EEVblog Electronics Community Forum
Electronics => Repair => Topic started by: FlyingHacker on September 17, 2015, 10:26:55 pm
-
Hi,
I have a Data Precision 2480 DMM I am trying to fix. The schematic is here: http://meanmutha.com/DataPrecision2480Schem.pdf (http://meanmutha.com/DataPrecision2480Schem.pdf)
Note that any component labeled "FSV" in the schematic is not present on the board. I don't know what "FSV" means. Also, this unit has the RMS option for AC, if that matters.
Everything works fine EXCEPT the DCV 100V setting. Ohms works fine.
If I try to measure 5v DC it reads correctly when I have the range set to 1v or 10v or 1000v. However, when I set the range to 100v it reads ~6.89v (it varies sometimes). If I feed it 1.26v it reads 0v in the 100v range. The error offset is not linear that I can tell.
I checked the input voltages per the limited service info in the manual, and they read fine. (Manual is paper, and I do not have the whole thing scanned.) I checked the reference voltages, and they were fine too.
The logic seems to be working fine, as the readout is displaying the voltage as seem at E37 on the schematic (right after the attenuator at H12). The manual also provided some logic waveforms, and they checked out OK on the scope.
The input to the meter is on the left hand side of the schematic. Everything DCV related is nicely labeled DCV.
A4 is a precision resistor pack used in the attenuator. When in 100v range pin 4 reads the voltage displayed by the meter (the wrong voltage). When it any other range it reads the correct voltage. As the A4 resistors are also used in the Ohms measurement I do not believe they are at fault. Also, other ranges work fine, and some would use the same resistors.
I tried removing C2, C3, C4, C5, C8, C9, C10 (used for AC tuning for ACV). This had no effect on the problem.
The decimal places work correctly (lower right side of the schematic, controlled by the same switch.) The 100v range switch is S10 (you can see these ranges listed right next to the attenuator on the schematic).
I am beginning to suspect it is possibly an issue in the switch S10. I put tons of contact cleaner in there, and in all the other switches, but that did not help. These are those ganged push button switches, where pushing one in makes the others pop out. They are a massive pain to remove, and that would be my last resort after seeking help here.
Any help would be appreciated. Any suggestions for where to look would be helpful.
The since the difference between the correct voltage and the displayed voltage is not consistent this makes me think a transistor or opamp is involved. Would some kind of crud in a switch cause the offset to vary like I described above?
Thanks in advance.
-
Everything works fine EXCEPT the DCV 100V setting. Ohms works fine.
Just to confirm, ohms works in the same 100 gang switch setting?
-
Yes. Ohms works at every range.
-
How about ACV?
And just to be sure you can still measure A4. Pull all switches off (push little pit then switch should pop back) when doing this.
-
Some more info...
I was wrong in my initial statement. ACV is also incorrect at the 100v range. 6.899v reads as 11.8. All the ohms ranges work correctly, though.
Adjusting R12 (G9 on the schematic) as part of the calibration procedure for ACV seems to have no effect.
If I measure A4 with the 100v range button in then it is higher, like the voltage. If I lightly push the button so that all range buttons pop out then the voltage on the A4 pins goes back to where it should be.
-
Measure the input impedance of the meter with another multimeter. That will tell you if the input divider is fried.
-
Measure the input impedance of the meter with another multimeter. That will tell you if the input divider is fried.
Just to be sure I do this right.... You are saying hook another meter up to the inputs of this meter and measure impedance (ohms), right? What kind of values should I be seeing in that case?
Thanks.
-
^^^ Correct.
Should be 10Meg ohm.
Currently reading through the manual.
-
This is what I dug up, wish the schematics were online.
There is no gain needed for the voltage ranges, which is one kess thing to worry about.
All rain switching is done with the input resistor string. I guess it is manual ranged.
-
I posted the schematic in the original post.
http://meanmutha.com/DataPrecision2480Schem.pdf (http://meanmutha.com/DataPrecision2480Schem.pdf)
-
^^^ Correct.
Should be 10Meg ohm.
Currently reading through the manual.
Measurements with power on the DUT off:
Range: reading
100mV: 160kOhm
1V: 160kOhm
10V: 5.7 MegaOhm
100V: 5.6 MegaOhm
1000V: 8.6 MegaOhm
With power to the DUT on:
100mV: 4.7 MegaOhm
1V: 4.6 MegaOhm
10V: 5.4 MegaOhm
100V: 5.4 MegaOhm
1000V: 6 MegaOhm
-
Some more info...
I was wrong in my initial statement. ACV is also incorrect at the 100v range. 6.899v reads as 11.8. All the ohms ranges work correctly, though.
Adjusting R12 (G9 on the schematic) as part of the calibration procedure for ACV seems to have no effect.
If I measure A4 with the 100v range button in then it is higher, like the voltage. If I lightly push the button so that all range buttons pop out then the voltage on the A4 pins goes back to where it should be.
Measure A4 resistance between each pin pair. 6-5 5-4 4-3 etc. Of course do this power off. And all range button out.
-
In what way do the ranges (1v, 10v,...) effect the resistance measurements?
-
In what way do the ranges (1v, 10v,...) effect the resistance measurements?
Measuring resistance using the meter itself works fine at all ranges. The correct value is shown even at the 100V (100kOhm) range.
This is why I was not suspecting the A4 resistor set.
Thanks.
-
It does look like the ohms measurements go through the A4 resistor pack as well, right? Since ohms works fine at every range we can rule out problems with A4, no?
-
If the 1000 V range is really OK (less than 2% off) and the ohms ranges are working Ok, the divider itself should be OK. The problem should be in on of the switches, especially S10.
It's also strange to get a 5 M resistance reading - this should be closer to 10 M.
-
I believe kleinstein is correct, I think one(or more) of the switches has gone high resistance(but not open) as after the switches the signal goes through a few 51K resistors and through Q7Fet.
-
I was worried it might be the switch... So I guess I desolder the one in question and try to clean it up.
-
Why don't you just measure switch first :-//
-
Why don't you just measure switch first :-//
You think I can measure it in circuit? You are saying measure the resistance across the different poles of the switch, right? And compare it to one of the other switches... That might work.
-
You can at least measure the switch of interest here. There are high value resistors connected, so they won't make a difference when it comes to measure <1 Ohms a closed contact should have. No need to compare with other switches.
-
....
Note that any component labeled "FSV" in the schematic is not present on the board. I don't know what "FSV" means. Also, this unit has the RMS option for AC, if that matters.
....
I believe that "FSV" means "factory selected value" for components added before shipment as part of calibration.
-
You can at least measure the switch of interest here. There are high value resistors connected, so they won't make a difference when it comes to measure <1 Ohms a closed contact should have. No need to compare with other switches.
Oddly the switches seem to measure fine. Either open or about 0.2ohms in the off or on positions.
The only things that read as an in between value were between S10 pins A2 and A3 with all switches out, which read just over 1k, and S11 pins A2 to A3, which read just over 1k as well. Those pins are used in current measurement only I think.
I fed a voltage into the input with the meter under test turned off and the resulting voltage was still incorrect when I switched in the 100V range. This makes me think it is indeed something passive... But where?!?
-
Try this, if you have a esr meter measure C38,39 if not swap C36,37 15uf 10V. I'll bet they have developed high esr. The caps are connected to 100V selector switch.
-
Try this, if you have a esr meter measure C38,39 if not swap C36,37 15uf 10V. I'll bet they have developed high esr. The caps are connected to 100V selector switch.
I will check them out. Though I believe those switches are basically four dual throw (changeover) switches in one. I guess quad pole changeover... And the pins B1, B2, and B3 are all one of the four subswitches used for switching on the decimal place character on the display. When the switch is out B1 is connected to B2. When in B3 is connected to B2.
So I think those caps are isolated from the measurement circuit. But will check them out anyway. This thing is driving me nuts. Thanks.
-
Try this, if you have a esr meter measure C38,39 if not swap C36,37 15uf 10V. I'll bet they have developed high esr. The caps are connected to 100V selector switch.
I will check them out. Though I believe those switches are basically four dual throw (changeover) switches in one. I guess quad pole changeover... And the pins B1, B2, and B3 are all one of the four subswitches used for switching on the decimal place character on the display. When the switch is out B1 is connected to B2. When in B3 is connected to B2.
So I think those caps are isolated from the measurement circuit. But will check them out anyway. This thing is driving me nuts. Thanks.
The caps and switches are connected to voltage comaparator opamp Z5/ (.1REF) voltage reference and linked to R49 27K to a quantity symbol ( diameter1) back to AD convertor 48-4058? If it is not this then
check quantity ( diameter1) pin 13 of AD convertor signal chain.
-
Try to measure voltage in different points. First put 5V in jacks and but DCV and 100V range. Then measure points
S6/A3 - S4/B1 = 5V
S2/B3 - A4/1 = 5V
A4/6 - A4/1 = 5V
A4/5 - A4/1 = 0,5V
A4/3 - A4/1 = 0,05V (From your first post with fault it read about 0,0689V)
If there is something funky business (there should be) try different range and see those results.
You should also try wave the switches and see if that affects reading. That way you might see if switch is faulty or something.
I was also thinking that fault might be in switch S11 between A6-A5. If there is a leakage it will affect just 100V range but it should drop voltage not raise it.
-
The caps and switches are connected to voltage comaparator opamp Z5/ (.1REF) voltage reference and linked to R49 27K to a quantity symbol ( diameter1) back to AD convertor 48-4058? If it is not this then
check quantity ( diameter1) pin 13 of AD convertor signal chain.
Wow. I missed that! It is amazing what your brain does when you think you understand something. It prunes it away for efficiency. But when you really don't it bites you in the rear...
I will definitely check those out. Thanks.
-
Try to measure voltage in different points. First put 5V in jacks and but DCV and 100V range. Then measure points
S6/A3 - S4/B1 = 5V
S2/B3 - A4/1 = 5V
A4/6 - A4/1 = 5V
A4/5 - A4/1 = 0,5V
A4/3 - A4/1 = 0,05V (From your first post with fault it read about 0,0689V)
If there is something funky business (there should be) try different range and see those results.
You should also try wave the switches and see if that affects reading. That way you might see if switch is faulty or something.
I was also thinking that fault might be in switch S11 between A6-A5. If there is a leakage it will affect just 100V range but it should drop voltage not raise it.
I have been doing that without figuring out why I am getting the readings I am getting.
What do you mean by "wave the switches" ?? You mean swap two of them? Or shift them all down or something? I have to figure out to remove the mechanical interlock (not that hard), and then I can remove them with a desoldering station.
Thanks.
-
I mean that when you get wrong reading you touch switches and see if that affect that reading. Like shake or wiggle. I do not know how to say that more clearly. (Sorry my english of course)
Every measurement you make you can take those in the paper and post those in here. That way we get more info and it is easier find the fault. It is very difficult say were the fault is if we not know those measurements.
-
Try to measure voltage in different points. First put 5V in jacks and but DCV and 100V range. Then measure points
Agreed. If you can poke around inside with the unit on then the best thing to do is to follow the voltage you're putting in. I mean, put 5v in then measure the voltage at the first point of contact to the PCB, then systematically trace the voltage through until you find the voltage drop...