Author Topic: DMM leakage cleaning  (Read 1822 times)

0 Members and 1 Guest are viewing this topic.

Offline Harvs

  • Super Contributor
  • ***
  • Posts: 1143
  • Country: au
DMM leakage cleaning
« on: December 01, 2013, 11:59:33 am »
Having got a HP 3456A back up and running, I've been checking various things the best I can.

I rigged up a test to get a rough idea of the input impedance.  Using the input resistance of a handheld DMM to measure current (as I've discussed previously on my blog.)

It took a fair bit of mucking around running leads in the air to get a zero input current reading without the meter connected, it's surprising how much leakage is cause just resting a lead on top of the meter's chassis.

Anyway, I ended up with ~2.5GOhm as the input impedance with a 10V input, and it's pretty linear as voltage is decreased.  Spec is for >10GOhm.

While I tried to be careful about what I touched when I had the covers off, the reality is it took quite a while to find the problem so I have little doubt that I've contaminated the input amplifier along the way.

So my question, how difficult is it to get the board clean enough to restore the spec?  Or is likely that me going around with IPA and cotton tip is likely to cause more problems than solve?
« Last Edit: December 01, 2013, 12:01:14 pm by Harvs »
 

Offline Lightages

  • Supporter
  • ****
  • Posts: 4295
  • Country: ca
  • Canadian po
Re: DMM leakage cleaning
« Reply #1 on: December 01, 2013, 01:00:37 pm »
It doesn't sound to me that you can do a reliable measurement of the input impedance that way. You said you wanted a rough idea and I think that is what you got and that is all you can get with that setup. You are well within an order of magnitude so I would say you are there.

If you really want to do this a little bit better than I would suggest stringing a bunch of really high resistors together to get 10gigOHM and then connect this to the input of the meter and supply a voltage. If the measured voltage is around, or less than half, of the input voltage than you are probaby close to the 10 gigOHM you are looking for.
 

Offline Harvs

  • Super Contributor
  • ***
  • Posts: 1143
  • Country: au
Re: DMM leakage cleaning
« Reply #2 on: December 01, 2013, 01:32:47 pm »
It doesn't sound to me that you can do a reliable measurement of the input impedance that way. You said you wanted a rough idea and I think that is what you got and that is all you can get with that setup. You are well within an order of magnitude so I would say you are there.

Why not?

With 10V across it, I measured 4.7nA.  That's 47mV across the handheld DMM.  For it to be within spec that'd be down to < 10mV.

In my experience of putting this sort of test together, it's actually far more reliable doing it this way than stringing a bunch of resistors together, because you can make the leads very short so you don't end up with large antenna's picking up noise.  You also don't have to worry about keeping all the resistors in the air.

If you've got a solid reason why using the 10Meg input impedance of a decent handheld DMM is not suitable for measuring low current please let me know.
 

Offline sync

  • Frequent Contributor
  • **
  • Posts: 799
  • Country: de
Re: DMM leakage cleaning
« Reply #3 on: December 01, 2013, 01:48:14 pm »
Too measure the input impedance you only need a stable voltage source and a 1Mohm resistor. Connect the resistor directly to the HI input terminal of the DMM. Connect the voltage source to the resistor and LO terminal. Short the resistor with an alligator test lead or something else. Use high resolution mode on the DMM. Set the voltage source to 10V. Wait until the reading has stabilized. Read the value or null the meter. Then remove the short over the resistor. Read the value. The difference is the voltage drop over the 1Mohm resistor. The rest is ohms law. Test -10V too.
 

Offline Harvs

  • Super Contributor
  • ***
  • Posts: 1143
  • Country: au
Re: DMM leakage cleaning
« Reply #4 on: December 01, 2013, 02:04:46 pm »
Ok I just re-did the measurements.  This time with 10x 10M resistors in series, and the handheld DMM trick with a battery powered 10V reference.

I must of had a lead touching something, it's such an easy measurement to screw up.  This time with everything suspended in the air, I got 12-13Gohm on the +10V and about 11Gohm on the -10V.

Both the DMM trick and the 100Mohm resistor pretty much agree to within a few 10's of Meg.

So I'm happy...
 

Offline Lightages

  • Supporter
  • ****
  • Posts: 4295
  • Country: ca
  • Canadian po
Re: DMM leakage cleaning
« Reply #5 on: December 01, 2013, 02:06:54 pm »
Great! It is nice to have confirmation with the two methods.  ;)
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf