Fluke 8050A - AC Measurement

[rstor22]

On my Fluke 8050A I noticed that if I apply a low voltage AC signal (7-8 volts @ 60Hz) and measure it on higher ranges, such as
the 200V or 750V range the readings are way off (by several volts). I noticed that if I measure a higher voltage such as mains 120V on the 200V range it reads accurately (around 120V) and on the 750V range it reads about 117V. On my Fluke 8600A and 8800A low AC voltages appear to read correctly even on the higher ranges.

Has anyone also experienced this on their 8050A unit?
(I realize there is no point in measuring lower voltages on these higher ranges, I just wanted to make sure my unit is functioning
correctly)

In the manual it specifies an input voltage for the 200V range as 10-200V and for the 750V range 100V-750V. Is it true that there is a minimum voltage that can be measured on higher ranges? I thought you just lost resolution on the higher ranges and there was no minimum voltage...

[retiredcaps]

I thought you just lost resolution on the higher ranges and there was no minimum voltage...

I don't have a 8050A, but seeing how your other two Flukes read correctly at the higher range, I would say something might be wrong with the 8050A.

Suggestions.

1) I'm not a fan of the ganged switches and I have one 8020A that acts funny when switching between voltages and ohms.  When you switch the 8050A to a higher voltage, perhaps there is contamination or dirt/debris that is causing low readings?  Cleaning it might help?

2) Check the voltage divider to make sure you have the proper ratios when switching to the higher ranges.

3) Check the pcb for dirt/contamination.

4) Does the same problem happen in the DCV range?

5) Does the same problem happen in the ohms range?

6) Since I don't have a 8050A, I haven't spent any time reading the service manual, but the older service manuals had excellent troubleshooting sections that might yield some clues.   I'm going to download and read it for my own learning.

[retiredcaps]

I had a quick look at the 8050A service manual

http://assets.fluke.com/manuals/8050a___imeng0200.pdf

Table 4-9 shows AC calibration and the procedures required.  I'm not saying your meter is out of calibration, but for the 200V range, C2 is the adjustment.  Check C2 on your pcb to see if there is any obvious dirt, debris or contamination.

I bought an used 8060A and 5 out of the 10 electrolytic capacitors had leaked from the bottom.  I replaced them all, but my ACV reading was way off.  After several days of troubleshooting, with help from modemhead, I finally decided to turn the ACV adjustment pot.  A couple of turns left and right all the way, ACV was measuring properly.  One electrolytic caps near the ACV adjustment pot had leaked it electrolyte and wicked its way up the pot.  It wasn't visible to the naked eye and I think turning that pot left and right a couple of times got rid of that dried up electrolyte?

Anyway, that should give you enough ideas, tests, etc to get started with so we can narrow down the problem (if this is a problem on the 8050A since I haven't read the entire manual yet).

PS. modemhead has a 8050A (or two).

[ModemHead]

All RMS converter circuits have offsets that limit their accuracy on the lower end of their input range.  If you look at the 8050A specs, you'll see that its accuracy is specified for values above 5% of the range.  That's 10V for the 200V range, and 38V for the 750V range.

For reference, my 8050A measures a 5.00V 60Hz sine at 4.97V on the 200V range, and 4.6V on the 750V range.

If yours seems too far off, you could try going through the AC cal and RMS converter offset adjustment procedures.  You'll need a source for *stable* AC voltages and a trustworthy reference meter.  You can get by with a function generator or computer sound card output for the low frequency, low voltage bits, but unless you have means of generating stable 19VAC and 100VAC at 10kHz, leave the capacitive trimmers alone.

[rstor22]

1) I'm not a fan of the ganged switches and I have one 8020A that acts funny when switching between voltages and ohms.  When you switch the 8050A to a higher voltage, perhaps there is contamination or dirt/debris that is causing low readings?  Cleaning it might help?

2) Check the voltage divider to make sure you have the proper ratios when switching to the higher ranges.

3) Check the pcb for dirt/contamination.

4) Does the same problem happen in the DCV range?

5) Does the same problem happen in the ohms range?

I've checked the DCV and the ohms range, and they appear fine. I would therefore think that the voltage divider and switch contacts are fine. Attached are my results.
I also opened the unit and it looks very clean inside.

Table 4-9 shows AC calibration and the procedures required.  I'm not saying your meter is out of calibration, but for the 200V range, C2 is the adjustment.  Check C2 on your pcb to see if there is any obvious dirt, debris or contamination.

I looked through the C2 calibration hole and from what I can see, I do not notice and dirt or debris.

All RMS converter circuits have offsets that limit their accuracy on the lower end of their input range.  If you look at the 8050A specs, you'll see that its accuracy is specified for values above 5% of the range.  That's 10V for the 200V range, and 38V for the 750V range.

When the manual mentions the AC accuracy and indicates between 5% of range and full range is that how you arrived at the 10V for 200V range and 38V (37.5) for the 750V range?  Does this mean that for the 200V range, the AC accuracy specified is only valid for measurements above 10V on the 200V range and above 37.5V for the 750V range?

Also  when the manual indicates "input voltage" for the 200V range as 10-200V and for the 750V range 100V-750V what does this mean?

For reference, my 8050A measures a 5.00V 60Hz sine at 4.97V on the 200V range, and 4.6V on the 750V range.

If yours seems too far off, you could try going through the AC cal and RMS converter offset adjustment procedures.  You'll need a source for *stable* AC voltages and a trustworthy reference meter.  You can get by with a function generator or computer sound card output for the low frequency, low voltage bits, but unless you have means of generating stable 19VAC and 100VAC at 10kHz, leave the capacitive trimmers alone.

Thanks for these reference voltages. My attached measurements are off compared to yours. Especially on the 750V range. I noticed in the manual for AC calibration (table 4-9) does not mention an adjustment for the 750V range. It mentions C1 and C2 are interacting adjustments. Would C1 and C2 also adjust the 750V range?

Though I have a function generator (amplitude seems to slowly increase on one, and the amplitude on the other varies slightly back and forth), I do not have a means of producing a stable high voltage for C1 and C2.

Perhaps at this point it is best to leave it as is and  I should just make a note that the AC 200V/750V ranges are out of calibration and leave it at that....
Or do you think it could be something defective in the true RMS module?

[ModemHead]

It might be helpful to refer to Section 2-34 "Insignificance of Inherent Meter Offset" on page 2-5 of the 8050A manual.

RMS converters produce a DC voltage proportional to the RMS value of the input.  The output will always have a small DC offset due to circuit imbalances.  This is an error term which becomes less and less significant as the applied input voltage goes up.  For that reason it is not appropriate to use the relative offset feature (REL) to get rid of the error.  It is also why accuracy is specified from a certain percentage of full range, up to full range.  Fluke is saying 5% for this lower limit, which basically means at least 10mV on the 200mV full-scale A/D converter should swamp any errors.

It was a mistake on my part earlier to use 750V as the full-range value, because that's basically an imposed safety limit.  For the highest range, full range is mathematically 2000.0V, even if it might arc over or exceed component ratings.  5% of 2000V is 100V, so to stay in band, it would be 100V and up.  Likewise, 10V and up for the 200V range.

From my DIY/hobbyist experience, the meter will be sufficiently calibrated for mains-frequency AC on all ranges with the two interacting potentiomenters R7 and R29 and the 100mV and 1.9V test inputs.  The tolerance and stability of the voltage divider resistors should take care of higher ranges.

At low frequencies, the resistive components of the voltage divider dominates its impedance. But as frequency increases, reactance components from stray capacitances start to become non-negligible.  The C1 and C2 trimmers are there to compensate, and that's why the procedure calls for 10kHz inputs for their adjustment.  I do recommend leaving those alone.

Your 8050A is actually within spec at 5VAC and 110VAC on the 20V and 200V ranges, assuming the 8800A is golden, which of course is not guaranteed.  It still might be interesting to try checking at the 100mV and 1.9V AC levels.  A little tweak might improve the "downrange" readings.  And for the RMS converter, check the reading with shorted inputs in the 2VAC range, the book says it should be less than 40 counts from zero.  Mine is 3 counts from zero (0.0003V).

[rstor22]

It might be helpful to refer to Section 2-34 "Insignificance of Inherent Meter Offset" on page 2-5 of the 8050A manual.

I was unable to locate this section in the manual : http://assets.fluke.com/manuals/8050a___imeng0200.pdf
Perhaps you have a different revision?

I found on page 2-19, section 2-98 the section "Affects of Offset in AC Measurements". Would this be the area you are referring to?

RMS converters produce a DC voltage proportional to the RMS value of the input.  The output will always have a small DC offset due to circuit imbalances.  This is an error term which becomes less and less significant as the applied input voltage goes up.  For that reason it is not appropriate to use the relative offset feature (REL) to get rid of the error.  It is also why accuracy is specified from a certain percentage of full range, up to full range.  Fluke is saying 5% for this lower limit, which basically means at least 10mV on the 200mV full-scale A/D converter should swamp any errors.

It was a mistake on my part earlier to use 750V as the full-range value, because that's basically an imposed safety limit.  For the highest range, full range is mathematically 2000.0V, even if it might arc over or exceed component ratings.  5% of 2000V is 100V, so to stay in band, it would be 100V and up.  Likewise, 10V and up for the 200V range.

So if I am to understand correctly, each AC range has a floor (i.e. 10mV for 200mV, 0.1V for 2V, 1V for 20V, 10V for 200V, and 100V for 750V). If one was to measure a value below the floor for the respective range, then Fluke does not guarantee what that reading would be, correct?

Your 8050A is actually within spec at 5VAC and 110VAC on the 20V and 200V ranges, assuming the 8800A is golden, which of course is not guaranteed.  It still might be interesting to try checking at the 100mV and 1.9V AC levels.  A little tweak might improve the "downrange" readings.  And for the RMS converter, check the reading with shorted inputs in the 2VAC range, the book says it should be less than 40 counts from zero.  Mine is 3 counts from zero (0.0003V).

I took some more readings at around 10V and 15V on the 200V range. It appears (assuming the other meters are accurate), that the 8050A is out of spec (see attached).
Also attached are the results for 100mV and 1.9V input (unfortunately the function generator does not appear to be stable to I recorded the lowest reading on each meter.)

With the inputs shorted on the 2VAC range I get -.0013

[ModemHead]

I found on page 2-19, section 2-98 the section "Affects of Offset in AC Measurements". Would this be the area you are referring to?

Sorry for the confusion, I have a hardcopy and it is Rev 1 8/80, the PDF is Rev 2 7/84 .  Yes section 2-98 is the relevant part, but sadly it seems they cut out a lot of the verbage and made it a bit more obtuse.  But the point is they're showing how the meter still displays "10.00" for a 10mV input even with a "typical" offset of 20 counts.  The formula is Vrms = SQRT(DC^2 + AC^2), ie. the square root of the sum of the squares of the DC and AC components of the signal.

So if I am to understand correctly, each AC range has a floor (i.e. 10mV for 200mV, 0.1V for 2V, 1V for 20V, 10V for 200V, and 100V for 750V). If one was to measure a value below the floor for the respective range, then Fluke does not guarantee what that reading would be, correct?

You got it.

I took some more readings at around 10V and 15V on the 200V range. It appears (assuming the other meters are accurate), that the 8050A is out of spec (see attached).
Also attached are the results for 100mV and 1.9V input (unfortunately the function generator does not appear to be stable to I recorded the lowest reading on each meter.)

With the inputs shorted on the 2VAC range I get -.0013

The 100mV / 1.9V cal values in the 200mV / 2V ranges look really good compared to the 8800A, but it definitely drops off fast as you go up from there.  I'd say something is amiss, but I don't know what the cause could be.  Maybe it's worth doing the RMS Converter Offset check/adjustment on pg 4-11, even though 13 counts offset at zero is well within limits.  You'll need a stable source of 1V AC @400Hz.  If the function generator isn't stable enough, try using Audacity or similar software to generate a sine wave and play it through a computer sound card.

If the RMS convertor checks out OK, then you might be back at the beginning, ie. dirty switch contacts or PCB.  It doesn't have to "look" dirty, the slightest bit of contamination can cause leakage currents.

[rstor22]

The 100mV / 1.9V cal values in the 200mV / 2V ranges look really good compared to the 8800A, but it definitely drops off fast as you go up from there.  I'd say something is amiss, but I don't know what the cause could be.  Maybe it's worth doing the RMS Converter Offset check/adjustment on pg 4-11, even though 13 counts offset at zero is well within limits.  You'll need a stable source of 1V AC @400Hz.  If the function generator isn't stable enough, try using Audacity or similar software to generate a sine wave and play it through a computer sound card.

If the RMS convertor checks out OK, then you might be back at the beginning, ie. dirty switch contacts or PCB.  It doesn't have to "look" dirty, the slightest bit of contamination can cause leakage currents.

I used Audacity  as suggested and output a 400 hz .99996V to 1.00021V RMS signal. I followed the procedure on pg 4-11 and the voltage on pin 7 of the RMS module was measured at -4.6 mV DC (step 6). The voltage at pin 6 was measured at -3.8 mV DC (step 7). It mentions if the voltage is greater than +/-0.5mV of the recorded value in step 5 (I believe they meant step 6) to perform adjustment. In my case the difference is 0.8mV. I think at this point I'll leave it as I do not plan to measure low value AC voltages on the higher ranges (i.e. 200V and 750V ranges).

When I used my laptop to output a signal via the sound card I found that if I left the laptop plugged in the measurements on the counter were incorrect on some units. I have two function generators (Sampo FG1617 and Instek FG-8016G, and a BK 1801 Frequency Counter). The Sampo unit BNC is not tied to earth gorund, the other two units are.

I got the following results when outputting a 200 hz signal:

Laptop Plugged in
=============
BK Precision 1801 - 238 - 255 hz
Sampo FG1617 - 200 hz
Instek FG-8016G - 1.4 to 1.5 khz


Laptop Unplugged
============
BK Precision 1801 - 200 hz
Sampo FG1617 - 200 hz
Instek FG-8016G - 275 to 294 hz

Using a  PC to output a signal via the sound card
=================================
BK Precision 1801 - 203 hz
Sampo FG1617 - 200 hz
Instek FG-8016G - 4.4 to 4.6 khz

I am using a stereo to RCA cable and then a RCA to BNC adapter. Is this the correct setup? Why is it that when the laptop is plugged in to mains, the BK unit reads incorrectly? Why is the Instek unit reading incorrectly in all cases? If I output a signal directly from the Sampo FG1617 and connect it directly to the Instek counter input via a BNC to BNC cable, the Instek reports the value correctly.

edit: I since found that if I push the 1/10 button on the Instek it reads the 200 hz correctly regardless if the laptop is plugged into mains or not and also reads correctly from the PC. Why is it necessary to do this?

I found the BK 1801 to read around 1.4khz from the laptop when the laptop is plugged into mains when testing again. I also found that if I output a signal from my iPhone when it is not charging the BK will read ok, if I plug it into the charger it is way off (65 khz) . Why is this and why does having some devices plugged into mains such as the laptop or iPhone effect the output but not on others (such as from the PC) ?

[ModemHead]

Oh my, didn't mean to open up another can of worms with the PC-generated waveform suggestion.  You may have some common-mode noise on the mains in your environment that the frequency counters are responding to.  Dave has an interesting video on this subject: http://www.eevblog.com/2013/03/22/eevblog-441-how-to-track-down-common-mode-noise/

Just for grins, I checked an 8050A here and got -2.38/-2.48 mV on pins 6/7, for a difference of 0.10mV.

[rstor22]

When you calibrate your Fluke meters, are you using a function generator or DMM calibrator?

Actually its good that I went through this excercise as I didn't realize I could have an issue with common-mode noise. I watched Dave's video you linked which was helpful in my troubleshooting, thank you! I decided to use a scope probe instead of the previous cables mentioned above. I proceeded to move the Instek and BK test equipment and laptop to another area of the house with an outlet on a different circuit. Outputting 200hz from the laptop (when plugged into mains) showed around 300 hz on the Instek and on the BK it was around 204 hz. A significant decrease (from the values noted previously). I did notice that for the Instek, if I switched the plug with the laptop in the same power receptacle, it reads around 600-800hz. I don't know why...

I then brought everything back in the original room and used an isolation transformer and tried various combinations of floating only the measuring equipment, only the laptop, and both devices. I found that if I float both the test equipment and the laptop, then the results are somewhat similar to when I measured them in the other room (i.e. the BK 1801 read 230 hz, the Instek read 800 hz [when a 200hz signal is applied from the laptop])

I found in all cases (e.g. in the original room, with no isolation transformer, everything powered from the mains), where values are way off i.e. 2-5 khz when a 200hz signal is applied, if I probe using a scope probe with 10x, then the results *always* read correctly as 200hz on both the Instek and BK.

I am guessing this is similar to using the 1/10 button on the Instek. I am thinking the effect has something to do with the amplitude attenuation at 10x? Would you know why everything reads correctly when probing with 10x, but when probing on 1x the counter readings are way off?

[ModemHead]

DMM calibrators are a bit pricey for my budget, so I use a DDS function generator (Instek SFG-1013).  It's reasonably stable after it warms up an hour or so.  Good enough for DIY repair projects on 3-1/2 digit meters.

Your noise problems seem a bit extreme.  You should probably scope the signals to see what they look like.  When you attenuate the signal, the noise components may then fall under the threshold of the counter's sensitivity, leaving only the desired signal.  Also, your generator is unloaded by the high-impedance inputs of a counter and/or DMM, leaving the line subject to reflections and noise.  Try terminating the line with a 50-ohm load at the end.

[rstor22]

I looked at the 200hz signal on the scope as you suggested and located the source of the noise. Thank you! I found that there was an old computer plugged in that was not turned on but was causing noise in standby (see attached image). Though the scope reports 200hz in both cases, when the noisy computer is unplugged, the BK Precision 1801 counter reports the 200 hz signal correctly.

For the Instek unit the issue persists when trying to read a 200hz signal. The issue is worse using the 3.5mm to RCA setup (1100hz reading) . I found that if I emit a 200 hz signal from the iPhone (with it plugged into mains) and probe using my Rigol scope probes at 1x then the reading is better -- about 340 to 370 hz. If I attenuate either on the counter or via the probe the issue is resolved. If do not attenuate but instead use a 50 ohm terminator, the unit does read 200 hz correctly. However; when trying the same 200 hz signal emitted from a laptop plugged into mains, the best I can get is 280-300 hz with the 50 ohm terminator. With the laptop only on battery and with a 50 ohm terminator, the reading is around 220hz. Again, with the Instek, if I do not use the 50 ohm terminator, if I attenuate the signal, then regardless if the laptop is plugged in or not, it reads the 200hz signal correctly.

The Instek manual mentions it has a high input sensitivity of 20mv RMS. In another area it mentions the sensitivity as <= 20mV rms.  Do you suspect a component to be defective in the Instek? Or could it be that it is overly sensitive to any minor noise?

[ModemHead]

Frequency counters are out of my area of expertise, I don't even own one.  But yours does seem to be overly sensitive to noise.  Maybe someone else can chime in on that?

[retiredcaps]

Maybe someone else can chime in on that?

@rstor22, for those questions, it is probably best to start a new thread with a different subject header.

[rstor22]

Thank you. I've started a new thread here (also includes some more details that I tried and also part of the response from Instek):

https://www.eevblog.com/forum/beginners/frequency-counter-measurements-and-noise/

[Napalm2002]

My dc measurements are way off. As are my ohms measurements as well. I pulled it apart and didn't see any contamination on the board? Good way to clean the switches?  Replace the electrolytics anyway?