USA calibration club

[nikonoid]

I still do not have my logging working. I am planning to use some of what you bitseeker did. Thanks.


Sent from my iPhone using Tapatalk

[bitseeker]

nikonoid, I used a version of BenchVue from last year on my laptop for the meter logging as it was relatively straightforward. The only downside was that the logging times out after an hour, at which point you have to start a new file. The current version of BenchVue only has a trial mode and I'm not sure what the limitations are during the trial period.

At some point, I'd like to move logging operations to my Pi or BBB.

[orin]

Orin, I cannot honestly answer the 3455A jitter question, as I discovered the +12V regulator has a fault both producing some very ugly ripple, and loading down the power supply by an extra 200mA.  I located a NOS replacement but it is hanging in limbo at a Houston post office.  No idea when it will arrive.  I'm half tempted to use a TO-220 regulator and just bolt it to the dang chassis, haha! 

Regards,
Chris

Chris,

This is a reply to your pm about the 12V power supply in the 3455A outguard.  Pre-regulator (J5 pin 2) is 18.6V and about 1.2V ripple.  Post-regulator (J5 pin 5) is 11.9V and 200mV noise.  Nothing really surprising there.  'scope shots attached at same vertical resolution*... not much point posting a higher resolution shot of the post-regulator noise.

Orin.

*multiply scope readings by 10 - someone forgot to set the scope probe setting.

[bitseeker]

Since the heatwave has passed, I did another run with the same meters. As before, I logged for two hours, but only pay attention to the second because the meters aren't really stable during the first hour even though I warm everything up half an hour prior to starting.

All logging at 100 NPLC as before. Data files will be uploaded to the xdev's share.

34401A

Average	10.00027
Std Dev	0.000002
Max	10.00028
Min	10.00026

34410A

Average	10.00026
Std Dev	0.000006
Max	10.00027
Min	10.00024

Based on the temp/humidity logging over the entire two hours, the temp was rising slowing from 27.0 to 27.7°C. The 34410A's voltage was similarly rising, unlike the 34401A. The next time the roving standard comes around I hope to have the ability to do all-night logging to better see the stability.

During this run, I also had a Keithley 2700 attached to the SVR-T. I don't have it set up for logging right now, but it was even closer to 10V than the 34410A. Right now, the displayed values are as follows:

34410A:    10.00010
34401A:    10.00026
K2700:      10.00007  (Cal unknown)

Here are some other meters I hooked up for a quick check. The Keithley 196 and Fluke 8100A warmed up for about half an hour. The handheld Keysights and Greenlee warmed up for about five minutes.

K196:        10.00035 (Cal unknown)
F8100A:    10.002     (Cal unknown)
U1272A:     9.978      (Cal exp. 6/2013)
U1282A:     9.998      (Cal exp. 12/2017)
U1252B:   10.000      (Cal exp. 10/2017)
DM-820A: 10.00        (Cal unknown)

As for the 9764.02 pF capacitor in the box, here are the values I got.

34410A:   9.7689 nF
U1272A:   9.691  nF
U1282A:   9.789  nF
U1252B:   9.764  nF
DM-820A: 9.77    nF
M12864:   9711  pF

The M12864 is a version of the "$20 transistor tester".

Note: I've updated the TiSensorLogger.488 program on the USB stick and attached to my post. Changes include the following:

- Code tidying
- Flush bad data prior to the start of logging (the first temp and humidity samples are always wrong)
- Configurable logging interval (in seconds)
- Throw an error if unable to connect to COM port
- Blink right-most LED on EZGPIB UI while logger is running

@RandallMcRee: I'll be dropping the package off at the Post Office on Monday. You'll probably get it Tuesday or Wednesday.

[Conrad Hoffman]

A note on the capacitor- the value is for the device plugged into standard banana jacks, with the fixed capacitance of the jacks subtracted out. That means I would have measured the open circuit capacitance of my 1615A, probably about 1.2 pF if I remember right, and subtracted that from the reading. The banana method is OK, but not nearly as good as a cap in a shielded box with two shielded connectors. The problem is almost nobody has a bridge or other method of doing the proper shielded measurement.

[bitseeker]

Thanks for the reminder, Conrad. I didn't check for null on the meters that I measured the cap on. They all have banana jacks and I just plugged the thing in. I guess I can go back and see if they're already zero when open circuit and subtract accordingly. The U1252B was bang on based on the value on the little tag.

[CalMachine]

I used our HP4284A for my reported value of the capacitor.  Open/Short corrections were done beforehand, as well.

[bitseeker]

I already packed everything up. The value I posted was from the little tag. I believe yours was on the larger paper, correct? If so, then where did the value on the little tag come from? I don't remember how much you had noted via the HP 4284A, but I think it was a higher value.

[Conrad Hoffman]

Pretty sure I put a toe tag on the cap with the 1615A value. I have access to several GR 1000 pF standards that agree pretty well, so I trust the 1615 is well calibrated, if not traceable.

[bitseeker]

Yes, the toe tag read 9764.02 pF

[bitseeker]

@RandallMcRee: When you receive the package, can you post the capacitor value on the larger tag (the one that isn't 9764.02 pF)? Thanks!

[RandallMcRee]

Package received.

Capacitor tags:
Large tag: 9776.11 pF, D=0.000081 measured @1V 1Khz w/ HP4284A
Small tag: Cs=9764.02 pF, D=0.000092 (73F at end)

More later.

Randy

[bitseeker]

Thanks, Randy. So, the large tag is from CalMachine and the small one from Conrad.

[CatalinaWOW]

I am sort of infected with the volt nut disease.  But often run into cognitive dissonance.  I grew up in the slide rule era.  Discussing the third digit of capacitor readings, or the fifth and sixth digits of voltage readings still seems a bit bit surreal to me.  All of the attention required to temperature, humidity, terminal material, aging and stabilization times continually reminds me why in the more general world of electronics things beyond the second digit are rarely important.

[bitseeker]

Yeah, having recently finished the measurements logging, looking at 3.5- and 4.5-digit meters is rather serene. Hardly any fluctuations, no worry about noise, and moving in/out/around the room doesn't cause a disturbance. I don't know if I could handle more than 6.5.

[cellularmitosis]

I am sort of infected with the volt nut disease.  But often run into cognitive dissonance.  I grew up in the slide rule era.  Discussing the third digit of capacitor readings, or the fifth and sixth digits of voltage readings still seems a bit bit surreal to me.  All of the attention required to temperature, humidity, terminal material, aging and stabilization times continually reminds me why in the more general world of electronics things beyond the second digit are rarely important.

I was thinking about the notion of ppm the other day and came up with the following visualization to put it into perspective for myself:

Think about the length of a football field.  1 ppm is roughly the thickness of a sheet of paper.

[zhtoor]

I am sort of infected with the volt nut disease.  But often run into cognitive dissonance.  I grew up in the slide rule era.  Discussing the third digit of capacitor readings, or the fifth and sixth digits of voltage readings still seems a bit bit surreal to me.  All of the attention required to temperature, humidity, terminal material, aging and stabilization times continually reminds me why in the more general world of electronics things beyond the second digit are rarely important.

I was thinking about the notion of ppm the other day and came up with the following visualization to put it into perspective for myself:

Think about the length of a football field.  1 ppm is roughly the thickness of a sheet of paper.

or say a scale 1KM long, 1ppm is 1mm on that scale.

regards.

[bitseeker]

Think about the length of a football field.  1 ppm is roughly the thickness of a sheet of paper.

Good visualization aide.

[bitseeker]

D'oh, I just saw that I forgot to try out my HP 3469A on the SRV-T. Next time.

[kj7e]

I am sort of infected with the volt nut disease.  But often run into cognitive dissonance.  I grew up in the slide rule era.  Discussing the third digit of capacitor readings, or the fifth and sixth digits of voltage readings still seems a bit bit surreal to me.  All of the attention required to temperature, humidity, terminal material, aging and stabilization times continually reminds me why in the more general world of electronics things beyond the second digit are rarely important.

I was thinking about the notion of ppm the other day and came up with the following visualization to put it into perspective for myself:

Think about the length of a football field.  1 ppm is roughly the thickness of a sheet of paper.

A bit off topic but that reminded of this, fold said sheet of paper 103 times = the thickness of the universe;

[MartyZ]

I just got an HP 3468A. The internal battery was from 1985 and dead. I soldered in a CR123. The device is at Error 1, and was before.  I tried following the Service Manual on Calibration. On the first step with the leads shorted, It said Zero Done.  I attached the 3.0 volt source and pressed CAl and then SGL TRIG and this is where I got the problem.  It says invalid Voltage for Cal. Can anyone help me?

Second, is the device usable un-calibrated?

I did find nice adapters to convert to shielded leads on Amazon.

[kj7e]

I just got an HP 3468A. The internal battery was from 1985 and dead. I soldered in a CR123. The device is at Error 1, and was before.  I tried following the Service Manual on Calibration. On the first step with the leads shorted, It said Zero Done.  I attached the 3.0 volt source and pressed CAl and then SGL TRIG and this is where I got the problem.  It says invalid Voltage for Cal. Can anyone help me?

Second, is the device usable un-calibrated?

I did find nice adapters to convert to shielded leads on Amazon.

Welcome Marty,  I may suggest posting this question as a new topic in the Test Equipment section, here https://www.eevblog.com/forum/testgear/

I'm sure those who have run into the same situation will respond.

[RandallMcRee]

In my case silence is good. I have been busy getting all the bits and bobs to do what they are supposed to be doing:

New prologix gpib converter woking on both computers   
EZGPIB working 
Installed CatalinaWows Temp-Humidity sensor 
Modified his script to work on my laptop 
Logging started 

Preliminary data confirms what we already know: Keithley 6.5 digit meters (K2015) have 1ppm/degree TCR and even older, uncalibrated meters like to show 10.00007 at 23C.

Perhaps someone can point me to a good (non-excel, prob. not python) windows or cloud way to show  csv data graphically. Previous tools are all work-related and forbidden. I see that TiN has d3.js on xdevs but not sure how that works. Pointers?

o-o,
Randy

[bitseeker]

You can use the LibreOffice suite. Calc is similar to Excel (but free) for loading up and graphing your log file.

[CatalinaWOW]

I have found Libre Office slow for large files like these log files.  Octave, an open source Matlab look alike is recommended by many.  I have loaded it and am coming up to speed so can't really personally recommend it.-