Hi,
I was not sure where to post this thread, as it mainly pertains to a signal transformer question, but will have a tube amp application and I figure there should be many knowledgeable people in here for my question.
With that said, I have a paire of Cinemag CMMI-5C step up transformers. They are rated to operate with a secondary load of 39kohm, which requires a driving impedance of 150ohm. I know that using lower impedances will quickly degrade the performances of the transformer, but what about higher impedances?
My application is to boost signal prior to my tube amp's input. It's an RCA/Victor console amp I resotred, but the thing has only 6dB of gain and therefore needs strong input signals if I hope to achieve it's full ~10-12W into 8ohm. Since I plan to use it with a CDP and passive pre, 6dB gain is not enough.
The input impedance of my tube amp is approximately 367kohm. The CMMI-5C being 1:5 step-up, reflects that as a ~14.7kohm load on the primary, wich is easily driven by a CDP. I did subjective listening tests and it generally seems fine, no distortion, and seemingly intact frequency response.
Using such a transformer with lower impedances would kill the HF response due to inductance, but higher impedances, as is my case should not cause a decrease in performance the way I see it. Does that sound correct? Are there issues you can see with what I'm doing?
Thanks!
IG
I was not sure where to post this thread, as it mainly pertains to a signal transformer question, but will have a tube amp application and I figure there should be many knowledgeable people in here for my question.
With that said, I have a paire of Cinemag CMMI-5C step up transformers. They are rated to operate with a secondary load of 39kohm, which requires a driving impedance of 150ohm. I know that using lower impedances will quickly degrade the performances of the transformer, but what about higher impedances?
My application is to boost signal prior to my tube amp's input. It's an RCA/Victor console amp I resotred, but the thing has only 6dB of gain and therefore needs strong input signals if I hope to achieve it's full ~10-12W into 8ohm. Since I plan to use it with a CDP and passive pre, 6dB gain is not enough.
The input impedance of my tube amp is approximately 367kohm. The CMMI-5C being 1:5 step-up, reflects that as a ~14.7kohm load on the primary, wich is easily driven by a CDP. I did subjective listening tests and it generally seems fine, no distortion, and seemingly intact frequency response.
Using such a transformer with lower impedances would kill the HF response due to inductance, but higher impedances, as is my case should not cause a decrease in performance the way I see it. Does that sound correct? Are there issues you can see with what I'm doing?
Thanks!
IG
Hi,
It will be interesting to see an experts reply .....
If you are using it at higher voltage levels than its designed for then I
understand things balance out. e.g. if its a 1:5 microphone transformer
then this would well with higher voltages and impedances, the core (sic)
point being the designed normal operating flux or current in each winding.
/Sreten.
It will be interesting to see an experts reply .....
If you are using it at higher voltage levels than its designed for then I
understand things balance out. e.g. if its a 1:5 microphone transformer
then this would well with higher voltages and impedances, the core (sic)
point being the designed normal operating flux or current in each winding.
/Sreten.
Last edited:
Actually, Steve Eddy more than me.
In any case, having the load higher than nominal will often cause ultrasonic ringing. If it sounds OK to you, then proceed no further, If you want to tweak and have some fun, you can optimize things by driving the input with a square wave generator and (if needed) a series resistor to pad the source impedance to match the source impedance of your CDP, then attach a series RC network (Zobel) to the transformer secondary. While looking at the secondary voltage with a scope, adjust the R and C values to optimize the square wave response. As starting values, I'd try 47k in series with 100pF.
In any case, having the load higher than nominal will often cause ultrasonic ringing. If it sounds OK to you, then proceed no further, If you want to tweak and have some fun, you can optimize things by driving the input with a square wave generator and (if needed) a series resistor to pad the source impedance to match the source impedance of your CDP, then attach a series RC network (Zobel) to the transformer secondary. While looking at the secondary voltage with a scope, adjust the R and C values to optimize the square wave response. As starting values, I'd try 47k in series with 100pF.
If there was ringing, I could not hear it, but I'll definitely test for it, because it cannot be a good thing to have around, wether you hear it or not.
The Cinemag datasheet already shows a Zobel in their test network, optimized for their rated impedances though. I'll look into this...when I get my O'scope working...nearly every one of my projects is getting stalled by this lately.
Thanks for the contribution!
IG
I have no experience with this particular transformer, but microphone input transformers when made optimal (and I believe Cinemag transformers are) are not good for higher voltages and resistances. 150 Ohm primary would have too low inductance causing roll-off and saturation on low end of frequencies.
I would suggest to go on Edcor website and buy a couple of transformers under $10 each for levels of signals you need.
I would suggest to go on Edcor website and buy a couple of transformers under $10 each for levels of signals you need.
As for signal level, with my CDP's 2V output, I'm slightly under the maximum input signal level of +10dBu for my particular model, so that aspect should be OK. As for higher impedances on their own, I would not know right now.
Thanks for your input. I'll have to do some actual testing when I repair my scope, those tips from all of you guys will give me stuff to look for.
Edit - I might just check out some of these Edcor you suggested, they are pretty cheap! Thanks for the suggestion
Cheers,
IG
Last edited:
Hi,
Higher voltages I agree would cause saturation, but I cannot see the case
for when this is combine with higher impedances. If the currents are the same
the voltages do not matter, a transformer is a current operated device IMO.
Capacitance of some sort would be the major issue into higher impedance.
rgds, /Sreten.
FWIW I think you have to chuck nominal loading out of the window and
re-zobel with a far higher resistance + capacitor to keep currents down.
Last edited:
Sure. And current depends on frequency.
What inductance can you expect from 150 Ohm primary of a microphone transformer?
Can you calculate it's impedance on low end of an audio band?
Can you imagine what happens to signal if it's source is loaded on such an impedance?
Hi,
The primary impedance is reflected from the secondary, its not fixed.
In this case its around 1.5kohms not 150ohms, bass will be extended.
The problem area AIUI will be the treble end.
rgds, /Sreten.
Last edited:
--What job does your husband have?
--Transformer.
--??
--He gets 240, brings home 110, and buzzes on the rest.
It's similar explanation to what you give. In reality, transformers are 2 coils on the single core, that's why load of one coil is reflected on the 2'nd one, but you should remember that they are coils, and impedances of coils may be calculated using this nice online thingy:
Pronine Electronics Design - Inductor Impedance Calculator
--Transformer.
--??
--He gets 240, brings home 110, and buzzes on the rest.
It's similar explanation to what you give. In reality, transformers are 2 coils on the single core, that's why load of one coil is reflected on the 2'nd one, but you should remember that they are coils, and impedances of coils may be calculated using this nice online thingy:
Pronine Electronics Design - Inductor Impedance Calculator
Hi,
Sreten is right, transformers do no more than transform voltages and currents.
You can speak of impedances when the source or load impedance is fixed.
For instance, there is "no" 3k2 SE output transformer unless the load impedance is an 8 ohm loudspeaker and the winding ratio 20:1 (the same transformer is 6k4 with a 16 ohm loudspeaker).
When a primary impedance like 150 ohm is specified it points in the direction of the source, a microphone with a 150 ohm output impedance in this case.
It also indicates that the primary induction of the transformer should be at least 1 Henry for sufficient bass response.
High frequency peaks are a function of load impedance and winding technique.
With lower source impedance and simple winding technique the HF peak(s) will be more pronounced.
With higher source impedance and/or multi-layer winding techniques there is more damping of peaks.
With too high source impedance high frequency audio bandwidth will suffer because the winding capacities are too high for the source impedance.
Pieter
Tribute Audio Transformers
Sreten is right, transformers do no more than transform voltages and currents.
You can speak of impedances when the source or load impedance is fixed.
For instance, there is "no" 3k2 SE output transformer unless the load impedance is an 8 ohm loudspeaker and the winding ratio 20:1 (the same transformer is 6k4 with a 16 ohm loudspeaker).
When a primary impedance like 150 ohm is specified it points in the direction of the source, a microphone with a 150 ohm output impedance in this case.
It also indicates that the primary induction of the transformer should be at least 1 Henry for sufficient bass response.
High frequency peaks are a function of load impedance and winding technique.
With lower source impedance and simple winding technique the HF peak(s) will be more pronounced.
With higher source impedance and/or multi-layer winding techniques there is more damping of peaks.
With too high source impedance high frequency audio bandwidth will suffer because the winding capacities are too high for the source impedance.
Pieter
Tribute Audio Transformers
Hi Pieter;
at 20 Hz 1 Henry means about 135 Ohm of inductive impedance. It is in parallel with reflected load of the primary.
Now, please calculate frequency response for 10K source loaded on 135 Ohm of inductive impedance and you may understand why I order my transformers from Altran and Edcor, and never from Tribute Audio Transformers.
Peace Profound!
Anatoliy Lisovskiy
Hi,
I'm sorry, I cannot see your point, that may be my fault. All I know is
the maximum throughput power loading of a transformer is related to
the input / output voltage and current and it usually needs derating
for the situations that are different, and 10:1 is very different.
e.g. a 110v isolating transformer can be used for 220v derated,
and the opposite situation, 220V for 110V again derated (in VA).
I'm sure I'm missing something the basic winding ratios do not indicate,
but I still think you can use transformers at higher voltages with higher
impedance loading to keep power throughput the same, and the main
issue with this case is going to be at the high frequency end.
As I understand it saturation is related to currents only in the transformer,
and its the secondary loading that determines the input/output voltage levels,
related to the energy stored in the core flux caused by the coil currents.
rgds, /sreten.
Did not see PT's post and WB's reply when posting this.
Last edited:
You can't do that. If you wanted to use a 110V isolation xfmr at 220V, the only way you could do that is to double the frequency.
Vn= 4fSB (B in Webers/M^2; S in M^2; f in Hz)
If you double Vn, you must double f otherwise the core will saturate badly, and the thing will probably burn up even with no load.
As for going lower in voltage, that's do-able with a decrease in efficiency since core losses will be bigger than necessary. Unless, of course, you halve the operating frequency.
As for impedance matching, what's important is the turns ratio for impedance transformation, and the inductive reactance of the primary and secondary. Trying to go higher than the design nominal impedances may cost you on the low end of the range since you'll run out of Xl at higher frequencies. Going higher could impair high frequency performance due to leakage inductance and stray capacitance.
I think that Waveborn's point is that the primary inductance that would be sufficient to support full bass response for a source with say a 1K ohm source impedance would necessarily be higher than that required for a 150 ohm source. Thus unless the input transformer is designed with a ten times margin of inductance over what is needed for the original application you are likely to lose some bass response.
Why not see if Cinemag can tell you the primary inductance of this transformer. Just off the cuff I am thinking that you are going to want at least eight or ten Henries.
Why not see if Cinemag can tell you the primary inductance of this transformer. Just off the cuff I am thinking that you are going to want at least eight or ten Henries.
Hi,
Note that I said for both cases derated, I did not state the amount of deration.
rgds, /Sreten.
Cinemag guaranteed no more than 0.8 dB drop on 20 Hz, if the transformer is used according to specs. You may estimate minimal inductance they guarantee.
Speaking of a ten times margin, that would require higher number of turns both in primary and secondary, that means more losses and worse response on the high end of the frequency band. Audio frequency band is so wide so it is always a compromise between multiple criteria of optimization when designing and building real things.
Speaking of a ten times margin, that would require higher number of turns both in primary and secondary, that means more losses and worse response on the high end of the frequency band. Audio frequency band is so wide so it is always a compromise between multiple criteria of optimization when designing and building real things.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.