How should I clean a 400 k ohm resistor?
I want to change a 400 k ohm through hole resistor in a HP 34970B data
acquisition unit
The resistor is only 1% tolerance, but has a temperature coefficient is 2
ppm/deg C. Once the instrument is calibrated, changes in the value of this
resistor will mess up the resistance measurements on the data acquisition
unit.
What is the best way to fit the resistor in such that there are not any
avoidable changes in value?
I did think of cleaning the resistor with IPA once it Is soldered in, but
IPA readily absorbs water, so I am concerned IPA this will leave residual
around.
Dave.
Dr. David Kirkby,
Do you know that YOUR resistor has 2 ppm/deg C drift, or is that the maximum
allowable for the part number? Many times they perform BETTER than
specified.
-----Original Message-----
From: volt-nuts volt-nuts-bounces@lists.febo.com On Behalf Of Dr. David
Kirkby
Sent: Tuesday, June 25, 2019 7:51 AM
To: volt-nuts@lists.febo.com
Subject: [volt-nuts] How should I clean a 400 k ohm resistor?
I want to change a 400 k ohm through hole resistor in a HP 34970B data
acquisition unit
https://www.keysight.com/en/pd-1000001313%3Aepsg%3Apro-pn-34970A/data-acquis
ition-data-logger-switch-unit
The resistor is only 1% tolerance, but has a temperature coefficient is 2
ppm/deg C. Once the instrument is calibrated, changes in the value of this
resistor will mess up the resistance measurements on the data acquisition
unit.
What is the best way to fit the resistor in such that there are not any
avoidable changes in value?
I did think of cleaning the resistor with IPA once it Is soldered in, but
IPA readily absorbs water, so I am concerned IPA this will leave residual
around.
Dave.
Dr. David Kirkby,
volt-nuts mailing list -- volt-nuts@lists.febo.com To unsubscribe, go to
http://lists.febo.com/mailman/listinfo/volt-nuts_lists.febo.com
and follow the instructions there.
On Tue, 25 Jun 2019 at 13:02, Mitch Van Ochten Mitch@vincentelectronics.com
wrote:
Do you know that YOUR resistor has 2 ppm/deg C drift, or is that the
maximum
allowable for the part number? Many times they perform BETTER than
specified.
I believe 2 ppm/deg C is the specification. I am not even 100% sure of
that, as I only have a Keysight part number 0699-3404 <0699-3404%20400>
https://www.keysight.com/my/faces/partDetail.jspx?partNumber=0699-3404
If 400 k ohm was not such an obscure value I would have fitted a higher
specification resistor, but there’s very little choice in 400 k ohm.
There is a service note on this instrument that indicates the 400 k ohm
resistor drifts, in some serial numbers. My instrument is missing the
serial number, so I purchased a resistor just in case mine is fitted with a
dodgy resistor.
Dave.
--
Dr. David Kirkby,
Hello,
you know that even a metal foil resistor can drift several 10 ppms due
to soldering?
(and may need several months to stabilize).
https://www.eevblog.com/forum/projects/vishay-bulk-foil-drift-after-soldering/msg445297/#msg445297
Of course the effect may be less on oil filled metal can resistors.
But after exchangeing the reference resistor I would wait some months
before next calibration.
With best regards
Andreas
Am 25.06.2019 um 13:50 schrieb Dr. David Kirkby:
I want to change a 400 k ohm through hole resistor in a HP 34970B data
acquisition unit
The resistor is only 1% tolerance, but has a temperature coefficient is 2
ppm/deg C. Once the instrument is calibrated, changes in the value of this
resistor will mess up the resistance measurements on the data acquisition
unit.
What is the best way to fit the resistor in such that there are not any
avoidable changes in value?
I did think of cleaning the resistor with IPA once it Is soldered in, but
IPA readily absorbs water, so I am concerned IPA this will leave residual
around.
Dave.
In message 782317b5-781c-e5ec-1cf8-44be9339663e@t-online.de, Andreas Jahn wri
tes:
you know that even a metal foil resistor can drift several 10 ppms due
to soldering?
You can bring that down both in PPM and time by a "degauss"
temperature-cycling sequence [+N, -N, +(N-2), -(N-2), +(N-4), -(N-4) ...]
--
Poul-Henning Kamp | UNIX since Zilog Zeus 3.20
phk@FreeBSD.ORG | TCP/IP since RFC 956
FreeBSD committer | BSD since 4.3-tahoe
Never attribute to malice what can adequately be explained by incompetence.
On Wed, 26 Jun 2019 at 06:05, Poul-Henning Kamp phk@phk.freebsd.dk wrote:
In message 782317b5-781c-e5ec-1cf8-44be9339663e@t-online.de, Andreas
Jahn wri
tes:
you know that even a metal foil resistor can drift several 10 ppms due
to soldering?
You can bring that down both in PPM and time by a "degauss"
temperature-cycling sequence [+N, -N, +(N-2), -(N-2), +(N-4), -(N-4) ...]
--
Poul-Henning Kamp | UNIX since Zilog Zeus 3.20
The description of the resistor on the Keysight website is
Resistor-Fixed 400K Ohm +-1PCT 0.25W TC+-2 thin film THT
I assume that it 400 k, 1%, 0.25 W, +/- 2 ppm/deg C, thin film, but I
don’t know. The resistor fitted looks like one of the expensive Vishay
0.01% resistors I have around here. Given the instrument is about 20 years
old, it has crossed my mind whether I am better leaving the original
resistor in place, despite I have have bought a replacement from Keysight -
the only source I could find of the 400 k ohm value.
I assume from your equation you mean heat the resistor a little, cool it a
little, then heat it to a higher temperature , then cool it to a lower
temperature. Unfortunately I don’t have any environmental chamber. The only
thing I could realistically do is to power cycle it every 15 minutes
whilst leaving it in an unheated garage. At night the temperature would
probably fall to around 15 deg C and during the day it would probably reach
25 degrees C. That’s not what your equation suggests, but it is probably
about the only way I have of causing temperature changes.
The temperature inside this data acquisition unit will charge when I next
power it on as I have replaced the enclosure for a ventilated enclosure and
replaced the fan for a different type. Those changes are suggested in one
of the service notes. The unit has internal temperature sensors for cold
junction compensation of thermocouples. Apparently the changes to the case
and fan improve the cold junction compensation.
I am amazed that the original enclosure design works at all as the outlet
of the fan is a couple of mm from the side panel of the case, which has no
ventilation holes! I don’t know who could have thought of that bright idea.
The new case has ventilation holes right by the fan outlet, as well as a
number of other ventilation holes. The shape of the impeller on the new fan
is different from the original fan. The fan is very small - around 40 x 40
x 8 mm from memory.
Dave.
Dr. David Kirkby,
Are we still talking about the HP 34970A data acquisition unit?
I have a service note for the 34401A DMM, ref 34401A-05A dated October 1994, that refers to this 400K resistor and/or a 40K resistor of the same type causing problems in some 34401As by ageing out of spec, but although the 34970A internal DMM is based on the 34401A I do not recall having seen a similar service note for the 34970A.
I may of course have missed it, and I don't know when my 34970A was manufactured although what might be an original hP cal sticker suggests c2001, but I don't recall either seeing any reference the 34970A in HP catalogues prior to c1998 so have always assumed that no 34970As would be affected.
Nigel, GM8PZR
-----Original Message-----
From: Dr. David Kirkby drkirkby@kirkbymicrowave.co.uk
To: Discussion of precise voltage measurement volt-nuts@lists.febo.com
Sent: Wed, 26 Jun 2019 8:47
Subject: Re: [volt-nuts] How should I clean a 400 k ohm resistor?
On Wed, 26 Jun 2019 at 06:05, Poul-Henning Kamp phk@phk.freebsd.dk wrote:
In message 782317b5-781c-e5ec-1cf8-44be9339663e@t-online.de, Andreas
Jahn wri
tes:
you know that even a metal foil resistor can drift several 10 ppms due
to soldering?
You can bring that down both in PPM and time by a "degauss"
temperature-cycling sequence [+N, -N, +(N-2), -(N-2), +(N-4), -(N-4) ...]
--
Poul-Henning Kamp | UNIX since Zilog Zeus 3.20
The description of the resistor on the Keysight website is
Resistor-Fixed 400K Ohm +-1PCT 0.25W TC+-2 thin film THT
I assume that it 400 k, 1%, 0.25 W, +/- 2 ppm/deg C, thin film, but I
don’t know. The resistor fitted looks like one of the expensive Vishay
0.01% resistors I have around here. Given the instrument is about 20 years
old, it has crossed my mind whether I am better leaving the original
resistor in place, despite I have have bought a replacement from Keysight -
the only source I could find of the 400 k ohm value.
I assume from your equation you mean heat the resistor a little, cool it a
little, then heat it to a higher temperature , then cool it to a lower
temperature. Unfortunately I don’t have any environmental chamber. The only
thing I could realistically do is to power cycle it every 15 minutes
whilst leaving it in an unheated garage. At night the temperature would
probably fall to around 15 deg C and during the day it would probably reach
25 degrees C. That’s not what your equation suggests, but it is probably
about the only way I have of causing temperature changes.
The temperature inside this data acquisition unit will charge when I next
power it on as I have replaced the enclosure for a ventilated enclosure and
replaced the fan for a different type. Those changes are suggested in one
of the service notes. The unit has internal temperature sensors for cold
junction compensation of thermocouples. Apparently the changes to the case
and fan improve the cold junction compensation.
I am amazed that the original enclosure design works at all as the outlet
of the fan is a couple of mm from the side panel of the case, which has no
ventilation holes! I don’t know who could have thought of that bright idea.
The new case has ventilation holes right by the fan outlet, as well as a
number of other ventilation holes. The shape of the impeller on the new fan
is different from the original fan. The fan is very small - around 40 x 40
x 8 mm from memory.
Dave.
Dr. David Kirkby,
volt-nuts mailing list -- volt-nuts@lists.febo.com
To unsubscribe, go to http://lists.febo.com/mailman/listinfo/volt-nuts_lists.febo.com
and follow the instructions there.
On Wed, 26 Jun 2019 at 16:33, Nigel Clarke via volt-nuts <
volt-nuts@lists.febo.com> wrote:
Are we still talking about the HP 34970A data acquisition unit?
Yes Nigel,
I have a service note for the 34401A DMM, ref 34401A-05A dated October
1994, that refers to this 400K resistor and/or a 40K resistor of the same
type causing problems in some 34401As by ageing out of spec, but although
the 34970A internal DMM is based on the 34401A I do not recall having seen
a similar service note for the 34970A.
There are quite a number of service notes for the 34970A here
The particular one I should probably address is "10 MegOhm and 100 MegOhm
ranges drift out of specification due to the Ohm's current source reference
resistor. This resistor may change in value with time"
http://literature.cdn.keysight.com/litweb/pdf/34970A-13.pdf
It applies to these serial numbers
MY41003000 / MY41014500
SG41002300 / SG41007600
I have no idea if my S/N falls into one of those ranges, and I don't have
any 10 or 100 M ohm resistors of known value. I wish I had a 3458A! The
best course of action is probably to replace the resistor. Mine is an old
HP unit, so I suspect it was made in the USA, so its S/N it outside those
ranges, but I don#t know.
There's another issue that may affect you too, regarding a problem on
resistance measurements on 50 Hz mains supplies. That needs a firmware
upgrade. I have version 12 of the firmware if you are able to blow the
PROM, but you will need a pretty decent programmer for that, as its not a
common DIL chip. A version 13 PROM is available for about $45, but Keysight
UK will replace the PROM free of charge if the instrument is sent for
calibration.
Another problem is a safety issue.
I may of course have missed it, and I don't know when my 34970A was
manufactured although what might be an original hP cal sticker suggests
c2001, but I don't recall either seeing any reference the 34970A in HP
catalogues prior to c1998 so have always assumed that no 34970As would be
affected.
Nigel, GM8PZR
Dave
--
Dr David Kirkby Ph.D C.Eng MIET
Kirkby Microwave Ltd
Registered office: Stokes Hall Lodge, Burnham Rd, Althorne, CHELMSFORD,
Essex, CM3 6DT, United Kingdom.
Registered in England and Wales as company number 08914892
https://www.kirkbymicrowave.co.uk/
Tel 01621-680100 / +44 1621-680100
Oh boy, that's quite a collection of service notes, and quite some can of worms:-)
Thanks for the link, I've not seen any of those before and wonder why they don't include them on the 34970A documentation page?
For me the good news is that my later unit with vented case has a US... serial number so shouldn't be affected by the drifting resistor problem.The not so good news is that I do have firmware version 11.
Do you have a type number for a suitable eeprom?, I have a variety of adapters for my Leaper48 programmer so might be able to program it, although that issue does look to be a quite specific problem that might be easy enough to avoid once aware of it.
Another bit of not so good news, I've just noticed my rotary encoder has become very intermittent.
Ho hum, there's just never an end to the fun:-)
Nigel, GM8PZR
-----Original Message-----
From: Dr. David Kirkby drkirkby@kirkbymicrowave.co.uk
To: Discussion of precise voltage measurement volt-nuts@lists.febo.com
CC: gandalfg8 gandalfg8@aol.com
Sent: Wed, 26 Jun 2019 17:44
Subject: Re: [volt-nuts] How should I clean a 400 k ohm resistor?
On Wed, 26 Jun 2019 at 16:33, Nigel Clarke via volt-nuts <
volt-nuts@lists.febo.com> wrote:
Are we still talking about the HP 34970A data acquisition unit?
Yes Nigel,
I have a service note for the 34401A DMM, ref 34401A-05A dated October
1994, that refers to this 400K resistor and/or a 40K resistor of the same
type causing problems in some 34401As by ageing out of spec, but although
the 34970A internal DMM is based on the 34401A I do not recall having seen
a similar service note for the 34970A.
There are quite a number of service notes for the 34970A here
The particular one I should probably address is "10 MegOhm and 100 MegOhm
ranges drift out of specification due to the Ohm's current source reference
resistor. This resistor may change in value with time"
http://literature.cdn.keysight.com/litweb/pdf/34970A-13.pdf
It applies to these serial numbers
MY41003000 / MY41014500
SG41002300 / SG41007600
I have no idea if my S/N falls into one of those ranges, and I don't have
any 10 or 100 M ohm resistors of known value. I wish I had a 3458A! The
best course of action is probably to replace the resistor. Mine is an old
HP unit, so I suspect it was made in the USA, so its S/N it outside those
ranges, but I don#t know.
There's another issue that may affect you too, regarding a problem on
resistance measurements on 50 Hz mains supplies. That needs a firmware
upgrade. I have version 12 of the firmware if you are able to blow the
PROM, but you will need a pretty decent programmer for that, as its not a
common DIL chip. A version 13 PROM is available for about $45, but Keysight
UK will replace the PROM free of charge if the instrument is sent for
calibration.
Another problem is a safety issue.
I may of course have missed it, and I don't know when my 34970A was
manufactured although what might be an original hP cal sticker suggests
c2001, but I don't recall either seeing any reference the 34970A in HP
catalogues prior to c1998 so have always assumed that no 34970As would be
affected.
Nigel, GM8PZR
Dave
--
Dr David Kirkby Ph.D C.Eng MIET
Kirkby Microwave Ltd
Registered office: Stokes Hall Lodge, Burnham Rd, Althorne, CHELMSFORD,
Essex, CM3 6DT, United Kingdom.
Registered in England and Wales as company number 08914892
https://www.kirkbymicrowave.co.uk/
Tel 01621-680100 / +44 1621-680100
volt-nuts mailing list -- volt-nuts@lists.febo.com
To unsubscribe, go to http://lists.febo.com/mailman/listinfo/volt-nuts_lists.febo.com
and follow the instructions there.
n Wed, 26 Jun 2019 at 19:38, Nigel Clarke via volt-nuts <
volt-nuts@lists.febo.com> wrote:
Oh boy, that's quite a collection of service notes, and quite some can of
worms:-)
Thanks for the link, I've not seen any of those before and wonder why they
don't include them on the 34970A documentation page?
Nigel, GM8PZR
I did manage to find the service notes myself from the Keysight website, so
I guess they are linked somewhere, but I must admit they are not easy to
find.
I put version 12 of the firmware here - the latest is version 13.
https://www.kirkbymicrowave.co.uk/Support/34970A/
so if you want to update the firmware to version 12, you can do. I put some
notes about possible PROMS. To be honest, given the latest (version 13)
PROM is only $41.29 from Keysight, you might consider it less hassle and
better to just buy the chip and fit it. Then you have the latest version,
and no need to mess about. If you send the instrument to Keysight UK for
calibration, you can get the firmware updated free. although they kick up a
bit of argument about it, but I got my 4285A LCR meter updated, and they
have agreed to do likewise for the 34970A.
The 34970A does look a very useful instrument for any volt nut, time nut,
or anyone else that wants to collect electrical data and process it.
Dr David Kirkby Ph.D C.Eng MIET
Kirkby Microwave Ltd
Registered office: Stokes Hall Lodge, Burnham Rd, Althorne, CHELMSFORD,
Essex, CM3 6DT, United Kingdom.
Registered in England and Wales as company number 08914892
https://www.kirkbymicrowave.co.uk/
Tel 01621-680100 / +44 1621-680100