I have been searching for a ShibaSoku signal generator to no avail. Meanwhile I picked up an H-Hite 4400A and a 4402B generator.... for IM tests. Might see if they can be lowered in THD also. After I measure them with ShibaSoku 725D analyzer. BTW - before the QA400 left this cold place for somewhere warmer in the universe -- the null used on the 725D appeared to be at least -150dB. !!
Thx-RNMarsh
Thx-RNMarsh
I have a kh4402b, manual has a simple proceedure for adjusting/tuning. I would be interested in any improvements beyond tuning. Please keep us posted.
Ken
I tweaked mine and got a substantial improvement on the second harmonic. I have been reluctant to change parts since its quite good. The harmonics were all below -110 dB I believe.
I added the injection lock and that really helps the close in phase noise when locked from a clean source. Too much and the harmonics start going up. I'll find the notes I made and post them if anyone is interested.
I added the injection lock and that really helps the close in phase noise when locked from a clean source. Too much and the harmonics start going up. I'll find the notes I made and post them if anyone is interested.
K-H 4400A
Just for the heck of it, I put in a socket and changed opamps for the oscillator IC in a K-H 4400A. I monitored the output with a 300Mhz scope. Some didnt work well at all.... the LM49710 didnt like the circuit values, i suppose... distortion was much higher than others. BTW - I had to replace the distortion null pot from 500 to 1K Ohms to get the range needed to null some of the other opamps I tried.
The LM318 has pretty low thd across the board but a lot of harmonics.... odd and even. The LT1468 was near the floor of -145dB --- about -130-135 everywhere except a high 3H. [Relatively high at -125dB].
This is a quick way to find out how your favorite opamp looks with its harmonic structure.... Testing is at no load for now -- 1Khz fundamental used. Residual taken from the ShibaSoku 725D (QA400 was doing this fine but it quit on me. I'm back to the analog HP until I can get QA400 working again).
Just for the heck of it, I put in a socket and changed opamps for the oscillator IC in a K-H 4400A. I monitored the output with a 300Mhz scope. Some didnt work well at all.... the LM49710 didnt like the circuit values, i suppose... distortion was much higher than others. BTW - I had to replace the distortion null pot from 500 to 1K Ohms to get the range needed to null some of the other opamps I tried.
The LM318 has pretty low thd across the board but a lot of harmonics.... odd and even. The LT1468 was near the floor of -145dB --- about -130-135 everywhere except a high 3H. [Relatively high at -125dB].
This is a quick way to find out how your favorite opamp looks with its harmonic structure.... Testing is at no load for now -- 1Khz fundamental used. Residual taken from the ShibaSoku 725D (QA400 was doing this fine but it quit on me. I'm back to the analog HP until I can get QA400 working again).
Last edited:
Hi Richard,
work has been a Bi... the last few months. Not to complaint. I have work.
It is my intention to open a QA400 thread. I have only had time to hook up my QA400 in loop back mode. I'm pretty sure there is a SW bug in the generator SW.
However, do you know what killed your, if that is what happened?
Btw, I loved your Audio Amatuer articles. In the late 80's I thought I had invented the intput stage you used and the folded cascode, which I later discovered I had not
Best regards,
Mogens
work has been a Bi... the last few months. Not to complaint. I have work.
It is my intention to open a QA400 thread. I have only had time to hook up my QA400 in loop back mode. I'm pretty sure there is a SW bug in the generator SW.
However, do you know what killed your, if that is what happened?
Btw, I loved your Audio Amatuer articles. In the late 80's I thought I had invented the intput stage you used and the folded cascode, which I later discovered I had not
Best regards,
Mogens
Please let us know when you open the thread on the QA400. I have just tried the latest beta software and it seems to work pretty well. I have had a few lockups mostly related to USB. All of the features seem to work. My particular QA400 seems to have something odd on its input Z, it measures 10K, not 100K and that caused some confusion when I was calibrating it.
Attached is a preliminary schematic for an interface to make the QA400 compatible with higher level signals etc. in expresspcb. The layout should be done the beginning of the week. Not shown are the cables needed to interface, I'll draw those up this week as well.
Attached is a preliminary schematic for an interface to make the QA400 compatible with higher level signals etc. in expresspcb. The layout should be done the beginning of the week. Not shown are the cables needed to interface, I'll draw those up this week as well.
Hi Richard,
work has been a Bi... the last few months. Not to complaint. I have work.
It is my intention to open a QA400 thread. I have only had time to hook up my QA400 in loop back mode. I'm pretty sure there is a SW bug in the generator SW.
However, do you know what killed your, if that is what happened?
Btw, I loved your Audio Amatuer articles. In the late 80's I thought I had invented the intput stage you used and the folded cascode, which I later discovered I had not
Best regards,
Mogens
I think it was when i unplugged the usb at some important time for the QA400. Its been very quirky ever since. I cant get it to display data. and it indicates drop outs frequently. It says the usb is connected. Wont cal - though there is output. Messed up for sure.
The SW could be a bit smoother in operation. But the manual is pretty complete.
Having said that, I dont want to give the impression it is a bad product and I do have the old favorite emu0204 and arta software as back up. The QA400 measures at least 6dB better and the integration is very convenient for me. So, I went ahead and ordered one. The first one was a demo/eval unit that I didnt pay for.
Glad you liked the articles... I had the drive to work, play, build, test and write back then. Now I have more free time and might get to put down some more info once again.
Thx-RNMarsh
Last edited:
What your describing is a USB problem. Either a problem with the USB port on you comp or a USB driver problem. You can down load USB drivers from cypress semiconductor. Look for CYUSB.SYS. There is also a forum on Cypress's website discussing driver problems and resolutions. Remove all old QuantAsylum usb drivers before installing new.
Cheers,
I am still working on the "USB" problem. Meanwhile, I got a loan of a ShibaSoku AG16B oscillator from Demian when in San Fransisco yesterday. On the newest ShibaSoku analyzer, I measured the THD+N on second (at 1v rms out into high Z) as .00007% for 2H and .000025% for 3H. Using an HP3580A swept analog filter, the noise floor from the analyzer output and the HP is about -140. The harmonics are at that floor..... a 9H showed up at -143.
When I can average out the noise with the QA400 FFT, I can get a more accurate finding on the Harmonics but this is much better than the KH4400A. Or, the HP339A. A KH4402B is coming up next but this ShibaSoku combo is what I need to reach my goal of -140dB THD+N (withOut FFT).
Thx-RNMarsh
When I can average out the noise with the QA400 FFT, I can get a more accurate finding on the Harmonics but this is much better than the KH4400A. Or, the HP339A. A KH4402B is coming up next but this ShibaSoku combo is what I need to reach my goal of -140dB THD+N (withOut FFT).
Thx-RNMarsh
The Shibasoku oscillators are so much better than the KH4400 I first thought something was broken when measuring them with the 725. I don't think you will get to the AG16 performance with something conventional. I have another older version of the oscillator with identical performance. At this level its time to move on and just use the stuff.
I'm almost done with the QA400 interface. I'm getting it checked and will send out for boards by the weekend I hope. The next challenge is to make a lockable oscillator and matched notch filter to get close to this level of performance with a QA400.
I'm almost done with the QA400 interface. I'm getting it checked and will send out for boards by the weekend I hope. The next challenge is to make a lockable oscillator and matched notch filter to get close to this level of performance with a QA400.
Hi Richard,
I have been following this thread, and it is a great one. It looks like some great oscillators, and ways to measure their very low distortion, have been described.
It looks like most of the work has been done at 1kHz, which is entirely appropriate. However, very low frequencies and high frequencies are where it is much more difficult to get very low distortion. As we all know, at low frequencies like 20 Hz we have the problem of AGC ripple if we are to have reasonable settling times. At high frequencies we have more operational-amplifier-related problems like THD etc.
Do you have some data on this oscillator's THD and settling performance at 20Hz?
Similarly, do you have some distortion measurements for high frequencies? I realize that with analyzers like the 3580A and even really good sound cards we may only be able to see spectra out to about 50kHz. So can you tell us about 2nd and 3rd harmonic distortion with a fundamental of, say, 15 or 16kHz?
Cheers,
Bob