I do not recall if i changed input level to see the shape for a photo comparison orif it actually has some gain in the active. Note scale on left and right side are different. But I can change opamp gain to be same as passive or to zero, I suppose. Might drop the opamp thd also by same amount. I think it is a LME49990 IIRC.... might be unstable at Gain =1? I did this mod but have never had to use it in either mode. The hardware I now have is so good. But, have the notch If I need it.
On that -140dB ---- that is instrument limit. When the residual is below -140 then the screen goes blank instead of giving junk data numbers. I havent checked the accuracy between -130 and -140dB. It isnt bad though from many comparison tests to other test arrangements and gear. The internal noise of the 725 filters/analyzer/display seems to be higher than the 7722A. On either i can go to the monitor output port to a QA401 FFT for further eval.
THx-RNMarsh
Last edited:
Hi Richard,
I honestly don't know if measurements down that low are any use except as relative measurements to see if we are on the right track or not when making modifications to equipment. I guess you could see if the readings are repeatable. That would give you some measurement of confidence. Seeing spikes down there might be very difficult as most of the residual would be noise. Is that what you are seeing Richard?
-Chris
Yes, that was exactly what I was thinking
I honestly don't know if measurements down that low are any use except as relative measurements to see if we are on the right track or not when making modifications to equipment. I guess you could see if the readings are repeatable. That would give you some measurement of confidence. Seeing spikes down there might be very difficult as most of the residual would be noise. Is that what you are seeing Richard?
-Chris
Yes Chris you are right. However it does lighten up the uncertainty when we measure -120dB on our audio equipment. Otherwise who knows. We can only speculate.
Hi Richard,
Yes, absolutely. That would be very useful in my case as I'm only seeing noise for some of my projects. I hate not knowing!
If I can squeeze another 10 - 20 dB of range in the bottom end out of my 339A, that would tide me over. I guess it's coming close to the day when a 339A gets torn down for a restoration and improvement. I was thinking of using the AM Detector switch and BNC jack for a low noise monitor jack ... or the original jack can be repurposed for that job. Then I could use the AM jack as a switch location for the extended range.
-Chris
Yes, absolutely. That would be very useful in my case as I'm only seeing noise for some of my projects. I hate not knowing!
If I can squeeze another 10 - 20 dB of range in the bottom end out of my 339A, that would tide me over. I guess it's coming close to the day when a 339A gets torn down for a restoration and improvement. I was thinking of using the AM Detector switch and BNC jack for a low noise monitor jack ... or the original jack can be repurposed for that job. Then I could use the AM jack as a switch location for the extended range.
-Chris
The SOIC is 4.65USD or so. The part is reasonable but the shipping is not.
I only want a couple for R&D and don't want to be stuck with more.
An order of 100 would water down the shipping cost.
Chris if you have two 339A you can use one to improve the resolution of the other.
The 339A is one of the analyzers that can use a passive or active notch filter at the input.
You can tap the output of the notch in one and route it to the AM BNC. Use this signal for the input to another. Mix some fundamental back into the notched signal for the second analyzer to function on.
Hi David,
I'll have to think on that one. There is a space thing on the bench too. Mind you, two 339As take up less room than a Sound Technology 1700 series analyser.
One 339A is for day to day use. The other is supposed to be a backup for the first because a distortion analyzer is absolutely essential for the work I am doing. I decided to modify the backup 339A and swap it with the first one. If the mods work well, I'll do the old primary generator as well. Chances are that the work will go a lot faster and be neater since I had done one already. I'm not certain what I might do with the power supplies. Having a low noise power supply is far more important than some people might think. So that is a mod I will need to develop first.
-Chris
I'll have to think on that one. There is a space thing on the bench too. Mind you, two 339As take up less room than a Sound Technology 1700 series analyser.
One 339A is for day to day use. The other is supposed to be a backup for the first because a distortion analyzer is absolutely essential for the work I am doing. I decided to modify the backup 339A and swap it with the first one. If the mods work well, I'll do the old primary generator as well. Chances are that the work will go a lot faster and be neater since I had done one already. I'm not certain what I might do with the power supplies. Having a low noise power supply is far more important than some people might think. So that is a mod I will need to develop first.
-Chris
Okay. It's not very invasive to the 339A to do this and certainly reversible. Just a bit of
time. Let me know what you want to do. We can do this offline.
I tend to believe that the attention to detail needed to get a piece of equipment to perform at such low distortion levels tends to result in a better-sounding product. In other words, it may not be the distortion itself per se, but rather that the attention to detail may mitigate other effects that may degrade the sound.
Also, I tend not to get as excited about 1 kHz THD at such low levels, since in most designs there is a huge amount of feedback loop gain available at 1 kHz. There will usually be at least 20 times less loop gain at 20kHz. From a very simplistic view, then 20 kHz THD might be 26dB higher than 1 kHz THD.
So if you are good enough to get your amp down to -110 dB at 20kHz (0.0003%), then at 1kHz it might be at -136dB.
Cheers,
Bob
Hi Bob,
I agree with you. But taking that much time and effort with an amplifier build isn't going to be a mass production situation. So there is one place where a small company can do better than the big ones with a nice lab. That's once they get into the same ballpark that the bigger company is playing in.
-Chris
I agree with you. But taking that much time and effort with an amplifier build isn't going to be a mass production situation. So there is one place where a small company can do better than the big ones with a nice lab. That's once they get into the same ballpark that the bigger company is playing in.
-Chris