• The move to the new server is done. There are some software and database maintenance updates in process. This has us passing the hat around to help out. We appreciate any donations. Seriously, even a dollar helps. The payment page may be found here - https://www.audiokarma.org/support.html

Amp square wave testing question, or Ignorance is Bliss

Schmidlapper

Schmidlapper

Super Member
In testing my tube amp's output I have read much on square wave testing and in my own amp I have a good 1k square wave at ~ 1 watt. Now understand my testing is limited since my Eico 377 died, and I now only have a mac mini sound card along with an online signal generator to use, With that though I have a good rise time, no corner roll over, no tilt and the output mirrors the input, also true at 400 hz. So Ive been happy with that until I read about Kward getting his great square waves at 10Khz, and so I start reading and another guy says for audio amps fundamental frequencies of 50hz and 1khz are sufficient to test, with 10k included in lab equipment category. So I go and check my amp and for 50hz I get a good bit of tilt [determined the Mac mini signal gen was outputing them this way, not the amp], and for 10khz I get a sine wave [likely sig gen too]. Now I'm thinking at 50hz fundamental the odd order harmonics necessary to make a 50hz square wave are going to be well out of most amps functioning range, and should I really expect a 10khz square wave? Is my mac mini even capable of making a 10khz square wave? So now I'm left wondering. The amp is a well modded 8802 with donor output transformers from a Rowe AMI jukebox amp. I get great sine waves from 30hz to 20khz, with the 1khz settling in at 5.2 watts max, and the 30 and 20k less.
 
Last edited:
Register to hide this ad
You are going to get tilt at 50Hz squarewave. The lowest corner frequency is about 10Hz or so, you are not going to get a perfect square wave at 50Hz, you are not going to get a perfect square at 100Hz. I know because I was testing that on my SS amp that has low corner frequency less than 10Hz.

You are going to have to increase the values of all the coupling caps to get better shape at 50Hz.......for nothing!!!

There is not hard rule that you have to use square, in your particular case, if you want to check the low frequency roll off, use sine wave and start at 1KHz, note the amplitude on the scope. Then decrease the frequency to the point where the amplitude reduces to 0.7 of the original amplitude, note that frequency, that's your low rolloff frequency. Usually it's about 7 to 20Hz and you are good.
 
Yes, using sine wave. The exact number is 0.707 which is 1/(sqrt 2)

This is also call -3dB point. Where dB= 20 log (Vout/Vin)=20log(0.707)=-3dB.

This is how amps are spec. Say they say the amp's frequency response is from 20Hz to 20KHz, what they say is the low -3dB point is 20Hz and the high -3dB point is 20KHz.
 
Last edited:
LF rolls off somewhere below 30Hz, as that was the lowest I had measured sine wave begins to distort a lot any lower, and HF is rolling off at ~19Khz.
 
For tube amp, that sounds about right. But I don't know why the waveform distort if you go lower than 30Hz. It might be the low frequency limit of the output transformer. Transformer becomes very lossy at low frequency and strange things might happen. I am not an expert on tube amps, that's my guess.
 
It can go lower without distorting, it just can't go lower at or near full volume. I actually have older o scope screen shots I will upload.
 
You have some oscillation in the 30Hz signal, that you might want to take a closer look.

Oscillation.JPG

I want other people that has more experience in tube amp to comment on the distortion, I think it's the output transformer. You are reaching the lower limit of the transformer. But I'll wait for someone else to comment.

Can you do a square wave of about 500Hz, crank the frequency up and down and take pictures of the ugliest waveform. I want to take a closer look at the ringing. The oscillation worries me more than any of the things you asked.
 
This is the 1khz square wave, scale set to 2v the Mac mini's output has these symmetrical overshoots also, so I ignore them for diagnosis until I can get a proper signal generator.
9A5364F5-7608-472B-868C-E266FDA66DA3.jpeg
 
You have some stability issue, from the picture, looks like you are on 200uS/div, you have about 8 ringing cycle in 3 division, so the ringing has period of 600uS/8=75uS. That translate to about 13.3KHz. You definitely have some stability issue.

You have your ground lead of the scope probe connect to the ground? Make sure you do that.


This is what I use square wave for testing, looking for ringing and stability.

Also crank the square frequency up and down, do you see the amplitude increase around 13KHz?
 
That oscillation bothered me too, but it only starts to appear at 30Hz and then if I go above 3/4 volume. 20Hz is worse, but it too is clean up to about 1/4 volume on log taper pot. I’m hesitant to look at ringing without a real signal gen. I think it is the transformers too, they are from a 7868 tube amp and I had to choose about a 7k load tap as the best choice.
 
If it sounds good, don't worry about it. The oscillation thing got to be around the day you have the amp. If the amp always sounds the same, it's all good.
 
Here is why I don’t worry about the square wave wiggles, and said the amp is faithfully reproducing what is going in. This is 1 kHz square wave straight from Mac mini output. The amp is not ringing like that.
35D77ED8-B770-49FA-936B-AAC8F0D6BA07.jpeg
 
Had one, a Eico 377, it crapped out one day, I tried recapping it, but it turned out it wasn’t that simple and I was done putting money into it. Guess where the choke, power on lamp and several of the power supply filter caps went to. They were all consumed by this little amp in the conversion to solid state power supply phase of upgrades.
 
You must log in or register to reply here.
Back
Top Bottom