(71B) HP 3421A DATA AQUISTION/CONTROL UNIT calibration

[Erwin]

Hi,

Now I started to work with the DAQ-Unit I planned 2 years ago. I own a HP 3421A Data Acquisition Unit and want to use it for a temperature measuring project in the future. Now with the FRAM71B I have the possibility to have all my necessary programs on board of the HP71B.
Before using the HP 3421A it is necessary to calibrate the multiplexer boards. HP does not provide a calibration program for the HP71B (only in a full automated setting with some additional instruments) and I'm wondering there is not something out in the wild. HP offers two possibilities for calibrating. One is the manual calibrating with the temperature calibration board.
The effort for manual calibrating the multiplexer with the temperature calibrator board is relatively high. So I wrote a small basic program to overtake this manual part and carries out the calibration semi-automatically.

1. prepare the multiplexer boards like the HP description says. It is important to follow HP's instructions (switch 8!!) The program makes a hint on this switch, the time to wait (10 seconds), and so on - cause I often forgot this.
2. start the basic program - first it looks for the HP3421A in the HPIL-loop and checks in which slot the multiplexer board is situated (I have two of them)
3. Then choose the board you want to calibrate. On every pass it displays the calibrated and the reference value and the number of passes
   
   Code:
   

   Pass No:slot C:tt.tt [calibrated] R:TT.TT [reference]. The values are in degree celsius (you can change this in line 10)
4. the program stops if the board is calibrated or with a limited run of 9 passes - normally the board should be calibrated otherwise you can start it again (my second board takes two sequences) or change to another board. The program shows the last values for checking the progress.

I did it fast and so it's not a code optimized program but it works for me and hopefully for others too. Maybe I'll do some code clearance later on. There is no additional comfortable error handling implemented. At the moment there is no calibration report - will be added in the future. I am happy about your feedback.

best regards
Erwin

EDIT 2017-01-06 small code improvements, add check SQR error code, put program in LIF file

Planning next steps:

• Menue driven
• Error check during program execution
• First draft for semi automatic DC calibration
  

Code:

10 ! HP 3421A DATA AQUISTION/CONTROL UNIT
20 ! Manual temperatur calibration of HP44462A Multiplexer
30 ! using the temperatur calibrator board
40 ! (c) Erwin Grabler, 2017
50 ! BINLEX and DATA-AQ-Module is necessary
60 !
70 DELAY 2 @ DESTROY ALL @ OPTION DEGREES C
80 A$=ADR3421$ @ DISP "** INIT3421 "&A$&" **"
90 DELAY 9 @ PRINT 'Switch 8 on 3421 UP"
100 INIT3421
110 DELAY 9 @ PRINT "Switch 8 on 3421 DOWN"
120 ! Check cardtype in slots
130 GOSUB 'CHECK'
140 DELAY 2 @ DISP "* Start calibration *"
150 DISP "* Load thermo data *"
160 A=.001285498378
170 B=.000230998857
180 C=9.324409398E-8
190 DISP "* Calibration loop *"
200 'LOOP':
210 INPUT "Check slot no: ";I
220 F=VAL(STR$(I)&STR$(4))
230 FOR L=1 TO 10
240 R=CMD3421(S$(I))
250 D=1/(A+B*LOG(R)+C*LOG(R)^3)-273.16 @ D=INT(D*100+.5)/100
280 E=REFTEMP(F) @ OPEN F
270 DISP STR$(L-1)&":"&STR$(I)&" "&"C:"D;"R:";E
280 GOSUB 'KALI'
290 IF D=E THEN 'END3421'
300 NEXT L
310 'EN03421':
320 DELAY 9 @ DISP "** End calibration **"
330 DISP "C"&STR$(I)&":";D;"/ R"&STR$(I)&":";E
340 DISP "* Check SQR register *" @ DELAY 9
350 OUTPUT A$;"SR" @ ENTER A$ ;K,L,M,N
360 DISP "Status Reg: ";DTBIN$(K,8)
370 DISP "Error Reg: ";DTBIN$(L,8)
380 DISP "Hardware Reg: ";DTBIN$(M,8)
390 DISP "Calibr Reg: ";DTBIN$(N,8)
400 DELAY 2 @ DISP "Close port ";F @ OPEN F
410 INPUT "Other slot (Y/N)";X$
420 IF X$="Y" THEN GOTO 'LOOP'
430 DELAY 9 @ PRINT "Switch 8 on 3421 UP"
440 OUTPUT AS ;"SR"
450 DELAY 0
460 END
470 'CHECK': DELAY 2 @ DISP "**** Check slots ****"
480 DELAY 9 @ OPTION BASE 0 @ DIM S$(3)
490 FOR I=0 TO 2 @ PRINT "Slot";I;
500 IF NOT CHANTYPE(I*10) THEN DISP "is empty"
510 IF CHANTYPE(I*10+2)=1 THEN DISP "relay board" @ S$(I)="TWO"&STR$(I)&STR$(4)
520 IF CHANTYPE(I*10)=2 THEN DISP "digital board"
530 NEXT I
540 RETURN
550 'KALI': DELAY 2 ! DISP "** Calibration loop **"
560 OUTPUT A$ ;"REF“&S$(I)
570 OUTPUT A$ ;"C00"&STR$(D)
580 OPEN F
590 WAIT 15
600 RETURN

[Dave Frederickson]

Hi Erwin,

Coincidentally, I was thinking of writing a calibration routine for the 3421A using the 71B. Unfortunately I don't have access to the necessary standards. I'll give your program a try.

I see that the Data Acquisition ROM is required. FRAM71 comes in handy in this regard.

Dave

[Erwin]

(01-05-2017 07:40 PM)Dave Frederickson Wrote:  Hi Erwin,

Coincidentally, I was thinking of writing a calibration routine for the 3421A using the 71B. Unfortunately I don't have access to the necessary standards. I'll give your program a try.

I see that the Data Acquisition ROM is required. FRAM71 comes in handy in this regard.

Dave

Hi Dave,

I did it like the HP papers said to do it in the manual way (if you don't have the paper I can make a copy and send it to you) - for me its running. The number of loops through the program depends on the hardware - I have two of them and one takes 25 times, the other 7 times. It's important that there is a constant temperature during the measurement and a "burn in time" of the 3421A minimum from 1 hour.

I have no possibility to check the result with an external instrument (this have to Bussi Erwin the next step) so I can't say the program is doing it right. But the figures showed it (but never trust only figures).
In the meanwhile I changed the waiting time from 10 to 15 seconds to be on the save side - its only necessary after writing the reference but I did it after the read cycle too.
I don't have the test-extension for the digital-board so I wasn't able to include this test.

I'm happy about your feedback and to sharpen the program
Erwin

[Dave Frederickson]

Hi Erwin,

Be aware that the manual temperature calibration procedure uses the internal voltmeter of the 3421A. For the temperature calibration to be accurate, the DC volts needs to calibrated first.

Edit: The temperature calibration procedure makes resistance measurements which is part of the voltmeter calibration, so resistance needs to be calibrated also, after DC volts.

Edit #2: Before diving into what is now becoming a much bigger task, are your batteries good? The CAL Data battery can be replaced with a Panasonic BR-2/3AE5SPN, Digi-Key P226-ND. The 6V 3.2Ah SLA battery can be replaced with a Powersonic PS-632. There's a nice procedure for replacing the CAL battery here: http://www.eevblog.com/forum/repair/owon...t=prev#new
Note: This is exactly the same battery as the one in the 3468 DMM.

Dave

[Erwin]

(01-05-2017 09:09 PM)Dave Frederickson Wrote:  Hi Erwin,

Be aware that the manual temperature calibration procedure uses the internal voltmeter of the 3421A. For the temperature calibration to be accurate, the DC volts needs to calibrated first.

Edit: The temperature calibration procedure makes resistance measurements which is part of the voltmeter calibration, so resistance needs to be calibrated also, after DC volts.

Dave

Hi Dave, yes I know this - but this is the more complicated part, cause it needs a multi-scale voltage reference (to check all the DC ranges), a 6 (5) digit multimeter. Also AC calibration should be done - with different frequencies. High precision resistor test and this should be done before temperature calibration.

Then each calibration data is set in the calibration RAM.

There exist a tape for the HP85 I think it should be a BASIC program. But I could not get hands on one of this - could be easy to transfer the program to the HP71B and perform most of the tests. Maybe some have a source for this?

regards
Erwin

[Dave Frederickson]

Hi Erwin,

I inquired about the calibration program about 3 weeks ago with hpseries80@groups.io.

https://groups.io/g/hpseries80/message/1...40,3784341

I also asked Larry Atherton who repairs Series 80 computers and calibrates some HP equipment. No luck.

Dave

[Dave Frederickson]

Hi Erwin,

The B parameter in line 170 should be 0.0002360998857.

Regards, Dave

[Erwin]

(03-31-2017 02:12 AM)Dave Frederickson Wrote:  Hi Erwin,

The B parameter in line 170 should be 0.0002360998857.

Regards, Dave

Hi Dave, thanks
I'm on a project outside my home so I'm not able to finish my work on the calibration program for a longer time. Thanks for the hint to the faulty numbers. I did a rewrite of the program - but still not finished - with additional measurement tasks. The performance verification is ready but not tested also the counter check.
Not ready is the calibration for DC, AC and Ohms ... I did this all without my HP 4321A.

The program at the moment consists of the two parts: performance and calibration of the main unit and performance and calibration of the MUX board.

If it would help I can PM you the actual version of the program. Maybe we could finish the program as a 'combination' project if you want.

best regards
Erwin

[Dave Frederickson]

Hi Erwin,

I'd be happy to take a look at your program, but I'm not up to collaborating on a project. I don't have the calibration standards necessary to validate a performance verification or calibration program.

Right now I'd just be happy if I could clear the CAL RAM errors on my 3421 and 3468. If I could just figure out how the undocumented CAL command works.
http://hpmuseum.org/forum/thread-8061.html

Regards, Dave