Low Distortion Signal Generator

Jump to Latest
#61
I built Jim Williams oscillator, spent many hours making it work. Its a lot more tricky than it looks to be. I wasn't happy with the final outcome and just had to stop.

There is another source for the optocouplers, low distorion ones in fact. http://www.silonex.com/audiohm/index.html I found them after I moved on to other things. If I didn't have the KH 4400 or the Boonton 1120 I would still be looking at this.

The 10 KHz oscillator looks interesting. I don't have time to build one but I could hook it to the CLT1 and measure the 3rd harmonic accurately. It would be interesting.
-Demian
 
#65
The DM4084 is stand-alone (there is another model that is a 500-series plug-in, I can't remember at the moment). Technically, I guess the DM4084 is kind of a plug-in too, as it releases the same way a regular plug in does, but it cannot be used in a 500-type plug in, and 500-type plug-ins cannot be used in the power supply that comes with the DA4084, or at least this is the way I was told and the way I understand it. The SG505 is the only one in the TM501, obviously.😉

I wish I had bought a TM502, and I could look for a FG501 or a 501A or a FG502, and sold that cheapie CFG253.

If I do decide to buy a FG plugin, which one ought I to look for??
 
#67
EchoWars said:
The DM4084 is stand-alone (there is another model that is a 500-series plug-in, I can't remember at the moment). Technically, I guess the DM4084 is kind of a plug-in too, as it releases the same way a regular plug in does, but it cannot be used in a 500-type plug in, and 500-type plug-ins cannot be used in the power supply that comes with the DA4084, or at least this is the way I was told and the way I understand it. The SG505 is the only one in the TM501, obviously.😉

I wish I had bought a TM502, and I could look for a FG501 or a 501A or a FG502, and sold that cheapie CFG253.

If I do decide to buy a FG plugin, which one ought I to look for??
Click to expand...

Just parenthetically, sometimes it's better to donate your unwanted equipment than to sell it (from a tax standpoint) -- figure this, the tax benefit is (1- tax rate) * fair market value = tax benefit. Fair market value is likely to be a lot higher than the scrap value.

Not that we give stuff to 501C-3's just for the tax benefit, but it is something to consider. I've outfitted an entire inner city high school electronics lab just with discarded Tek PS503's and DM501's -- and they are a lot better than the stuff they get from Pascoe or Vernier. If the economic benefit doesn't quite make it, at least some under-equipped high school teacher will think more kindly of you!
 
#71
Great news Jackinnj, that looks very nice! Offcourse the big question is... can I buy a PCB from you... ... and.. hahaha I'll be happy to check your finished version for you 😎 .... but seriously, I been wanting to build the superoscilator for over a year now, but the lack of a pcb has hold be back.. I would be very interested in buying a PCB from you...

Kindest regards,
Thijs
 
#72
ghg said:
Perfect, and the question is ....

Do you have access to an AP2 to proof the specs ?

BR

Gary G.
Click to expand...

You mean 2700, or ATS-2? The 2700 will measure down to 0.00025%

follow the procedure that Bateman used with his ULD oscillator in EW --- a passive 60 dB notch filter followed by an active notch filter.

My Boonton 1120 and Tektronix SG505 can both be tweaked to
~ 0.0007%.
 
#73
Thijs -- there are a couple tweaks to do here -- the 50k amplifier "gain adjust" pot should be 10 or more turns -- this adjustment is a little "nervy". I have used ceramic caps since that was all I had on hand -- will see if i can fit some polystyrene ones (on order) in their place. I used an SMA connector for this board -- I am sure that most folks would like BNC -- and I am further sure that 1kHz would be preferred to 10kHz -- but 10kHz has its own challenges and rewards.

you have seen this site -- http://www.amplifier.cd/Technische_Berichte/Amplifier_reiner_Sinus/amplifier_clean_sine.htm
 
#74
some hints -- I have made some measurements -- not in a shielded case yet (and using a noisy TEK PS5010 supply) -- all of the filter circuits should be 1% metal film, all of the filter capacitors should be polystyrene (you could use mica I suppose) -- and the 3 trimmer pots should be 12 turns -- we are getting down into the 0.00X% THD readings which will improve when it is shielded and has a noiseless power supply -- this really does amplify anything nearby !!!

when I simmed the device's circuits in Multisim I get a 11.3 kHz -- crossover -- in reality the device is oscillating at 7.6kHz. -- curious -- that's roughly a factor of 1/sqrt(2)
 
#76
Hi Jackinnj,

Sorry,.. haven't been able to post in a while... i'm terrible busy at my work... but I'm reading everything .... I got the last IC for the uber-oscilator last month or so... nothing left to keep me from building one this summer holiday.. keep us posted on your results..


mean while I managed to buy a Krohn-Hite 4400A oscilator from ebay.. low thd, fully adjustable frequency and amplitude.. should be nice....

cheers m8,
Thijs
 
#78
i briefly got it down to where it should be operating -- actually the Boonton (which is in cal) said 0.00012% -- perhaps for a second or so -- then it reverted to 0.0029% -- it's running on 10 "D" cells now -- i am not happy with the stability -- will run for 30 minutes or so at very low THD and then break out.
 
#79
Looks like you made more progress that I did on the oscillator. I still don't believe they built it in an afternoon.

I have never been able to get any of my Boontons below .0007%. I believe its a noise limit more than anything else. I have also noticed that its possible for the measurement routine to "lockup" and require a restart (press the local button to reset the system)

I am attaching the distortion residuals from both the internal oscillator in the Boonton and the KH4400. These are FFT's of the output of the Boonton analyzer

Boonton:
 

Attachments

  • boonton dist.jpg
    boonton dist.jpg
    32.9 KB · Views: 957
#80
Here is the KH4400 (I haven't figured out how to attach two images to one post yet.)

Both read .0008% THD w/ 30 KHz low pass filtering, using inverting input and all of the other tricks I have figured out for lowest readings.

The display is 0-25KHz and the marker is at 2 KHz.

Krohn Hite:
 

Attachments

  • kh4400 dist.jpg
    kh4400 dist.jpg
    33.6 KB · Views: 949
You must log in or register to reply here.
Top Bottom