[volt-nuts] Fluke 510A

Dick Moore richiem at hughes.net
Fri Jan 22 19:56:33 UTC 2010


Hi, Bill -- thanks for the observation re the integrator; I'll look into that in a couple of months, when we get back from Hawaii. My distortion analyzer can read reliably to better than 0.0005% at 400Hz, ie 5ppm, and I've set the 400Hz 510 for about 0.002%, which seems to be pretty good for both frequency and level -- I can see however that that may be too fine a set-point, perhaps resulting in a little metastability problem around the set-point. 

I haven't quantified the readings in detail for level since my last cal of the 510, when I fooled around with the various adjustments. I had it set for around +30ppm (the tweak is very dodgy at these levels) according to the 3458; now it's over +500ppm, about three months later, and varies a lot with the night/day cycle of the heat in my shop. I wonder how tight the control loop is around the peak-detector and whether that is actually the source of the variation? In any case, I had to set the DC reference higher than the rec'd value to get 10V out -- I'll have to look at my notes to see how much higher. So that may indicate leakage at the integrator....

Best,
Dick Moore
> 
> Message: 2
> Date: Thu, 21 Jan 2010 17:15:40 -0800
> From: WB6BNQ <wb6bnq at cox.net>
> Subject: Re: [volt-nuts] First post
> To: Discussion of precise voltage measurement <volt-nuts at febo.com>
> Message-ID: <4B58FC3B.AC5BD168 at cox.net>
> Content-Type: text/plain; charset=iso-8859-1
> 
> Hi Dick,
> 
> Well, first the obvious question.  Have you gone through the complete calibration procedure ?
> 
> When you say ". . . doesn't actually result in a 10VAC output," just how far off do you think it is ?
> 
> Have you looked at the output waveform on the scope ?  Actually you should measure it with a distortion analyzer, if you have one.  If it isn't clean that would through off the RMS value.  It could also, possibly, fool the peak detector.
> 
> If you think about it, the 14.142136 is 10 times the square root of 2 (close enough anyway) and would be the precise peak value of the 10 volt RMS level (RMS being 1/2 SQRT of 2 times the peak value).  If the waveform is "clean," perhaps there may be an issue with the integrator that follows the peak detector.
> 
> Also, your statement about the 510 ". . . which seem very drift prone and highly temperature sensitive," how much variation are you are seeing ?
> 
> Again, I am wondering if there may be an issue with the integrator.  Perhaps the integrating capacitor has a problem or maybe there is some additional leakage around the integrating capacitor and the input of the integrator circuit that should not be there.  It could have been sprayed with something that has left a film that reacts to moisture.
> 
> If that is the case, then spraying it with a good soap like 409, a little brushing and rinsing with distilled water should take care of that.  Just make sure not to get the transformers wet.  If they do get a little wet, then baking it out in a cardboard box with a 100 watt incandescent light as a heat source for 2 to 3 days.  Obviously you may want to have some venting of the box to control the temp so the plastic parts do not melt.  But you should be able to have around 150 degree F without a problem.
> 
> Bill....WB6BNQ
> 
> 
> 
> Dick Moore wrote:
> 
>> Hi Bill -- The HP3458A has an AC measurement mode that for frequencies of 50kHz or less does an A-D conversion of sample points then calculates the RMS value, instead of the usual mode of rectifying the signal and the calulating the RMS. The outstanding linearity of the A-D in the 3458 means that it is very accurate, with much lower uncertainty than the 510s, and the cal data for the 3458, which is fairly recent tends to support this. So, I'm using it to look at the 510s outputs. I'm also a bit mystified why setting the 510s internal reference to the recommended 14.142136 VDC doesn't actually result in a 10VAC output; The peak detector-comparator circuit seems to have pretty good performance. I've tried varying the internal adjustments of the 510 one at a time to gauge effects on output level, and the only one that really seems to matter is the reference level setting. Any light youc an shed would be welcome.
>> 
>> Ulrich, thanks for the kind offer. I hope to get into GPIB in the late spring or early summer.
>> 
>> Best,
>> Dick Moore




More information about the volt-nuts mailing list