Author Topic: HP 34401a - Drift in the 2W R mode / 100 Ohm range  (Read 16311 times)

0 Members and 1 Guest are viewing this topic.

Online ZuccaTopic starter

  • Supporter
  • ****
  • Posts: 4298
  • Country: it
  • EE meid in Itali
HP 34401a - Drift in the 2W R mode / 100 Ohm range
« on: July 21, 2014, 12:43:45 am »
Hello guys.

I already (partially) fixed previously this unit, see the full story here:

https://www.eevblog.com/forum/testgear/34401a-cal-error-707/

but the cal lab is still reporting a drifting in the 2W R mode in the 100 Ohm range. Only in the 2W R mode not in the 4W :wtf:!

Since the unit was sent to me from ebay in a worst than terrible package (stupid seller and all those who thinks a package is not important), I think some joints on the board could have some issue (as already found in my previous repair..). Since the failure reported is only in the 2W I have a big hope to fix this problem.

Now I will start the investigation and report my findings soon.

Stay tuned,

Z
« Last Edit: July 21, 2014, 01:22:03 am by zucca »
Can't know what you don't love. St. Augustine
Can't love what you don't know. Zucca
 

Online ZuccaTopic starter

  • Supporter
  • ****
  • Posts: 4298
  • Country: it
  • EE meid in Itali
Re: HP 34401a - Drift in the 2W R mode / 100 Ohm range
« Reply #1 on: July 21, 2014, 03:35:10 am »
Since I have a good calibrated 34401a, I am going to compare the two side by side.

First of all a sanity check with a 2W R and a 0 Ohm resistor (this one http://www.pomonaelectronics.com/pdf/d_MDP-S_100.pdf).

(ignore the unit on the top)


The good is connected with RS232, the bad one with GPIB (no ther devices on the bus).

Settings:
Auto Range:   On
Auto Zero:   On
Measurement:    Resistance (2W)
NPLC:   10
Range(Ohm):   100
minumum sample interval, 10min test

Here the graphs, left one the good one calibrated and right the bad one (not calibrated)



Statistics of the good one (front side)

Mean   0.002524814
Standard Error   4.86997E-06
Median    0.00252
Mode    0.002609
Standard Deviation    0.000196435
Sample Variance    3.85869E-08
Kurtosis    -0.288419334
Skewness   0.063261047
Range    0.001202
Minimum   0.001919
Maximum    0.003121
Sum   4.107873
Count    1627
Largest(1)    0.003121
Smallest(1)    0.001919
Confidence Level(95.0%)   9.55207E-06


and bad one (front side):

Mean    0.010605372
Standard Error   6.77113E-06
Median   0.010596
Mode    0.010571
Standard Deviation   0.000269573
Sample Variance   7.26694E-08
Kurtosis   -0.036873015
Skewness   0.054733768
Range   0.001944
Minimum   0.009568
Maximum   0.011512
Sum   16.809514
Count   1585
Largest(1)   0.011512
Smallest(1)   0.009568
Confidence Level(95.0%)   1.32813E-05


Oh, the GPIB (158,5 sample/min) is slower and the RS232 (162,7 sample/min)  :-DD, let´s know investigate the rear side with the same sample numbers 1650 instead the same time 10min.

Here the new settings:

Settings:
Auto Range:   On
Auto Zero:   On
Measurement:    Resistance (2W)
NPLC:   10
Range(Ohm):   100
minumum sample interval, 1.65K Samples

same story,  left one the good one calibrated and right the bad one (not calibrated):



Statistics of the good one (rear side)

Mean   0.00705325
Standard Error   4.10731E-06
Median   0.007048
Mode   0.007124
Standard Deviation   0.00016684
Sample Variance   2.78355E-08
Kurtosis   0.594448789
Skewness   0.135870213
Range   0.001535
Minimum   0.006293
Maximum   0.007828
Sum   11.637863
Count   1650
Largest(1)   0.007828
Smallest(1)   0.006293
Confidence Level(95.0%)   8.05609E-06



and bad one (rear side):


Mean   0.016549085
Standard Error   6.83344E-06
Median   0.016528
Mode   0.01654
Standard Deviation   0.000277576
Sample Variance   7.70483E-08
Kurtosis   0.484874451
Skewness   0.2889019
Range   0.002157
Minimum   0.015462
Maximum   0.017619
Sum   27.305991
Count   1650
Largest(1)   0.017619
Smallest(1)   0.015462
Confidence Level(95.0%)   1.34031E-05


Conclusions:
At 0 Ohm 2W the bad one is a little bit off of about 8m Ohm (not all the POMONA MDP-S are the same???). No significant drift present at 0 Ohm, although the Standard Deviationi is slightly bigger in the bad one.

Soon the 100 R Range Test with a 90 R connected.

Take care,

« Last Edit: July 21, 2014, 03:37:24 am by zucca »
Can't know what you don't love. St. Augustine
Can't love what you don't know. Zucca
 

Online ZuccaTopic starter

  • Supporter
  • ****
  • Posts: 4298
  • Country: it
  • EE meid in Itali
Re: HP 34401A - Drift in the 2W R mode / 100 Ohm range
« Reply #2 on: July 25, 2014, 04:49:32 am »
Ok, it was a pretty nasty measure but I finally got it.

Since the target is to test a drift on the R measured, I needed a way to check not only his value over time but also the current generated by the 34401A.

Here my setup:



I have the 34401A measuring a precision 90R in parallel with a 34461A to check the voltage across it (related to the current generated by the 34401A DUT).
Basically the voltage measured by the 34461A divided by 90R will provide us the current sourced by the 34401A. The coax cables will minimize the noise pickup.

Now I had the 34461A USB connected to the PC, and and the other 34401A connected as well using GPIB or RS232.

I had two 34401A units (GOOD #1 and GOOD #2) fresh from calibration no problem to use as reference to compare with the, according to the calibration lab, bad one (BAD).
Of course the units were properly warmed up and the measure time was over one hour period.

Here the setting used for the R measures:

Auto Range:   On
Auto Zero:   On
Measurement:    Resistance (2W)
NPLC:   10
Range(Ohm):   100


and for the current ones on the 34461A:

Auto Null Value:   FALSE
Auto Range:   On
Auto Zero:   On
Input Impedance:   10 MOhm
Measurement:   DC Voltage
NPLC:   10
Null State:   FALSE
Null Value(Vdc):   0
Range(Vdc):   0.1


Let´s dig into the results...

GOOD #1

2W R measured:



Statistic [Ohm]

Mean   90.07426003
Standard Error   3.79745E-06
Median   90.074236
Mode   90.074083
Standard Deviation   0.000375295
Sample Variance   1.40846E-07
Kurtosis   -0.493118701
Skewness   0.283102244
Range   0.002418
Minimum   90.073072
Maximum   90.07549
Sum   879755.2978
Count   9767
Largest(1)   90.07549
Smallest(1)   90.073072
Confidence Level(95.0%)   7.44379E-06


Current (90mV=1mA)



Statistic [A*90]

Mean   0.09014677
Standard Error   2.22337E-09
Median   0.090146746
Mode   0.090146714
Standard Deviation   2.22526E-07
Sample Variance   4.95176E-14
Kurtosis   -0.075682427
Skewness   0.473640809
Range   1.4409E-06
Minimum   0.090146106
Maximum   0.090147547
Sum   903.0001985
Count   10017
Largest(1)   0.090147547
Smallest(1)   0.090146106
Confidence Level(95.0%)   4.35824E-09


GOOD #2

2W R measured:



Statistic [Ohm]

Mean   90.07989073
Standard Error   3.09024E-06
Median   90.079946
Mode   90.079946
Standard Deviation   0.000302496
Sample Variance   9.15041E-08
Kurtosis   -0.177549751
Skewness   -0.477571608
Range   0.0019
Minimum   90.078848
Maximum   90.080748
Sum   863145.513
Count   9582
Largest(1)   90.080748
Smallest(1)   90.078848
Confidence Level(95.0%)   6.05752E-06


Current (90mV=1mA)



Statistic [A*90]

Mean   0.088696667
Standard Error   2.62486E-09
Median   0.088696721
Mode   0.088696711
Standard Deviation   2.62263E-07
Sample Variance   6.87817E-14
Kurtosis   -0.557971289
Skewness   -0.449007757
Range   1.5119E-06
Minimum   0.088695812
Maximum   0.088697324
Sum   885.4588291
Count   9983
Largest(1)   0.088697324
Smallest(1)   0.088695812
Confidence Level(95.0%)   5.14525E-09


BAD

2W R measured:



Statistic [Ohm]

Mean   90.02526044
Standard Error   5.78764E-06
Median   90.025232
Mode   90.025119
Standard Deviation   0.000564228
Sample Variance   3.18353E-07
Kurtosis   -0.456379127
Skewness   0.228340502
Range   0.00326
Minimum   90.02379
Maximum   90.02705
Sum   855600.0752
Count   9504
Largest(1)   90.02705
Smallest(1)   90.02379
Confidence Level(95.0%)   1.1345E-05


Current (90mV=1mA)



Statistic [A*90]

Mean   0.089503874
Standard Error   2.02854E-09
Median   0.089503873
Mode   0.089503842
Standard Deviation   2.02641E-07
Sample Variance   4.10633E-14
Kurtosis   -0.520592125
Skewness   -0.068937736
Range   1.3154E-06
Minimum   0.08950316
Maximum   0.089504475
Sum   893.159158
Count   9979
Largest(1)   0.089504475
Smallest(1)   0.08950316
Confidence Level(95.0%)   3.97634E-09


CONCLUSIONS

We need a side to side comparison.

2W R Measurements



Here the yellow values are the only one concerning me. Still not a huge drift in my eyes. The BAD shows a negative drift decreasing the R value. The GOOD 1 also does that, but the gradient is less than the BAD one.

Current Measurements

Pretty much here everything is acceptable, since the unit cal lab reported the drift only in the 2W R mode, I was not expecting a problem with the current:



Maybe the current generated by the BAD one is suspiciously pretty nervous and irregular in the graph.
 
Next steps

0) Check/Understand the Spec. Tolerance (let´s do tomorrow. Is already 00:51 AM)
1) Dig into the schematics and isolate the part of the circuit which can cause a drift only in 2W R mode
2) Perform a 4W R 1 hour test as above to verify the drift is not there in 4W
3) Check the 100 Ohm Range current generation, check if current is constant in 1K Range
3) Open the unit and look the circuit for bad joints on the part of circuits founded in point 1 and 3

I don´t know if I will fix this unit, it seems a nasty slightly out of spec. malfunction.

On the side: EMI CRAP?

I had the bad idea to work under the bench using my 12V impact drill during the 1 hour measurement and here what I got on a good one! (left current, right R measured)



Even when I turned on my monitors from sleep, I noticed similar jumps. I decide to measure when I was not at the bench during night time (everything off, just the PC and the DMMs).
 

« Last Edit: July 25, 2014, 04:57:05 am by zucca »
Can't know what you don't love. St. Augustine
Can't love what you don't know. Zucca
 

Online ZuccaTopic starter

  • Supporter
  • ****
  • Posts: 4298
  • Country: it
  • EE meid in Itali
Re: HP 34401a - Drift in the 2W R mode / 100 Ohm range
« Reply #3 on: July 26, 2014, 04:02:28 am »
I spent 3 hours on the service manual.

The Ohm current source block is working since the 4WR is not failing.
The Function Switch is working because the 4WR is not failing.
The U101-5 sense (voltage measured on the High Post), DC amplifier and DAC is working because is not failing the Voltage Tests.
Since the 1KOhm 2W Range is not failing also the 2W Voltage drop reading of a 1mA current source is also working.

Here an interesting further measure 1 hour 100NPLC 2WR, 100 Ohm Range with 100Ohm R. I also did a null for the the cable resistance...



I don´t like that exp transient...the device was already warmed up pretty good, it should not happen.

mmmmm interesting.



 
« Last Edit: July 26, 2014, 06:18:12 am by zucca »
Can't know what you don't love. St. Augustine
Can't love what you don't know. Zucca
 

Online ZuccaTopic starter

  • Supporter
  • ****
  • Posts: 4298
  • Country: it
  • EE meid in Itali
Re: HP 34401a - Drift in the 2W R mode / 100 Ohm range
« Reply #4 on: July 26, 2014, 06:08:34 am »
Another measure 1 hour 100NPLC 4WR, 100 Ohm Range with 100 Ohm R attached. No drift or transient at all.

Can't know what you don't love. St. Augustine
Can't love what you don't know. Zucca
 

Online ZuccaTopic starter

  • Supporter
  • ****
  • Posts: 4298
  • Country: it
  • EE meid in Itali
Re: HP 34401a - Drift in the 2W R mode / 100 Ohm range
« Reply #5 on: July 26, 2014, 01:17:03 pm »
Damm it, 1 hour 100NPLC 2WR, 1 KOhm Range with 1 KOhm R attached. No drift here  |O



PS: The red box was probably because the R was cold and it took a little bit to settle with the temperature.
Can't know what you don't love. St. Augustine
Can't love what you don't know. Zucca
 

Offline robrenz

  • Super Contributor
  • ***
  • Posts: 3035
  • Country: us
  • Real Machinist, Wannabe EE
Re: HP 34401a - Drift in the 2W R mode / 100 Ohm range
« Reply #6 on: July 26, 2014, 01:51:57 pm »
Here an interesting further measure 1 hour 100NPLC 2WR, 100 Ohm Range with 100Ohm R. I also did a null for the the cable resistance...



I don´t like that exp transient...the device was already warmed up pretty good, it should not happen.

mmmmm interesting.

Just some thoughts:

Are you thermally shielding the resistor and leads during this test?  The TCR of the copper test leads come into play in 2 wire mode. Also since there is no offset compensated ohms mode you are subject to errors caused by thermal EMFs. You have about a .004 Ohm span there which is only 4µV at the 1mA drive current of the 100 Ohm range so thermal EMFs will be significant.


Offline Dr. Frank

  • Super Contributor
  • ***
  • Posts: 2377
  • Country: de
Re: HP 34401a - Drift in the 2W R mode / 100 Ohm range
« Reply #7 on: July 26, 2014, 02:04:36 pm »
I spent 3 hours on the service manual.

....

The Function Switch is working because the 4WR is not failing.
....

Here an interesting further measure 1 hour 100NPLC 2WR, 100 Ohm Range with 100Ohm R. I also did a null for the the cable resistance...


mmmmm interesting.

I think, your logical conclusion is wrong!

Function switch?

Do you mean the Front/Rear switch, or the internal switching relays?

Both are in series with the 2W resistor DUT, but they are both NOT compensated , but they  are indeed in 4W Ohm!!

Therefore, if one of the 2-3 components in this inner 2W circuitry (I listed them already, in the original thread) is faulty, and this obvious, then it will affect your 2W measurement, but NOT your 4W one.

I still vote for the Front/Rear switch, which gets polluted by something and then has several ten milli Ohm change on its contacts if you actuate it.. Afterwards, it may also creep back to lower Ohm values, if you leave it alone, afterwards.

Just redo the tracing function in 2W, 100Ohm, then switch front/rear and to front again, and measure the difference, also over time.

I bet that will be not oK.

A new or cleaned switch will have a few milli Ohm only, and this will not change that much after actuation.

Frank
 

Offline free_electron

  • Super Contributor
  • ***
  • Posts: 8515
  • Country: us
    • SiliconValleyGarage
Re: HP 34401a - Drift in the 2W R mode / 100 Ohm range
« Reply #8 on: July 26, 2014, 02:43:48 pm »
Notice that the machine reads 0.05 ohms lower than the others..
That tells something.
Professional Electron Wrangler.
Any comments, or points of view expressed, are my own and not endorsed , induced or compensated by my employer(s).
 

Offline Dr. Frank

  • Super Contributor
  • ***
  • Posts: 2377
  • Country: de
Re: HP 34401a - Drift in the 2W R mode / 100 Ohm range
« Reply #9 on: July 27, 2014, 09:45:49 am »
Hi,
here are my drift measurements.

I have a massive short on the instrument, photo 1.
fixed range 100Ohm, NPLC 10 (6digit, fast).

In 2W and 4W mode, I first switch to rear and back to front, then I immediately start the measurement with benchview.

You see, that in 2W mode, it takes sveral minutes that the zero reading is going down from 13mOhm to 6mOhm .
At time of writing, the reading is down to 4mOhm, and it was initially 2mOhm, before I did the first actuation today.

In 4W mode, the reading is directly below 100µOhm, and the negative reading lets you estimate the real internal offset voltages.

You can repeat that as often as you like, but the switch, although cleaned, but not brought back to virgin condition, has some unsteady contact resistance.

I think, that's due to the contact force getting weaker.

I'd like to encourage you, to replicate my measurement.

Frank
 
« Last Edit: July 27, 2014, 09:56:09 am by Dr. Frank »
 

Online ZuccaTopic starter

  • Supporter
  • ****
  • Posts: 4298
  • Country: it
  • EE meid in Itali
Re: HP 34401a - Drift in the 2W R mode / 100 Ohm range
« Reply #10 on: July 27, 2014, 12:30:03 pm »
Free_electron, robrenz and Dr. Frank are officially invited at my home in South Carolina.
I will cook for you some Italian food and when afterwords some delicious wine we will finally see the reality in a proper way, we will move in my lab and discuss how to fix the 34401a.

Thank you soo much!

I am currently not in the lab, I will be back tonight.

I agree with all of you, I need to do more investigation. BTW while I was driving those days thinking about your posts, I think I got a nice idea. For the details I will post soon something!

Take care!
« Last Edit: July 28, 2014, 04:34:56 am by zucca »
Can't know what you don't love. St. Augustine
Can't love what you don't know. Zucca
 

Online ZuccaTopic starter

  • Supporter
  • ****
  • Posts: 4298
  • Country: it
  • EE meid in Itali
Re: HP 34401a - Drift in the 2W R mode / 100 Ohm range
« Reply #11 on: July 28, 2014, 04:17:40 am »
 Well... back in the lab, but it is 23:30 and it is time to sleep. Before that I want to respond to the you awesome guys who posted something

...
Are you thermally shielding the resistor and leads during this test?  The TCR of the copper test leads come into play in 2 wire mode.
...

You are correct, there is surely that error. Still I don´t know if it follows a exp(-t/T) decay over time...well may be because the temperature exchange between contact and the ambient will follow a exp(-t/T) to reach the equilibrium. If it is the case by inverting the high post with the low one on the DMM I should have a exp(-t/T) decay again but this time with R increasing. The EMF Thermal is always an offset in a defined direction, I mean it can´t be depended on the current direction. Tomorrow I´ll try.

Notice that the machine reads 0.05 ohms lower than the others..
That tells something.

I noticed that too immediately. Bad previous calibration? If the 2W R gain calibration value an old one? Still it must be investigated.


I think, your logical conclusion is wrong!

Function switch?
Do you mean the Front/Rear switch, or the internal switching relays?


Hi Dr. Frank I missed you a lot the past days. I was talking about the internal relays not the front/rear switch, pag. 95 of the service manual, and you your are right I was not 100% right. I think only K101 is not compensated in the 2W, so it is  definitely a candidate to be a bad guy. The others one don´t play a role, since they are not conducting the 1mA current and/or they don´t cause a voltage drop in the 2W measurement.

You see, that in 2W mode, it takes sveral minutes that the zero reading is going down from 13mOhm to 6mOhm .
At time of writing, the reading is down to 4mOhm, and it was initially 2mOhm, before I did the first actuation today.
...

I'd like to encourage you, to replicate my measurement.
 

Wow, also you are getting a exp(-t/T) decay! That warms my heart a little bit. Tomorrow I hope to have time to replicate your awesome tests and to go deeper in my theory/idea I got (i promise to explain it soon, not tonight.... have mercy on me). Anyway at the end of my theory/idea I agree with you: the front/rear switch needs to be cleaned and serviced.

According on a previous post:

https://www.eevblog.com/forum/testgear/hpagilentkeysight-34401a-2w-problem/15/

I got following suggestion regarding products to use there:

1) IPA bath, (obviously)

2)  Deoxit G100L

http://www.amazon.com/Deoxit-Gold-Solution-With-Brush/dp/B0032URX50/ref=sr_1_4?ie=UTF8&qid=1406520287&sr=8-4&keywords=Deoxit+G100L

 3) Meguiars NXT All Metal Polysh

http://www.premiumboatcare.com/mg13005.html?productid=mg13005&channelid=FROOG

I have some Stabilant available,

http://www.amazon.com/Stabilant-5ml-Kit-Makes-30ml/dp/B001E50GQS/ref=sr_1_1?ie=UTF8&qid=1406520641&sr=8-1&keywords=stabilant+22

but I am concern to mess up the switch property in such a precision measurement instrument.

What I should buy? I mean, I have already the IPA and the Stabilant, do I need anything else?

Good night and thanks again...

Z
« Last Edit: July 28, 2014, 04:36:50 am by zucca »
Can't know what you don't love. St. Augustine
Can't love what you don't know. Zucca
 

Offline Dr. Frank

  • Super Contributor
  • ***
  • Posts: 2377
  • Country: de
Re: HP 34401a - Drift in the 2W R mode / 100 Ohm range
« Reply #12 on: July 28, 2014, 06:36:08 am »
.. I was talking about the internal relays not the front/rear switch...

.. I agree with you: the front/rear switch needs to be cleaned and serviced...

.. but I am concern to mess up the switch property in such a precision measurement instrument.


Z

Hello,

before you clean the switch, you should test it, as described.

As I said, there are about 3 components in the 2W loop, which may fail, the other one is the relays, and the third, I don't know any more. Also, their solder junctions may be suspect.

But you can also test that by shorting temporarily these switches, and see what happens.

Concerning the switch, I'd also be careful, myself I cleaned it with conservative stuff like alcohol and mechanically.


Frank
 

Offline robrenz

  • Super Contributor
  • ***
  • Posts: 3035
  • Country: us
  • Real Machinist, Wannabe EE
Re: HP 34401a - Drift in the 2W R mode / 100 Ohm range
« Reply #13 on: July 28, 2014, 12:05:58 pm »
  The EMF is always an offset in a defined direction, I mean it can´t be depended on the current direction. Tomorrow I´ll try.

Thermal EMF polarity is current direction dependent. That is why the current reversal technique works.

Edit: That is not why the current reversal technique works :-[  My bad |O
From Keithley low level measurement handbook



« Last Edit: July 28, 2014, 02:03:42 pm by robrenz »
 

Online ZuccaTopic starter

  • Supporter
  • ****
  • Posts: 4298
  • Country: it
  • EE meid in Itali
Re: HP 34401a - Drift in the 2W R mode / 100 Ohm range
« Reply #14 on: July 28, 2014, 01:01:39 pm »
« Last Edit: July 28, 2014, 02:20:46 pm by zucca »
Can't know what you don't love. St. Augustine
Can't love what you don't know. Zucca
 

Offline robrenz

  • Super Contributor
  • ***
  • Posts: 3035
  • Country: us
  • Real Machinist, Wannabe EE
Re: HP 34401a - Drift in the 2W R mode / 100 Ohm range
« Reply #15 on: July 28, 2014, 02:04:44 pm »
Fixed my erroneous post above :palm:

Online ZuccaTopic starter

  • Supporter
  • ****
  • Posts: 4298
  • Country: it
  • EE meid in Itali
Re: HP 34401a - Drift in the 2W R mode / 100 Ohm range
« Reply #16 on: July 29, 2014, 03:19:14 am »
Dr. Frank,

this one is just for you, I want to replicate your tests and compare the results.

I was searching in my lab for something with a really low resistance... I found this:



but I was difficult to solder the banana jack on it:



So I used 8 Pomona MDP S:



Now here the result with 2W, exactly how you did:



and 4W:



mine 4W never goes negative in 4:25 min.

Looking also to my previous results, this unit is out of cal (slightly or not I don´t care) so I would not trust the reading too much. Maybe the cal lab played too much on it.
 
« Last Edit: July 29, 2014, 03:26:27 am by zucca »
Can't know what you don't love. St. Augustine
Can't love what you don't know. Zucca
 

Online ZuccaTopic starter

  • Supporter
  • ****
  • Posts: 4298
  • Country: it
  • EE meid in Itali
Re: HP 34401a - Drift in the 2W R mode / 100 Ohm range
« Reply #17 on: July 29, 2014, 04:11:42 am »
Let´s make it easy. The lab cal is telling us only the 2W 100 Ohm is failing, the rest is passing. If the 4W 100 Ohm test and all the others are ok, then the problem must be



a R High between the V Sense and the Hi post and/or a R Low between the Lo post and ground, since R Low and R High are bypassed in the 4W measurement (as Dr. Frank also says).

The idea is that (R High+R Low) is big enough to cause the 100Ohm test failure but small enough to pass the 2W 1KOhm.

Back to the schematics:



the first bad guy on the R High side is K101.

Moreover (I draw the current just for the front posts)



here we have another candidate for R High: S1-A. Finally the only bad guy for R Low is S1-G.

Of course it could also be some tracks or whatever. Now it´s the time to open the unit and dig into it.

The vibration stress theory

This unit had survived a journey in a terrible package, the resistors in the red circle had cracked joints... Now by looking at the board shape, all the area around those resistors could have been seen the same deadly vibrations since (at least in mine) the only mechanical holding point is the screw on the bottom right and the big red plastic posts enclosure block have some freedom to move.


 
Very interesting, all the suspected components are in that area.




 
« Last Edit: July 29, 2014, 04:47:01 am by zucca »
Can't know what you don't love. St. Augustine
Can't love what you don't know. Zucca
 

Offline robrenz

  • Super Contributor
  • ***
  • Posts: 3035
  • Country: us
  • Real Machinist, Wannabe EE
Re: HP 34401a - Drift in the 2W R mode / 100 Ohm range
« Reply #18 on: July 29, 2014, 11:38:05 am »
This post shows my low resistance short used for CC power supply testing. The jacks themselves are the weak point here. Very small resistance changes have quite an effect on what the minimum drive voltage is on CC mode.

Offline Dr. Frank

  • Super Contributor
  • ***
  • Posts: 2377
  • Country: de
Re: HP 34401a - Drift in the 2W R mode / 100 Ohm range
« Reply #19 on: July 29, 2014, 04:27:35 pm »
Hello Zucca,

you marked correctly the relevant components for 2W, that are the input jacks themselves, the S1-A and S1-G switch (my favorite) and the K101 relays.
And of course all of their solder junctions. 

During calibration, the lab has to short the front and rear jacks, then measure zero for all modes for front and rear.
Therefore they have to actuate the front / rear switch.
And they also put a mechanic force on the jacks, when they remove the shorts again.

Both may give rise to that change in resistance over many minutes, which you also measured.

The change can be as high 100mOhm, and  may be as small as a few milli Ohm, after the switch has actuated several times.

Then it's clear, that the 100Ohm calibration fails, as the zero Ohm value is not valid any more, after actuation.

If you look closely at your 100Ohm , 2W calibration protocol, the difference between the measured, failed  value is 99.9052, and the 'good' limit of 99.9806Ohm, this difference is about 80mOhm, or even 100mOhm compared to the nominal value, which can really happen by a bad switch, or bad solder junctions.

You may test the switch resistance by directly shorting both 'front' position switches S1-A and S1-G and measuring the real change in resistance.

The quality of the input jack may be measured by shorting High and Low first at the jacks  and afterwards directly at switch S1-A and S1-G, so that you circumvent the jacks.

Frank
 

Online ZuccaTopic starter

  • Supporter
  • ****
  • Posts: 4298
  • Country: it
  • EE meid in Itali
Re: HP 34401a - Drift in the 2W R mode / 100 Ohm range
« Reply #20 on: July 30, 2014, 03:39:31 am »
Quick and dirty update.

Between pin 20 and pin 2 of the switch, I measured 54 mOhm (4W with 34461a) the HO and LI Front were shorted by the Pomona MDP S.

The 54 mOhm were so distributed:

1) 07 mOhm between pin 20 S1 and pin 21 S1 (Front switch on)
2) 26 mOhm between pin 21 S1 and LO
3) 03 mOhm between LO and HI (across Pomona MDP S)
4) 10 mOhm between HI and pin 1 S1
5) 08 mOhm between pin 1 S1 and pin 2 S1 (Front switch on)

Just a roughly measure, tomorrow I will try to do something more accurate...

Can't know what you don't love. St. Augustine
Can't love what you don't know. Zucca
 

Online ZuccaTopic starter

  • Supporter
  • ****
  • Posts: 4298
  • Country: it
  • EE meid in Itali
Re: HP 34401a - Drift in the 2W R mode / 100 Ohm range
« Reply #21 on: July 31, 2014, 05:20:19 am »
Interesting evening...

I open that bastard and took live measures


I activated the 2WR mode with a short bar between HI and LO. Then with my trusted 34461A I took some µV measures referred to AGND0 (Analog ground).
Here the points I analyzed, let´s assign a letter for each one:







The point A drifts pointing the sky.



Here is my theory: if a constant current starts to flow through a R, the R temperature will rise for the Joule effect. Now if the temperature increase in a resistor and it has a positive temperature coefficient the resistor value will increase as well. Since the current is hold constant by the current source circuit, bingo, the voltage is linearly going to the sky until some other black magic happen in the R. I can be wrong.
Anyway after playing around and measuring other points I don´t know why but the voltage jumped down to 200µV again. After that is start to going up slowly but it seems to reach equilibrium at about 211µV.

Let´s measure all those points:



The K101 Relay-Reed 2A 500 Vdc 5 Vdc-Coil is now under arrest, 105mOhm (even 110mOhm I got later on) is too much. I checked also the coil voltage, 5VDC spot on. Definitely a relay contact topic.

Next step:

1) Reflow the K101 joints (no hope step, but maybe?)

2) High Contact Resistance Remediation

That´s cool!

Quote
High Contact Resistance Remediation

If switching very low currents with armature relays is unavoidable, there are still things that can be done.

1. Increase the Switching Current

As stated above, increasing the switching current can clean the adsorbed/polymerized material from the contacts. The amount of current will vary depending of relay size.  For small signal relays this could be as little as 10mA. There is no one correct answer for this value. The lowest value that can clean the contacts should be used.  A current that is too high can cause other problems that also degrade the contact resistance.

2. "Buzz” the Relay

Another technique for cleaning contacts is to “buzz” them. This method actuates the problem relay as fast as possible for about one second. The act of rapidly cycling the relay knocks deposits from the relay contacts. This method has the advantage that no fixtures or other measurement devices are needed.

Neither of these two methods is a permanent solution and may need to be performed periodically. Every application is different and it is not possible to say in advance how many switch cycles can be performed before maintenance is required.  When this is determined, the relay operations counter function found on most National Instruments switch products can be used to predict future cleaning cycles.

for more details:

http://www.ni.com/white-paper/7200/en/

I can´t wait to pump some current (YES 4A@1V?  >:D ) in that little bastard... For the "buzz" method I have to desolder it and build something around it.

3) Dremel and open it up for surgery...  :-X

4) Replace it

It seems like it´s not so easy to buy a new one...
https://www.eevblog.com/forum/chat/help-me-find-a-replacement-relay/

So I need to buy a new kick ass 5VDC Relay and build a fixture for it...





 
« Last Edit: September 30, 2014, 01:01:12 pm by zucca »
Can't know what you don't love. St. Augustine
Can't love what you don't know. Zucca
 

Online ZuccaTopic starter

  • Supporter
  • ****
  • Posts: 4298
  • Country: it
  • EE meid in Itali
Re: HP 34401a - Drift in the 2W R mode / 100 Ohm range
« Reply #22 on: August 01, 2014, 01:20:01 am »
No.

 :(

I checked other two calibrated 34401a and in both the R of a closed K101 is about 100/130 mOhm.
I also compared the others points and nope nothing really different. The point A is also at about 211µV in the others.

I have now to play the card of the sporadic failure (not reproducible under my eyes).

Let´s reflow the 2W joints and pray.

Take care, thank´s to all who posted here to help me.

PS: Worst case I will try to calibrate the unit at home by comparing with the 34461a numbers.
« Last Edit: August 01, 2014, 03:30:45 am by zucca »
Can't know what you don't love. St. Augustine
Can't love what you don't know. Zucca
 

Offline ManateeMafia

  • Frequent Contributor
  • **
  • Posts: 730
  • Country: us
Re: HP 34401a - Drift in the 2W R mode / 100 Ohm range
« Reply #23 on: August 01, 2014, 02:57:14 am »
Zucca,

Did you check L101,L102, and L106 for cracked joints? They appear to be in the path too and not far from the other resistors you found with cracked joints.

Also, Dr Frank mentioned jumping out the switch contacts, did you try that too? It should be easy to jump out across both contacts and watch the resistance change.
This method has helped me track down the same issue with a couple of my meters, though not this particular model.
 

Online ZuccaTopic starter

  • Supporter
  • ****
  • Posts: 4298
  • Country: it
  • EE meid in Itali
Re: HP 34401a - Drift in the 2W R mode / 100 Ohm range
« Reply #24 on: August 01, 2014, 03:29:01 am »
Hi ManateeMafia,

No I did not check those, but according to the theory L101,L102, and L106 will not cause just the 2W R Test to fail.
I did also not performed the jumping out test, since the result would have been pretty obvious once you got all the ladder of voltages from point A to P.

Anyway I reflow all the 2W joints, all the S1 switch pins and the tall capacitors since they are more sensible to vibrations. Now point A seems stable to 183µV in 2W R with a shortcut between HI and Lo.

Maybe I´m lucky this time.
« Last Edit: August 01, 2014, 03:34:25 am by zucca »
Can't know what you don't love. St. Augustine
Can't love what you don't know. Zucca
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf