The Whammy is an excellent headphone amplifier kit. I am trying to determine if opamp rolling makes the Whammy more or less excellent. I am using it as a reference. The selection process applies to other audio amplifiers.
Test Setup:-- Whammy with BitScope.
The BitScope output is applied to the right channel input. The resistor load is a connected to the phone jack to monitor the channel. The BitScope outputs a 1khz sine wave. I set the scope cursors to track the output amplitude.
Next I applied a 50hz sine wave to the right channel input. It's perfect on the input side of C1.
But the square wave is sloped on U1-1 and the output jack.
The 20hz sine output is reduced slightly.
The 10hz even more.
Tools:-
I have used a variety of scopes without reading the fine manual but the BitScope is not intuitive. There are obvious button to select modes but some of the colored rectangles are also buttons and more. For example, the function generator has a brown widget where you can right click to select sine. square, or ramp.
You can select a frequency with the red rectangle and then drag the mouse to sweep the selected frequency. This device enforces RTFM.
Some BitScopes have a DSP for Arbitrary Waveform Generation which would be more useful than my old unit. I am aware of sophisticated tools such as REW software for checking noise and distortion, which might help with opamp selection, but I am not ready for that yet.
- Whammy complete kit
- JRC4080DD (came with), JRC2604AP, JRC4562D, OPA2604AP, JRC2068D
- BitScope Micro BS05U
- Two 15 ohms in series to load the right channel.
Test Setup:-- Whammy with BitScope.
The BitScope output is applied to the right channel input. The resistor load is a connected to the phone jack to monitor the channel. The BitScope outputs a 1khz sine wave. I set the scope cursors to track the output amplitude.
Next I applied a 50hz sine wave to the right channel input. It's perfect on the input side of C1.
But the square wave is sloped on U1-1 and the output jack.
The 20hz sine output is reduced slightly.
The 10hz even more.
Tools:-
I have used a variety of scopes without reading the fine manual but the BitScope is not intuitive. There are obvious button to select modes but some of the colored rectangles are also buttons and more. For example, the function generator has a brown widget where you can right click to select sine. square, or ramp.
You can select a frequency with the red rectangle and then drag the mouse to sweep the selected frequency. This device enforces RTFM.
Some BitScopes have a DSP for Arbitrary Waveform Generation which would be more useful than my old unit. I am aware of sophisticated tools such as REW software for checking noise and distortion, which might help with opamp selection, but I am not ready for that yet.
Oops! I maybe mistaken about the BitScope cursor usage in the 1kz to 20khz levels. I will check tomorrow.
Well I was not mistaken about the freq response which was flat from 50hz to 20khz and beyond when I tested again. There are some measurements done at the end of the build guide which I overlooked when the build was finished.
https://guides.diyaudio.com/Guide/WHAMMY+headphone+amplifier/3?lang=en
Step 30 shows the frequency response.
So the freq response alone gives no evidence of variance in my small stockpile of opamps. However, it suggests that the bass could be more existing by replacement of the input caps which are beside the ALPS volume pot.
Frequency response is just one piece of the puzzle that makes up the sonic characteristic of an amp.
Distortion, harmonics, noise, slew rate, bandwidth, frequency response, square wave response, etc... they all play their part and looking at just one is like characterising an amp by saying 'how loud' one is compared to another.
Distortion, harmonics, noise, slew rate, bandwidth, frequency response, square wave response, etc... they all play their part and looking at just one is like characterising an amp by saying 'how loud' one is compared to another.
As pointed out earlier I'd look at the THD+N vs frequency and THD+N vs output power. I'd also look at the nature of the distortion by measuring the frequency spectrum of the output. That's commonly done with an FFT. Your scope may not be able to resolve the differences between the different opamps.
Assuming the amp can handle it, you could try something like a uA741 as a worst case test. It should show pretty high distortion, low slew rate, etc. Then try a modern opamp like the LM4562 or OPA1656.
Tom
Assuming the amp can handle it, you could try something like a uA741 as a worst case test. It should show pretty high distortion, low slew rate, etc. Then try a modern opamp like the LM4562 or OPA1656.
Tom
Even a 12bit BitScope will not be enough to see differences between typical opamps. Although the uA741 may punch trough.
A 12 bit FFT can look down to some -72dB, a modern opamp is orders cleaner.
Overwhelmingly, the freq response is determined by the circuit, although you may see minor changes at the high end, far beyond the audio range.
Jan
A 12 bit FFT can look down to some -72dB, a modern opamp is orders cleaner.
Overwhelmingly, the freq response is determined by the circuit, although you may see minor changes at the high end, far beyond the audio range.
Jan
Not sure but it seems to me that edges of the squares shown in the OP are not clean. It would be interesting to see it with 10x shorter time base.
I think you see the limited bit depth of the scope.
You see it also on the function generator output.
jan
You see it also on the function generator output.
jan
Last edited:
Thank you for recent comments.
A scope is useful for consistent setup of input and output signals. But getting low noise platform is a challenge. That is a work in progress.
A scope is useful for consistent setup of input and output signals. But getting low noise platform is a challenge. That is a work in progress.
Yes, you really do need a scope for any serious audio work. It is a wonderful piece of equipment that can answer many questions.
But as will all relatively advanced equipment, it takes time to get to know what it can and what it cannot show.
I got my hands on a scope first when I started to work at Philips, I think it was a Tek 535 or 545, big noisy tube filled box.
I remember my wonderment on how I could change the picture with just a few knobs - timebase and level. Fascinating.
Now I have 4 or 5, all with different strengths and weaknesses. One of them even has a mouse, another one has WiFi.
Jan
But as will all relatively advanced equipment, it takes time to get to know what it can and what it cannot show.
I got my hands on a scope first when I started to work at Philips, I think it was a Tek 535 or 545, big noisy tube filled box.
I remember my wonderment on how I could change the picture with just a few knobs - timebase and level. Fascinating.
Now I have 4 or 5, all with different strengths and weaknesses. One of them even has a mouse, another one has WiFi.
Jan
I been reading about precision 1khz oscillators that are hard to diy. For example,
https://sound-au.com/articles/sinewave.htm#s2 .. there are other papers on this topic
Here is a cheap no-name, no-blame, that declares to be good for THD.
https://sound-au.com/articles/sinewave.htm#s2 .. there are other papers on this topic
Here is a cheap no-name, no-blame, that declares to be good for THD.