its driving a ADC input so probably more than 10k no buffer needed. I suggested a transformer because the op dosnt build circuits and thats what every one else is suggesting. And theres other transformers that might be better suited.
By the way, when you put 600 ohm between the amplifier(s) and the transformer to get more distortion, and load the output with an attenuator with 600 ohm input impedance, the amplifier has to produce twice the voltage.
With 600 ohm between amplifier and transformer and a high-ohmic load, you get more transformer distortion at a given voltage level and have the amplifier and transformer work at essentially the same voltage level. I don't know what happens to the frequency response, does the datasheet say anything about 600 ohm source and high-impedance load?
With 600 ohm between amplifier and transformer and a high-ohmic load, you get more transformer distortion at a given voltage level and have the amplifier and transformer work at essentially the same voltage level. I don't know what happens to the frequency response, does the datasheet say anything about 600 ohm source and high-impedance load?
Not sure but here's a link to what I could find in the webs
https://www.google.com/url?sa=t&sou...kQFnoECAQQAQ&usg=AOvVaw0zZsPa_sZ9eHGISwZ4-M6a
So it's designed to be used with 150 ohm (windings in parallel) or 600 ohm (series) on both sides, no information on what happens with a high-ohmic load.
Most transformers need an RC series network termination when used with a high-ohmic load, to prevent peaking due to spreading inductance. The values depend on the transformer.
Most transformers need an RC series network termination when used with a high-ohmic load, to prevent peaking due to spreading inductance. The values depend on the transformer.
Actually 600 ohms with .018 would be a design target for this transformer. I would use 620 ohms and .022 uF. But I would use a 200 volt DC rating on the capacitor. For the resistor assuming an audio and not a test signal I would use 2 watts carbon composition. Or four 150 ohm 1/2 watt units as a series string.
However as the intent is to maximize the distortion you might want to drop the capacitor to a more standard .01 uF. Of course the same voltage rating.
The other technique to increase the distortion would be just to use one primary winding.
When the transformer was designed for telephone use, it was intended as a “Repeat” coil, no high impedances involved. Intended to keep line impedance constant and low to preserve frequency response but lose level. The source impedance came from the resistance of the long telephone cables in actual use or a vacuum tube amplifier.
When driving the coils from an audio power amplifier build out resistors are a must. Blowing open the windings most likely would be a bad thing.
Last edited:
600 ohms reactance at the upper frequency limit of 15,000 Hz.
Carbon composition not just for increased distortion but also in case of a power surge from doing something stupid like power amplifier starting or stopping surge, maybe even a live input disconnect. Carbon composition resistors have much higher surge capacity than film types. If not carbon composition then a six to ten watt resistor.
Carbon composition not just for increased distortion but also in case of a power surge from doing something stupid like power amplifier starting or stopping surge, maybe even a live input disconnect. Carbon composition resistors have much higher surge capacity than film types. If not carbon composition then a six to ten watt resistor.
Not important. The discussion has moved to the minutia that people like to talk about to...I don't know...make themselves look smarter?
If I thought any of this mattered I'd be testing a 111C right now. I have several. And I'm not.
Just put your rig together and give it a try.
Your Samson Servo 120a Power Amplifier can deliver 120 W into 8 ohm in bridged mode, which corresponds to 30.98386677 V RMS. If you connect it to the transformer via 600 ohm (or two times 300 ohm) and load the transformer with 600 ohm, the voltage at the transformer input will be only half of it, so 15.49193338 V RMS - theoretically anyway, chances are that the amplifier will deliver a bit more at light load and the transformer losses also do something, so let's guess it will be about 17 V RMS. That's about 26.8 dBu, while the target was 35 dBu. I don't know why jaddie considers missing the spec by 8.2 dB unimportant, it doesn't seem logical to me.
Two things you can do to get closer to the target level, or even above it:
1. Don't load the secondary, that will result in 6 dB extra. The disadvantage is that this may result in a peak somewhere in the treble range, or in the ultrasonic range if you are lucky. An RC network on the secondary side could fix this. simon7000 has a way to calculate what RC network you need, I think he overestimates the required capacitance. jaddie could measure it, but refuses to do so.
2. Put the primary windings in parallel rather than in series and use a 150 ohm resistor (or two 75 ohm resistors for symmetry) between amplifier and transformer. That will also result in 6 dB extra.
If you want the resistors to always survive, even with shorted transformer output and the amplifier driven into hard clipping, you need about 4 W power handling for the 600 ohm resistor or 2 W each for the two 300 ohm resistors between the amplifier and the transformer. With change 2, that becomes 16 W for the 150 ohm resistor or 8 W each for the two 75 ohm resistors.
It's a bit overdone when you use music, don't clip the amplifier and don't short the transformer, though. One eighth of those power values will then be more than enough. simon7000 prefers carbon composition resistors because they are good at handling short overloads.
Man, this is still French to me. You said if I put the amp in bridged mode, which I assume means only getting one channel out of the amp (it's a two channel)... I need both channels, one for left, one for right, so I won't be bridging it. That means it will be doing 60 per channel at 4 ohms. Does that change what you wrote above?
I don't know what loading the secondary means. I don't know what an RC network is. I don't know what primary windings are or if I have them wired in series or parallel. I don't understand how I would short the transformer, or anything else you said after that.
I know next to nothing about electrical engineering. I can do simple things like make cables and solder resistors in a path if I'm told where, but that is it. I do appreciate your input, I just don't understand it.
No I only wanted to use 1 amplifier that would power both channels.
I have been googling and am starting to understand some of things mentioned above. I know know RC is a resistor into a capacitor, I understand how the volts were calculated, and I understand that there are two sides to a transformer.
When I recalculated the volts for 60w at 4 ohms I still ended up with the same voltage per channel as the bridged amount. I'm guessing that's wrong.
My transformer is currently configured for 1:1 at 600ohms so I'm assuming that means the secondary isn't loaded?
That's about where I'm at. Can I still at least 30dbu per channel now that you know I'm only using 1 amplifier?
Also, I'm pretty sure the 60w per channel rating is peak, not rms.. but that might be ok because the +30dbm rating of the transformer is also a peak rating (I think?).
Seems to be what I find the Samson Servo 120A and Servo 200 are the same price! The Servo 200 would have more voltage so would be a better choice.
Mouser has 1/2 watt carbon composition resistors. So you might want to buy ten pieces of 33 ohm 1/2 watt units to drive your transformer pair at 150 ohms primary connection. As that is almost no money you should also get four .01 uF 200 volt DC film resistors and two 600 ohm 1/2 watt resistors. To make the order worth a bit more add a spool of solder of your choice. Maybe even the connectors you will need.
Mouser has 1/2 watt carbon composition resistors. So you might want to buy ten pieces of 33 ohm 1/2 watt units to drive your transformer pair at 150 ohms primary connection. As that is almost no money you should also get four .01 uF 200 volt DC film resistors and two 600 ohm 1/2 watt resistors. To make the order worth a bit more add a spool of solder of your choice. Maybe even the connectors you will need.
Last edited:
Yeah I got the Samson used already so no turning back there.
What are the 200 volt DC resistors for? The output of the transformer?
If I wanted to send the transformer 600ohms and keep it wired at 1:1 the way they are now, what watt per channel would I actually need? 120 per channel?
200 volt DC rating on the capacitor used to load the transformer output.
What you need is enough voltage to get the transformer into saturation distortion.
The voltage available from an amplifier will be the square root of eight times the power rating into 8 ohms.
So a cheap 120 watt amplifier is probably 60 watts per channel into 4 ohms or say 35 watts into 8 ohms. Thus the square root of 8 x 35 would be 16.7 volts. If it were rated at 60 watts at 8 ohms it would provide just under 22 volts. However as you might want to put a resistor in series with the transformer primary that will reduce the voltage, the exact amount will depend on the secondary load.
Wiring the transformer primary as 150 ohms instead of 600 ohms reduces the voltage to get into distortion to half the voltage.
You want to see at least 26 volts for wiring the input as 600 ohms. Much easier to get 13 volts wiring it as 150 ohms.
What you need is enough voltage to get the transformer into saturation distortion.
The voltage available from an amplifier will be the square root of eight times the power rating into 8 ohms.
So a cheap 120 watt amplifier is probably 60 watts per channel into 4 ohms or say 35 watts into 8 ohms. Thus the square root of 8 x 35 would be 16.7 volts. If it were rated at 60 watts at 8 ohms it would provide just under 22 volts. However as you might want to put a resistor in series with the transformer primary that will reduce the voltage, the exact amount will depend on the secondary load.
Wiring the transformer primary as 150 ohms instead of 600 ohms reduces the voltage to get into distortion to half the voltage.
You want to see at least 26 volts for wiring the input as 600 ohms. Much easier to get 13 volts wiring it as 150 ohms.
Last edited:
He doesn't need to load the transformer output with a capacitor. It's unnecessary in this application. Most 111C transformers in the world were never loaded with a cap.
You do realize that saturation is a curve, right? Level vs Frequency. And a rather steep one. I posted that curve for a Jensen 123, and the 111C is similar in shape. The target was +30dBu, but frankly, that's way too low if you want to saturate mid-band. Going back to Post #45, if you really want transformer "color", use a different transformer that will saturate at a sane level.
The Samson in question is 60/side into 4, 55/side into 8.
Everybody forgets that saturation, and the audible distortion that results, is a PEAK function, specifically Peak-to-peak. Everybody's calculating RMS levels. Yes, the distortion graphs reference RMS, but the actual distortion mechanism, saturation, happens on peaks, we only measure it at RMS values. It's audible at peak values, and audibility is dependent on the complex combination of frequency of saturation, level, and what else is in the spectrum that might either mask harmonics or be intermodulated. Unless the OP want's horrific amounts of distortion, 120w should be enough.
I don't disagree that the 111C could be wired for 150/600, and the termination resistor could be left off. And that's why you can also forget about the RC compensation net. This is now an "effects" box. All options are open. I'd even suggest a pair of back-to-back diodes and a couple of resistors, if you want so extra crunch.
Might as well say this now: when it comes to tranformer distortion boxes generating "glue", or whatever, I still don't get it. In the last 50+ years of audio technology development one of the primary goals has been to reduce or eliminate distortion of all kinds. We have some excellent examples of the resulting recordings. They didn't need any glue, gunk, or glop to "improve" anything. And with the dizzying array of device-simulating plug-ins that provude the user to a full palette of effects colors that are predictable, repeatable, and even removable, I fail to understand why anyeone would want to introduce deliberate, wild, nonlinearity into a perfectly clean system. I will eventually have my how high-level 111C driver to play with, so perhaps my days of not understand this are coming to a middle.
But hey, it's an effect. A rather permanent one, but never the less. So are back-to-back diodes, but nobody talks about that really cheap distortion generator.
