRs increase the impedances of both primary and secondary.
If you forget that fact, you thought you had a 5k primary, but the two DCR losses make it a 5500 Ohm primary; do your tubes "care" about that?
If that is not important to you, then remember, the output transformer insertion loss also reduces the Open Loop Damping Factor of your amplifier.
Did the above factors ever make you wonder why you got the lower performance from your amplifier?
$0.03
DCR also contributes to distortion: https://www.mennovanderveen.nl/imag.../Partridge_1939_Distortion_in_Xfrmr_Cores.pdf
smoking-amp often reminds us that AP distortion analyzers have output transformers (for galvanic isolation) and use (and patented) current sensing positive feedback to make a negative resistance driver for the output transformers, to minimize their distortion.
Not always a whole lot can be done about DCR, it kinda is what it is, but the diyAudio field is far from exhausted about this topic. Someday...
All good fortune,
Chris
smoking-amp often reminds us that AP distortion analyzers have output transformers (for galvanic isolation) and use (and patented) current sensing positive feedback to make a negative resistance driver for the output transformers, to minimize their distortion.
Not always a whole lot can be done about DCR, it kinda is what it is, but the diyAudio field is far from exhausted about this topic. Someday...
All good fortune,
Chris
OPTs are all about making compromisses between contradicting requirements, dcr is only one of those.
In SE, fed from a tube, especially a pentode without so called "schade fb" with fb taken on the secondary, the driving impedance is so high that a high dcr is mainly a power loss, distortion wont be any smaller with lower dcr.
Things are different when driven from zero impedance, then dcr really counts because the high current distortion directly translates into voltage distortion.
A zero dcr tranformer driven from a zero impedance source has zero voltage distortion (but offcourse current distortion is highest, so the source must be able to cope with that)
Also, in SE, core permeability is a second order issue, because the prmeability of air is only 1 anyway, so the combined effective permeability of core + airgap is low in in low hundreds anyway.
As to wires, Formvar has the lowest dielectric constant (3.5 or so), solderable wire has the highest dk (up to over 6), 200°C imide (around 4.5) not as low as formvar but at least you can buy it.
And b.t.w., once we had higher resistance 4% Si iron (translates into lower eddy current) than today wich does not only impact losses, but also effective core area at increasing frequencies.
How come, you may ask, well it is cheaper to produce lams from iron with lower Si content, lower toolwear...
Just my 2 cents
In SE, fed from a tube, especially a pentode without so called "schade fb" with fb taken on the secondary, the driving impedance is so high that a high dcr is mainly a power loss, distortion wont be any smaller with lower dcr.
Things are different when driven from zero impedance, then dcr really counts because the high current distortion directly translates into voltage distortion.
A zero dcr tranformer driven from a zero impedance source has zero voltage distortion (but offcourse current distortion is highest, so the source must be able to cope with that)
Also, in SE, core permeability is a second order issue, because the prmeability of air is only 1 anyway, so the combined effective permeability of core + airgap is low in in low hundreds anyway.
As to wires, Formvar has the lowest dielectric constant (3.5 or so), solderable wire has the highest dk (up to over 6), 200°C imide (around 4.5) not as low as formvar but at least you can buy it.
And b.t.w., once we had higher resistance 4% Si iron (translates into lower eddy current) than today wich does not only impact losses, but also effective core area at increasing frequencies.
How come, you may ask, well it is cheaper to produce lams from iron with lower Si content, lower toolwear...
Just my 2 cents
I do not wind my own transformers.
But I do a number of measurements on them, so that I have an approximate idea of how I expect them to perform in any amplifier that I design and build.
Boy, it was so much easier to do all the measurements when I had the Rohde & Schwarz Vector Network Analyzer and Precision Cal Kit that I could use.
For the output transformers that I owned, I found that the insertion loss at 1kHz was almost completely caused by the total loss of the primary DCR and secondary DCR (there were no other significant insertion losses at 1kHz).
If the DCRs are low enough, then the other losses at 1kHz will show up.
But I do a number of measurements on them, so that I have an approximate idea of how I expect them to perform in any amplifier that I design and build.
Boy, it was so much easier to do all the measurements when I had the Rohde & Schwarz Vector Network Analyzer and Precision Cal Kit that I could use.
For the output transformers that I owned, I found that the insertion loss at 1kHz was almost completely caused by the total loss of the primary DCR and secondary DCR (there were no other significant insertion losses at 1kHz).
If the DCRs are low enough, then the other losses at 1kHz will show up.
Last edited:
I annoyed ChatGPT to reveal to me the most important quantifiable quality parameters of tube output transformers.
Responeses (continued):
Measurements Output Transformer Quality
To evaluate the quality of a push-pull tube Hi-Fi amplifier's output transformer, several technically measurable parameters should be considered. These parameters give insights into the performance, efficiency, and sound quality of the transformer. Key measurable parameters include:1. Frequency Response
- Range: Measures the transformer's ability to handle low and high frequencies effectively.
- Flatness: How consistent the transformer is over the audio spectrum, typically 20 Hz to 20 kHz for Hi-Fi applications.
- Phase Shift: Indicates any phase lag between input and output, which affects timing and sound coherence.
2. Total Harmonic Distortion (THD)
- Harmonic Content: Excessive distortion, especially at low frequencies, is often a result of core saturation or inadequate bandwidth.
3. Power Bandwidth
- Measures how the transformer performs at full power across the frequency range without significant distortion.
- Lower Cutoff Frequency: This is often limited by the transformer's core inductance.
- Upper Cutoff Frequency: Limited by parasitic capacitance and leakage inductance.
4. Inductance
- Primary Inductance: Determines the ability to handle low frequencies effectively without significant core saturation.
- Leakage Inductance: Influences high-frequency performance and the damping factor.
5. Core Saturation
- Magnetic Core Characteristics: When the core saturates, the transformer can't properly transfer energy, causing distortion.
- Check the maximum flux density and its saturation limit.
6. Impedance Matching
- Primary to Secondary Turns Ratio: This defines the impedance transformation and is crucial for matching the tube output impedance to the speaker load.
- Impedance mismatching leads to power loss and degraded sound quality.
7. Damping Factor
- Relates to the transformer's ability to control speaker cone movement. A low damping factor could result in muddy or uncontrolled bass.
8. Phase Distortion
- Inconsistent phase shifts at different frequencies can impact stereo imaging and soundstage.
9. Efficiency and Power Handling
- Measures how efficiently the transformer transfers power without excessive losses due to heat, leakage, or resistance.
10. Inter-Winding Capacitance
- Capacitance between windings: High inter-winding capacitance can limit the high-frequency response and cause instability in the amplifier.
11. Temperature Rise
- Excessive temperature can indicate losses or inadequate core material.
12. DC Resistance (DCR)
- Primary and Secondary Resistance: Higher resistance leads to power losses and affects damping factor.
13. Crosstalk (in stereo transformers)
- Measures how much signal from one channel leaks into the other, especially in stereo transformers.
Conclusion
A high-quality output transformer should have:- Flat frequency response,
- Low distortion,
- High power bandwidth,
- Low leakage inductance,
- High primary inductance for good bass response,
- Minimal phase shift,
- Effective impedance matching.
4o
Du:
are there more significant/relevant parameters that can be measured for quality evaluation ?ChatGPT:
ChatGPTThe parameters mentioned earlier cover most of the critical aspects when evaluating the quality of a push-pull tube Hi-Fi amplifier output transformer, but there are a few additional or refined parameters that can further deepen the analysis. Depending on the level of performance expected and the precision of the evaluation, the following parameters may also be relevant:
1. Phase Margin (Stability)
- This parameter measures the stability of the amplifier when paired with the transformer. An output transformer can introduce phase shifts at high frequencies, affecting the amplifier's feedback loop stability.
2. Magnetic Hysteresis Loss
- Core materials have inherent hysteresis, which causes energy losses when the magnetic field is cycled. Excessive hysteresis loss leads to inefficient energy transfer, especially at low frequencies.
3. Core Saturation at Various Signal Levels
- While core saturation was mentioned earlier, measuring saturation at different power levels, especially during transients, is important. Tube amplifiers have varying load demands due to music dynamics.
4. Transient Response
- The ability of the transformer to reproduce rapid changes in the signal (transients) without introducing distortion is crucial for accurate sound reproduction.
5. Distortion at High Power (Saturation Characteristics)
- Testing the transformer’s distortion at high power levels reveals how it performs when pushed to its limits.
- While Total Harmonic Distortion (THD) is important, also look at Intermodulation Distortion (IMD), which measures how well the transformer handles complex signals with multiple frequencies.
6. Leakage Inductance Under Load
- Leakage inductance becomes especially critical under load conditions. While it is usually measured without a load, measuring it under operational conditions gives more insight into real-world performance.
7. Group Delay
- Group delay refers to the delay of different frequencies passing through the transformer. Significant group delay variations at different frequencies can introduce audible phase distortion, particularly in the midrange and high-frequency spectrum.
8. Voltage Regulation
- Voltage regulation refers to the transformer’s ability to maintain consistent output voltage despite changes in load. Poor voltage regulation can lead to significant drops in output levels, particularly at higher power outputs.
9. Parasitic Capacitance Effects
- Parasitic capacitance (inter-winding and intra-winding capacitance) can limit high-frequency response, causing signal attenuation or instability.
10. Symmetry of Magnetic Field in Push-Pull
- In push-pull designs, the symmetry between the two halves of the primary winding is essential for cancellation of even-order harmonics and balanced operation.
11. Intermodulation between Channels (in Stereo Designs)
- This refers to the bleed or interference between left and right channels in stereo output transformers. It’s important to maintain channel separation to avoid sound smearing or crosstalk.
12. Transformer Core Type and Materials
- While not a "measurable" parameter in itself, understanding the core material (e.g., silicon steel, amorphous metal, ferrite) helps assess the transformer's performance. Different core materials have varying loss characteristics, saturation points, and permeability.
13. Thermal Stability
- Thermal stability indicates how the transformer performs as it heats up over time. Excessive heating can lead to degradation of sound quality, especially in long listening sessions.
14. Winding Parasitics and Resonances
- The windings themselves can introduce parasitic resonances, especially at high frequencies. These resonances can create peaks in the response or even instability in certain circuits.
Conclusion
By expanding the evaluation to include parameters like group delay, parasitic capacitance, core hysteresis loss, and thermal stability, a more complete picture of the transformer’s quality can be achieved. The goal is to ensure that the output transformer not only meets basic specifications like frequency response and impedance matching but also excels in real-world operating conditions, delivering consistent, high-quality audio performance with minimal distortion and optimal efficiency.4o
Du:
are there still more ?ChatGPT:
ChatGPTYes, there are even more nuanced and advanced parameters that could be measured to thoroughly evaluate the quality of a push-pull tube Hi-Fi amplifier output transformer. These additional metrics delve into the transformer’s behavior under specific conditions and in complex audio environments. Though some are less commonly measured due to the need for specialized equipment, they can provide deeper insights into performance.
1. Parasitic Resistance (Copper Loss)
- Definition: This refers to the resistance of the transformer windings, which results in heat dissipation (copper losses) and affects the efficiency of the transformer.
- Measurement: Use a precise ohmmeter to measure the DC resistance of the windings and calculate copper loss under load conditions. Excessive parasitic resistance can lead to inefficient energy transfer.
2. Core Losses (Eddy Current Losses)
- Definition: Eddy currents in the transformer’s core generate heat and reduce efficiency, especially at high frequencies. These losses increase with frequency due to the skin effect.
- Measurement: Measure power loss across frequencies and use a core loss tester to estimate eddy current losses. Materials like laminated steel cores or amorphous cores tend to minimize these losses.
3. Stray Magnetic Fields
- Definition: Stray magnetic fields can induce hum and noise in nearby components, which affects the amplifier’s overall noise floor.
- Measurement: Use a Gauss meter to measure stray magnetic fields surrounding the transformer, particularly under load. Shielding the transformer can mitigate stray field issues.
4. Dielectric Breakdown Voltage
- Definition: The voltage at which the insulation between windings breaks down, causing a short or arcing. It is crucial to test for insulation quality and durability, especially in high-voltage tube amplifiers.
- Measurement: Apply increasing voltage to the transformer until the insulation fails (if done destructively) or verify with a high-pot tester to confirm that it withstands voltages well above operational levels.
5. Noise and Hum Rejection
- Definition: Measures the transformer’s ability to reject or minimize external noise (e.g., hum from mains or electromagnetic interference).
- Measurement: Evaluate the noise floor of the amplifier by measuring the signal-to-noise ratio (SNR) with the transformer under various loads. Transformers with poor shielding or grounding may introduce more hum into the system.
6. Power Efficiency at Low and High Frequencies
- Definition: The ability of the transformer to maintain power transfer efficiency across the entire frequency range. Transformers often lose efficiency at frequency extremes due to core saturation or inductive/capacitive limitations.
- Measurement: Measure power output vs. input across the frequency range, particularly in the low-frequency (20 Hz – 50 Hz) and high-frequency (10 kHz – 20 kHz) ranges.
7. Temperature Coefficient of Performance
- Definition: Indicates how the transformer’s performance changes with temperature. This is important for long-term reliability and sound consistency in thermally stressful environments.
- Measurement: Measure key performance parameters (frequency response, THD, impedance) at different temperatures (e.g., room temperature vs. elevated temperature) to see if the transformer exhibits performance drift.
8. Mechanical Vibrations (Microphonics)
- Definition: Output transformers can mechanically vibrate due to the alternating magnetic field. These vibrations can couple into the amplifier chassis and become audible as hum or noise, a phenomenon known as microphonics.
- Measurement: Use a vibration sensor or accelerometer to measure the mechanical movement of the transformer under load. Ideally, the transformer should not introduce mechanical noise.
9. Interleaving and Winding Geometry Effects
- Definition: The way the primary and secondary windings are interleaved affects the high-frequency response and leakage inductance. Proper interleaving reduces leakage inductance and improves the bandwidth of the transformer.
- Measurement: Analyze the winding structure using specialized equipment (e.g., a high-frequency LCR meter) to confirm interleaving effectiveness and the quality of the winding geometry.
10. DC Bias and Its Effect on Core Saturation
- Definition: Tube amplifiers often have some DC component in their output, and this can push the core closer to saturation, especially in the absence of a perfectly balanced push-pull stage.
- Measurement: Apply varying levels of DC bias to the primary and measure the impact on core saturation and low-frequency distortion.
11. Flux Leakage and Energy Storage
- Definition: The amount of magnetic flux that leaks outside the core instead of being fully transferred between the primary and secondary windings. Higher flux leakage reduces efficiency and affects the quality of the output.
- Measurement: Use a flux meter to assess the amount of magnetic flux leaking from the transformer, which correlates with overall performance loss.
12. Load Linearity
- Definition: Describes how well the transformer maintains a linear relationship between voltage and current across a range of loads. Nonlinearities can result in frequency-dependent distortion and inefficient energy transfer.
- Measurement: Apply varying loads to the transformer and measure the linearity of voltage and current output. Nonlinear behavior could indicate core or winding saturation.
13. Capacitance Coupling Between Windings
- Definition: Electrical coupling between the primary and secondary windings via capacitance can lead to unwanted signal transfer, especially at high frequencies, degrading the transformer's performance.
- Measurement: Measure the capacitance coupling using an LCR meter, and analyze its effect on high-frequency performance with a network analyzer.
14. Voltage Spike Handling and Clipping Characteristics
- Definition: When the amplifier drives the transformer hard, voltage spikes can occur, leading to transient distortion or clipping. This is especially critical in high-power tube amplifiers where signal peaks are more common.
- Measurement: Use a fast oscilloscope to observe the transformer’s behavior under high-power transient loads, looking for voltage clipping or saturation events.
15. Saturation Recovery Time
- Definition: After the core reaches saturation, the time it takes to recover affects transient response. Faster recovery times are preferred for maintaining signal integrity during dynamic passages in music.
- Measurement: Measure the transformer’s recovery time after core saturation using a pulse generator and oscilloscope. Faster recovery times correlate with better transient performance.
16. Magnetic Field Symmetry
- Definition: Push-pull transformers should maintain symmetry in the magnetic fields generated by each side of the push-pull pair. If the symmetry is not maintained, even-order harmonics won’t cancel out effectively.
- Measurement: Measure the harmonic distortion levels of the transformer and check for any imbalance in even-order harmonics, which would indicate magnetic asymmetry.
17. Hot Spot Temperature Monitoring
- Definition: The temperature inside the transformer, especially at winding junctions or “hot spots,” is critical for long-term reliability.
- Measurement: Use thermal sensors or thermocouples placed at different points inside the transformer to monitor hot spot temperatures under load conditions.
18. Resonant Frequency and Self-Resonance
- Definition: Every transformer has a self-resonant frequency where its inductance and parasitic capacitance create resonance, which can cause unwanted peaks in the frequency response.
- Measurement: Use a network analyzer to identify the transformer’s self-resonant frequency and assess its impact on high-frequency performance.
19. Ageing and Long-Term Drift
- Definition: Over time, transformer performance can drift due to mechanical wear, thermal cycling, and material degradation.
- Measurement: Conduct long-term stress testing (thermal, electrical) and periodically measure key parameters (frequency response, THD, leakage inductance) to observe any performance drift.
Conclusion
While the previously listed parameters are fundamental, these additional measurements focus on long-term reliability, mechanical properties, and advanced electrical characteristics that further enhance the overall understanding of the output transformer’s quality. A full evaluation should consider these factors depending on the specific application, load conditions, and listening environment to ensure optimal performance of the push-pull tube Hi-Fi amplifier output transformer.4o
Du:
are there still more ?ChatGPT:
ChatGPTYou've now explored a comprehensive list of key technical parameters to evaluate the quality of a push-pull tube Hi-Fi amplifier output transformer. However, if you're seeking even more exhaustive performance factors, we can go into extreme niche parameters that may be of interest to those focused on ultimate fidelity or specialized transformer applications. While these may go beyond typical Hi-Fi transformer assessments, they can still affect sound quality or reliability in critical scenarios. Here are a few ultra-specific and esoteric measurements to consider:
1. Corona Discharge Testing (High-Voltage Insulation Breakdown)
- Definition: Corona discharge is a localized dielectric breakdown of the insulation under high-voltage stress. It can lead to long-term degradation of the transformer, especially in high-voltage tube amps.
- Measurement: Use high-voltage corona detection equipment to detect partial discharge in the insulation system of the transformer, ensuring it is properly insulated for long-term reliability.
2. Transformer Electromagnetic Compatibility (EMC)
- Definition: Transformers can radiate electromagnetic interference (EMI) which can affect nearby circuits. Evaluating the transformer’s EMC performance helps in ensuring that the amplifier as a whole doesn't cause interference with other nearby equipment.
- Measurement: Conduct electromagnetic compatibility (EMC) testing according to standards (e.g., CISPR, FCC), measuring emitted and conducted EMI levels during normal operation.
3. Core Magnetostriction Noise (Acoustic Noise from Core)
- Definition: Magnetostriction refers to the slight dimensional change of magnetic materials when subjected to a magnetic field. In transformers, this can cause mechanical vibrations and low-level noise, affecting listening environments at high volumes.
- Measurement: Use acoustic microphones near the transformer to detect any mechanical hum or noise due to magnetostriction, particularly when the transformer is under load at low frequencies.
4. Laminated Core Quality Factor
- Definition: The quality of the lamination in the transformer core affects eddy current losses and overall efficiency. Poor lamination can lead to increased losses and core heating.
- Measurement: Conduct detailed metallurgical analysis of the laminated core to ensure proper material thickness and insulation between laminations. Use a core loss analyzer to measure the quality factor of the lamination stack.
5. Core Permeability vs. Signal Amplitude
- Definition: The magnetic permeability of the core changes with the signal amplitude. For low-level signals, permeability is higher, while at high levels, the core can saturate, lowering the permeability.
- Measurement: Measure the core's permeability as a function of the signal amplitude using specialized magnetometers to assess non-linearity in the transformer’s behavior under varying signal levels.
6. DC Offset Tolerance (Core Biasing Sensitivity)
- Definition: When a DC component is present (even small), it can prematurely bias the core, pushing it closer to saturation and leading to asymmetric distortion.
- Measurement: Inject a small DC current through the primary and measure its effect on core saturation, distortion, and low-frequency response. This is important in single-ended designs but can still be relevant in imperfectly balanced push-pull stages.
7. Thermoelectric Effects (Seebeck Effect in Windings)
- Definition: Small temperature differences across the transformer windings can generate voltage due to the Seebeck effect (thermoelectricity). Though typically negligible, this can be an issue in highly sensitive, low-noise designs.
- Measurement: Measure minute voltage generation across windings caused by temperature gradients during operation, especially in amplifiers designed for ultra-low noise floors.
8. Core Domain Wall Motion Analysis
- Definition: In magnetic cores, domain wall motion can contribute to small non-linearities, creating low-level noise or distortion at certain frequencies.
- Measurement: Use advanced magneto-optical sensors to visualize and measure domain wall motion in the core under varying signal conditions, identifying potential sources of subtle distortion.
9. Core Annealing Quality
- Definition: The annealing process of the core material affects the magnetic properties, including permeability, hysteresis, and core losses. High-quality annealing results in better magnetic performance, particularly in high-fidelity audio applications.
- Measurement: Use magnetic material testers to measure the quality of the core’s annealing by comparing hysteresis curves and saturation points before and after annealing.
10. Transformer Winding Tension
- Definition: The tension of the windings during manufacture can affect the parasitic capacitances, inductance, and even mechanical resonance properties of the transformer.
- Measurement: Inspect and measure the winding tension using mechanical testing tools or during fabrication to ensure uniform tension. Non-uniform tension can lead to frequency response anomalies and long-term reliability issues.
11. Effect of Saturation on High-Frequency Dynamics
- Definition: The saturation of a transformer core tends to affect low frequencies more significantly, but subtle distortions can also occur in the high-frequency domain when the core is driven hard at low frequencies.
- Measurement: Measure high-frequency dynamic performance under strong low-frequency excitation to observe whether saturation affects high-frequency detail, leading to frequency intermodulation distortion.
12. Surface Roughness of Core Lamination
- Definition: The surface roughness of each lamination layer impacts the efficiency of flux transfer between layers. Smoother lamination surfaces reduce eddy current losses and improve high-frequency performance.
- Measurement: Use surface profilometers to measure the roughness of the laminated core surfaces and compare it against industry standards for high-performance transformers.
13. Transformer Vibration Under AC Load (Dynamic Stability)
- Definition: Transformers can vibrate dynamically under alternating load conditions, potentially generating subtle mechanical noise or affecting the internal structure over time.
- Measurement: Measure vibrational patterns and mechanical stresses inside the transformer using piezoelectric sensors under varying AC load conditions, particularly at the transformer’s resonant frequencies.
14. Electrostatic Shielding Effectiveness
- Definition: Electrostatic shielding between windings helps minimize capacitive coupling, reducing noise and improving isolation between the primary and secondary windings.
- Measurement: Perform capacitance and noise reduction tests to determine the effectiveness of electrostatic shielding, particularly in designs where ground isolation is critical.
15. Loss Tangent (Dielectric Loss of Insulation)
- Definition: The loss tangent measures the energy lost in the insulation material itself. While this is a small factor, in high-performance transformers, dielectric losses can accumulate and impact efficiency.
- Measurement: Use high-frequency dielectric loss meters to measure the loss tangent of the insulation material, especially at higher signal levels or in environments with significant temperature fluctuations.
16. High-Altitude or Pressure-Sensitive Performance
- Definition: In very specific use cases, such as military or aerospace applications, the transformer may be exposed to high altitudes or pressure changes, affecting insulation and breakdown voltage due to air pressure.
- Measurement: Test the transformer in environmental chambers simulating high-altitude conditions or rapid pressure changes to ensure that insulation and performance remain stable.
17. Transformer Barkhausen Noise
- Definition: Barkhausen noise is the irregular movement of magnetic domains within the core as they transition between magnetic states. In sensitive audio systems, this can manifest as low-level noise.
- Measurement: Use magnetic noise analyzers to detect Barkhausen noise in the core during signal transitions, particularly at low signal levels.
Conclusion
At this stage, we've reached the most esoteric, niche, and highly specialized parameters that could be measured to assess a push-pull tube Hi-Fi amplifier's output transformer. These parameters are seldom measured in typical audio setups due to their complexity and the specialized equipment required, but they can be highly relevant for research and high-end audio manufacturers who seek absolute perfection in transformer design. Many of these are over-engineering for most Hi-Fi applications but can provide insights into extreme precision and longevity in critical environments.4o
Du:
are there still more ?ChatGPT:
ChatGPTAt this point, we’ve covered virtually every measurable aspect of a push-pull tube Hi-Fi amplifier output transformer evaluation. The parameters discussed so far are so extensive that they encompass electrical, mechanical, thermal, and even esoteric magnetic properties that might only be considered in highly specialized or extreme-use cases (e.g., aerospace, scientific audio testing).
Beyond what we've discussed, most further analysis would move away from practical transformer testing into areas of fundamental physics research or material science, where measurements could only be relevant in the realm of developing new materials or extremely advanced electromagnetic studies.
Here are a few ultra-rare, cutting-edge research directions that may apply:
1. Quantum Tunneling in Insulation
- Definition: At very small scales (nano-level insulation), quantum tunneling of electrons can occur between windings or across insulating barriers, which could theoretically lead to signal degradation in some extreme cases.
- Measurement: Quantum tunneling effects are not typically measured in audio transformers but would require quantum mechanical analysis and extremely precise electron microscopy or quantum tunneling measurement devices.
2. Nanostructure Analysis of Core and Windings
- Definition: On a microscopic level, the arrangement of atoms in the core and windings may have an impact on performance, particularly in terms of magnetic behavior or current flow at extreme conditions.
- Measurement: Utilize scanning electron microscopy (SEM) or atomic force microscopy (AFM) to analyze the nanostructure of the core materials and windings. This would be more relevant in material science development rather than traditional transformer performance evaluation.
3. Casimir Effect in Transformer Gaps
- Definition: The Casimir effect describes quantum forces that arise between closely spaced conductive surfaces in a vacuum. In a transformer, tiny gaps between windings or core materials could, in principle, experience these forces, though the effect would be extraordinarily small.
- Measurement: Only detectable in extreme laboratory conditions and highly unlikely to be measurable in a standard transformer. This effect would require quantum-level instrumentation and conditions.
4. Thermo-Magnetic Coupling Effects
- Definition: In some situations, especially under extreme thermal stress, magnetic properties and thermal expansion can interact in complex ways, affecting performance in unpredictable ways.
- Measurement: Advanced thermo-magnetic coupling models would be required, including simulations using finite element analysis (FEA) and detailed thermal imaging combined with magnetic field measurements.
5. High-Frequency Plasma Effects in Transformers
- Definition: Under very high frequencies or voltage, air molecules in the vicinity of the transformer may ionize, creating plasma that can affect the performance of the transformer. This is more relevant in high-power radio-frequency transformers.
- Measurement: Special plasma detection equipment would be needed to observe any plasma discharge events. This is almost entirely irrelevant for audio transformers but could be of interest in high-voltage or high-frequency transformer applications.
6. Nonlinear Chaotic Behavior in Magnetic Domains
- Definition: At very small scales, magnetic domain movements in the core material could exhibit chaotic, nonlinear behavior under certain conditions, leading to unpredictable performance degradation.
- Measurement: Use sophisticated magnetic domain analysis with neutron diffraction or Barkhausen effect sensors under extreme load conditions to detect this behavior.
7. Anomalous Magnetic Losses at Cryogenic Temperatures
- Definition: If a transformer were cooled to cryogenic temperatures (near absolute zero), magnetic properties of the core could change drastically, and anomalous magnetic losses might appear.
- Measurement: Cryogenic testing chambers and superconducting quantum interference devices (SQUIDs) would be necessary to detect and measure these effects. Again, this is highly specialized and generally irrelevant for audio applications.
8. Anisotropic Magnetic Permeability Analysis
- Definition: In some materials, the magnetic permeability can vary depending on the direction of the magnetic field. Anisotropy in core materials could theoretically impact transformer efficiency in unexpected ways.
- Measurement: Utilize anisotropic magnetic property testers to evaluate how magnetic permeability changes with the orientation of the magnetic field within the core.
9. Thermoelectric Magnetic Field Interaction (Peltier Effects in Windings)
- Definition: Temperature differences in the windings can induce small electrical currents (Peltier effect), and these currents might interact with magnetic fields in complex ways, leading to slight performance variances.
- Measurement: Advanced Peltier effect measurement instruments would be needed, typically far beyond what would be used in audio applications.
Final Thoughts
At this point, we've ventured into fundamental physics and advanced material science, which exceeds the realm of what would ever be practically relevant to a Hi-Fi audio transformer, and moves into cutting-edge research topics. These effects are largely theoretical, or at the very least, extremely niche, and likely undetectable with traditional transformer testing equipment.In conclusion, from a practical perspective, the comprehensive list of parameters you've already seen covers everything required to rigorously evaluate a transformer’s quality in the context of a push-pull tube Hi-Fi amplifier. These additional speculative measurements would only be relevant in highly specialized contexts—usually far removed from audio applications.
4o
Du:
are there still more ?ChatGPT:
ChatGPTAt this point, we have truly delved into the most obscure and theoretical aspects of transformer evaluation and physics. Moving beyond this scope would likely take us into completely experimental or uncharted territory of transformer analysis, where no practical methods exist or where the phenomena are purely speculative. However, there are some extreme, cutting-edge fields in both theoretical physics and materials science that could potentially still be mentioned, but they're likely irrelevant or unmeasurable with current technology for an audio application.
Let's explore a few final ultra-hypothetical or fringe areas for curiosity:
1. Zero-Point Energy Effects on Magnetic Fields
- Definition: Zero-point energy refers to the lowest possible energy that a quantum mechanical physical system may have. At an atomic scale, it's theoretically possible that the interaction between quantum zero-point fields and the magnetic core could produce minuscule effects on the transformer’s magnetism.
- Measurement: Currently, there are no feasible instruments to measure the influence of zero-point energy in transformers. This remains speculative and purely in the realm of quantum physics theory.
2. Vacuum Energy Fluctuations in Transformer Gaps
- Definition: In quantum electrodynamics (QED), vacuum energy fluctuations are constantly occurring even in a vacuum. In theory, these fluctuations could affect extremely tiny gaps between windings or core materials in ways that could influence performance at ultra-small scales.
- Measurement: This would require quantum field measurement techniques or advanced particle physics equipment, which are not designed for audio transformer applications.
3. Gravitational Effects on Transformer Behavior
- Definition: In extremely high gravitational fields (e.g., near a black hole), gravitational waves and space-time curvature could, in theory, distort electromagnetic fields, potentially affecting transformer behavior.
- Measurement: This is entirely theoretical since no measurable gravitational effects would influence a transformer under normal conditions on Earth. But in extreme astrophysical environments, it could theoretically alter behavior.
4. Time Dilation and Electromagnetic Propagation
- Definition: According to relativity theory, time dilation could, under extreme conditions (near the speed of light or in intense gravitational fields), affect how electromagnetic fields propagate within a transformer.
- Measurement: This would only be relevant in the context of relativistic physics, such as space travel near the speed of light, and cannot be tested with current audio equipment. It’s more of a physics thought experiment.
5. Dark Matter Interaction with Magnetic Fields
- Definition: Hypothetically, dark matter (which constitutes most of the universe's mass but doesn't interact with electromagnetic forces directly) could, under certain conditions, interact with magnetic fields or transformers in ways we don’t yet understand.
- Measurement: Currently, there is no equipment that can detect dark matter's potential interaction with magnetic fields in any practical scenario, and it would have no measurable impact on a push-pull amplifier’s performance.
6. Cosmic Ray Interaction with Transformer Materials
- Definition: Cosmic rays, which are high-energy particles from space, constantly bombard the Earth. In extreme cases, cosmic rays could theoretically cause minor, momentary disruptions in electrical systems, including transformers.
- Measurement: Specialized particle detectors are used in high-energy physics labs to measure cosmic rays, but the impact on a transformer would be so small that it would likely be unmeasurable in an audio setting.
7. Quantum Coherence in Transformer Materials
- Definition: In certain superconducting materials, quantum coherence (the maintenance of quantum states across the system) can be a factor. While this doesn't apply to ordinary transformers, in a hypothetical future where superconducting materials are used for windings, quantum coherence could affect efficiency.
- Measurement: This would require quantum computing or superconducting testing equipment and is far outside the realm of practical transformer design today.
8. Exotic Matter and Its Effects on Electromagnetism
- Definition: Exotic matter (hypothetical materials that have unusual properties, such as negative mass) could, in theory, interact with magnetic fields and transformers in ways that aren’t understood.
- Measurement: As exotic matter remains theoretical, there’s no way to measure its interaction with transformers at this point.
9. Magnetic Monopoles and Transformer Fields
- Definition: Magnetic monopoles are theoretical particles that carry a single magnetic charge (north or south) as opposed to dipoles. If such particles existed and interacted with transformers, they could dramatically alter the way magnetic fields are generated and controlled.
- Measurement: No confirmed detection of magnetic monopoles has occurred, and current technology cannot measure them. This remains purely speculative.
10. Higher-Dimensional Electromagnetic Effects
- Definition: In theories like string theory, there are additional dimensions beyond the ones we experience. If these extra dimensions exist, they could, in principle, have some unknown effect on electromagnetic fields in ways we cannot yet measure.
- Measurement: This is entirely speculative and remains within the domain of theoretical physics. No instruments currently exist to test for such effects in transformers.
Conclusion
The list we've now explored represents the absolute outer limits of what could be considered in evaluating a transformer’s performance. Most of these ideas are purely theoretical or speculative, often grounded in advanced quantum mechanics, relativity, or cosmology rather than audio or electrical engineering. These concepts are so far removed from practical Hi-Fi transformer testing that they may never be relevant, but they are interesting thought experiments about how physics at the most fundamental levels could potentially influence electromagnetic devices.In terms of real-world applications, all measurable aspects of a Hi-Fi transformer have been covered in detail already. The rest belongs to the realms of scientific curiosity or hypothetical future technology.
4o
Du:
are there still more ?ChatGPT:
ChatGPTWe've now reached the absolute boundaries of both practical and speculative transformer evaluation. Beyond this point, we’re entering territory where the exploration would be purely theoretical, philosophical, or even science fiction. But to entertain your question fully, let's dive into the most outlandish, unproven, or purely hypothetical ideas that, while intriguing, have no measurable foundation in current science or transformer engineering.
1. Interdimensional Magnetic Field Leaking
- Definition: In certain science fiction scenarios or speculative physics theories (like the multiverse or parallel dimensions), it's hypothesized that electromagnetic fields might "leak" between dimensions. This could theoretically affect a transformer's magnetic field if it was positioned at a point where these dimensions interact.
- Measurement: There are no known ways to measure or even confirm the existence of interdimensional field interactions. This concept resides in science fiction rather than science.
2. Time Travel-Induced Field Anomalies
- Definition: If time travel were possible, changes to the magnetic field in the past could theoretically ripple through to affect a transformer's performance in the present. This might manifest as subtle electromagnetic fluctuations that defy normal causality.
- Measurement: Measuring this would involve understanding and proving the mechanics of time travel, which remains far beyond the scope of current science.
3. Black Hole Effects on Electromagnetism
- Definition: If a transformer were placed near the event horizon of a black hole, space-time curvature and gravitational time dilation could dramatically alter its magnetic field and electrical performance. This is an extreme thought experiment in theoretical physics.
- Measurement: Currently, we have no practical methods for measuring electromagnetic performance near a black hole. This would require an entirely new branch of physics and space travel technology.
4. Warp Drive or Faster-Than-Light Field Effects
- Definition: In theories involving faster-than-light travel or warp drives, it’s possible that the magnetic and electric fields of a transformer would behave in unknown ways due to the distortions in space-time.
- Measurement: There is no experimental setup to simulate or measure the effects of faster-than-light travel on a transformer. It’s a speculative idea at best.
5. Simulated Reality Testing
- Definition: If we are living in a simulated reality (as suggested in certain philosophical or scientific hypotheses), the behavior of electromagnetic fields, including those in transformers, could be influenced by the underlying code or framework of the simulation.
- Measurement: Measuring this would involve proving the existence of the simulation itself and understanding how its programming works, which remains purely hypothetical.
6. Alien Technology Integration
- Definition: If we were to acquire and integrate advanced alien technology into transformers, it's possible that unknown forms of energy, fields, or materials could influence transformer behavior in ways we can't predict or measure with current human technology.
- Measurement: Theoretical at best, since we have no access to alien technology or the means to integrate and measure it.
7. Quantum Superposition of Transformer States
- Definition: If a transformer could exist in a state of quantum superposition (existing in multiple states at once, as theorized for particles in quantum mechanics), it might theoretically have multiple output states simultaneously.
- Measurement: This would require advanced quantum computers and physics to measure, and it's currently unobservable in the macroscopic world. Transformers as classical devices don’t exhibit quantum superposition.
8. Antigravity Transformer Testing
- Definition: In a hypothetical future where antigravity technology is developed, transformers might function differently due to the absence of gravitational forces, potentially altering the distribution of magnetic fields or the movement of electrons.
- Measurement: Antigravity technology doesn’t exist, so there’s no way to measure or simulate the effects of gravity removal on transformer performance.
9. Dark Energy Field Interaction
- Definition: Dark energy is a mysterious force theorized to drive the expansion of the universe. If dark energy could interact with electromagnetic fields in some unknown way, it might have subtle effects on transformers in specific cosmic conditions.
- Measurement: No known instruments or experiments have been able to detect or measure interactions between dark energy and electromagnetic fields. This remains a cosmological puzzle.
10. Magnetism in Alternate Universes
- Definition: If alternate universes exist with different physical laws, the nature of electromagnetism in these universes might differ significantly from our own. A transformer from another universe could behave unpredictably in ours, or vice versa.
- Measurement: This is entirely speculative and unmeasurable, as we have no access to alternate universes or knowledge of their physical laws.
Final, Ultimate Conclusion
We are now far beyond any scientific or engineering discipline that applies to transformer design, evaluation, or operation. The ideas listed above stretch into the realms of imaginative speculation, philosophical inquiry, or even science fiction, where there's no empirical foundation for measurement or testing. These concepts, while entertaining and thought-provoking, have no practical relevance to Hi-Fi amplifier transformers or any technology available today.In terms of real-world transformer evaluation, we have already covered all practical, scientific, and theoretical aspects. Any further inquiries would require breakthroughs in quantum mechanics, cosmology, or even alien technology, which are far removed from the scope of electrical engineering.
So, to definitively answer your question: There are no more practically measurable parameters left. We've reached the absolute limits of what's conceivable.
4o
Du:
are there still more ?ChatGPT:
ChatGPTAt this point, we've completely exhausted the measurable and speculative domains that could be applied to transformer evaluation. We've traversed from real-world engineering to quantum mechanics, cosmology, theoretical physics, and even science fiction scenarios. You’ve essentially asked, "What else exists in all of known and unknown physics that could relate to transformers?"—and we’ve gone as far as possible, venturing into thought experiments that defy our current understanding of reality.
To emphasize, there truly are no more measurable or relevant parameters left for the evaluation of a push-pull tube Hi-Fi amplifier output transformer. All practical, hypothetical, and speculative ideas have been covered.
If you still wonder about more, we would be stepping into the completely unknown, meaning we'd be inventing new physics entirely—untested, unexplored, and with no known basis for measurement or experimentation.
In conclusion: We are at the absolute limits of transformer analysis and science itself. There really are no more measurable factors.
long post of mine will be an exception. yes, really long, partially redundant list, esoteric in the end, interesting.
FrankieS,
Thanks for the definitive writeup on Valves / Tubes on that subject!
For the rest of us . . .
If we have all the necessary equipment to do all those tests, and the time to do all those tests . . .
Then we can afford to purchase:
10 sets of 2 SE output transformers from the manufacturer of your choice (10 Stereo amplifiers)
And
10 sets of 2 push pull output transformers of the manufacturer of your choice (10 Stereo amplifiers)
Or, just purchase:
2 SE output transformers from the manufacturer of your choice,
And
2 push pull output transformers from the manufacturer of your choice.
Let your favorite output transformer companies do all the work!
You will $$$ Save Money!
Thanks for the definitive writeup on Valves / Tubes on that subject!
For the rest of us . . .
If we have all the necessary equipment to do all those tests, and the time to do all those tests . . .
Then we can afford to purchase:
10 sets of 2 SE output transformers from the manufacturer of your choice (10 Stereo amplifiers)
And
10 sets of 2 push pull output transformers of the manufacturer of your choice (10 Stereo amplifiers)
Or, just purchase:
2 SE output transformers from the manufacturer of your choice,
And
2 push pull output transformers from the manufacturer of your choice.
Let your favorite output transformer companies do all the work!
You will $$$ Save Money!
thanks 6A3sUMMER.
I 'researched' this topic as a layman and beginner, via ChatGPT, to find out, which are the more important parameters, but what could also be 'relatively' important. yes, the instruments could be rather expensive and the results less important.
I think of starting (when?) a push pull stereo power amplifier project in the future, preferentially with EL84s. (or EL34s).
The Morgan Jones books are an excellent start.
For myself the question was/is: what are the single best factory-new output transformers for PP EL84 ? The price, really would be less important,
I must confess, but any very expensive OT would only be a choice, if the audio quality is RELEVANTLY better than the less expensive ones.
(then: what is 'relevantly' better? difficult.)
I feel Lundahl could be a very good choice, but also ask myself, whether the cheaper Toroidys would be even better technically and audio-wise ?
any input much appreciated.
cheers
I 'researched' this topic as a layman and beginner, via ChatGPT, to find out, which are the more important parameters, but what could also be 'relatively' important. yes, the instruments could be rather expensive and the results less important.
I think of starting (when?) a push pull stereo power amplifier project in the future, preferentially with EL84s. (or EL34s).
The Morgan Jones books are an excellent start.
For myself the question was/is: what are the single best factory-new output transformers for PP EL84 ? The price, really would be less important,
I must confess, but any very expensive OT would only be a choice, if the audio quality is RELEVANTLY better than the less expensive ones.
(then: what is 'relevantly' better? difficult.)
I feel Lundahl could be a very good choice, but also ask myself, whether the cheaper Toroidys would be even better technically and audio-wise ?
any input much appreciated.
cheers
- Home
- Amplifiers
- Tubes / Valves
- What Qualities Makes A Good Output Transformer