I would like some opinions about the HP 8903a analyzer. I don't have a distortion analyzer (realizing this does more than distortion) and there may be an opportunity to pick one up for under $300.00 that is in working condition. I also have the opportunity to check it out in person prior to purchase.
Thanks.
Thanks.
Distortion analyzers are nice as an adjunct to a good soundcard/interface. If you don't have the latter, I'd get that before the analyzer; the combo of a good soundcard/interface for spectrum analysis will allow reliable sub 0.001% measurement broken down by harmonic, as well as quantifying any anharmonic content. The distortion analyzer will then allow you to visually examine the residual as well as use the spectrum analyzer to further break it down. A useful combination of reasonably inexpensive instruments gives you a lot of measurement power.
Agree with SG, it's a good home lab grade unit and if it's in excellent shape, that's a good price.
Agree with SG, it's a good home lab grade unit and if it's in excellent shape, that's a good price.
I have an 8903B -- 'cuz I collect distortion analyzers -- Pete has written some code to get graphs with the instrument -- HP 8903 Software
Participatnts in the yahoo user groups "hp_agilent_equipment" and "tekscopes" exhibit a very high degree of knowledge on programming via the HPIB and GPIB.
Participatnts in the yahoo user groups "hp_agilent_equipment" and "tekscopes" exhibit a very high degree of knowledge on programming via the HPIB and GPIB.
$300 for an 8903A in good working condition?! That's a steal! I've refurbished and sold a few of them for twice that. Buy it!
I also recommend getting an operation and service manual if it doesn't come with one.
Connect a coax cable from generator out to analyzer in and measure the THD of a 1 kHz, 1 V signal. You should get below 0.01 %. Mine sits at 0.002 % as I recall.
~Tom
I also recommend getting an operation and service manual if it doesn't come with one.
Connect a coax cable from generator out to analyzer in and measure the THD of a 1 kHz, 1 V signal. You should get below 0.01 %. Mine sits at 0.002 % as I recall.
~Tom
Tricky to find, as they are not listed for sheet A13. Page 6-104 (page 97 in volume 1). They are TO-3 LM340-15 and LM320-15 which are equivalent to 7815 or 7915, but are rated at 1.5 A.
Well, it seems the -15v reg TO-3 is ok... there is/was a strange mechanical intermittent somewhere near the three electrolytic caps and the plug-in vreg board. As near as I can tell, something caused the crowbar to actuate, frying the -15v fuse! It would do this with a nice little tap on the board or the electrolytics... As of now, I seem to have eliminated the failure. If it happens again, I'll go back in and have at it again. I was suspicious of the connection to the cap opening up and causing an over voltage... or maybe a ground reference lifting, same effect.... dunno. Seems stable now after some poking about and touching up solder... Tnx for the input/info!
The loop through as of today was about 0.0029 iirc... seems like it was higher today than the first time I turned it on...
I am unsure of the effects of changing modes and levels upon the amplitude level. But I am mostly just pushing buttons to see how things work.
There may or may not still be some flaws in the operation. I'll have to run through the operation manual and see, then maybe the calibration proceedure...
I am wondering if the noise floor can be dropped by some better regulation and if the distortion measurement limit might be improved by putting in some new ultra low distortion opamps...
_-_-bear
The loop through as of today was about 0.0029 iirc... seems like it was higher today than the first time I turned it on...
I am unsure of the effects of changing modes and levels upon the amplitude level. But I am mostly just pushing buttons to see how things work.
There may or may not still be some flaws in the operation. I'll have to run through the operation manual and see, then maybe the calibration proceedure...
I am wondering if the noise floor can be dropped by some better regulation and if the distortion measurement limit might be improved by putting in some new ultra low distortion opamps...
_-_-bear
Bear,
Mine runs right at .0029, .003 as well. I would be quite interested to hear if you can improve things by rolling opamps. Please keep us posted.
Ken
I don't think that I did, but it's been a while since I messed with it. I purchased some extender boards but haven't used them yet. I got a used hp3562a spectrum analyzer and use it to look at the specific harmonics. So the 8903 has been relagated to more of a support role.
Bear,
The dynamic range is 0 to 100 kHz. I use Richiem Twin-T notch filter to null the fundamental about 50 db or so. http://www.diyaudio.com/forums/equipment-tools/188703-build-active-twin-t-notch-filter-distortion-analysis.html
The 3562 has a 85 db bandwidth so, if the fundamental is nulled to -50 then I can get down to -130 db resolution of harmonics. The practical limit is more like -125db. I haven't tried using the null of the 8903 for these measurements, but, now that you mention it, I will give it a try.
Initially I had difficulty using Pete's software, ended up buying a student version of VEE and Pete was kind enough to give me the un-compiled programs he created. Since then, I haven't had any issues and have written VEE scripts to run the 3562 in all sorts of neat ways. The only limitation of the student edition is that I can't compile the scripts.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.