Re: HP-3456A - Question Regarding Resistance Measurement
HI George,
Thanks for the information. I knew you'd have insight into this one.
I don't have a precision 10M resistor, but I did find a 10M in my parts drawer (I think it's metal film 5% but not sure about that tolerance band color). In any case, the 3456A measured it as 9.9172M.
I placed the questionable 90M and the 9.9172M in parallel and the 3456A showed the resulting resistance as 8.9441M. That makes the resulting value of the questionable 90M to be 91.1524M. Given that 90M a 1% could still be in tolerance at 90.9M, that's not too far off and given that my 3456A is not calibrated, then I'm guessing the 90M is probably within spec and the rest of that difference is attributable to lack of calibration.
Yes, I know, this is all very crude due to the lack of proper calibration but it at least gives me something that makes just slightly better sense than the directly-measured 90M value I was seeing.
BTW, did you ever find those Red Devil resistors? We discussed this very early this year and I think you said you thought you had some of those that could be used as a somewhat better resistance "standard" than what I happen to have but you were going to wait until your move was complete, etc., to locate them. If you do happen across those, I would still like to get one if that's still possible.
Thanks again,
Barry - N4BUQ
toggle quoted message
Show quoted text
----- Original Message ----- From: ghnatiuk@...
To: [email protected]
Sent: Tuesday, October 13, 2020 10:55:23 AM
Subject: Re: [HP-Agilent-Keysight-equipment] HP-3456A - Question Regarding Resistance Measurement
Barry,
You really need to calibrate your 3456A.? ?The Hewlett Packard / Agilent /
Keysight? DMM meters that were not specifically designed for high resistance
or low current applications as the 4329A or 34420A measure high resistances
by placing the 10 Meg Ohm voltage attenuator (used in the 100V and 1KV
range) in parallel with the unknown and computing the unknown value? from
1/ Rx =? ?1/ Rt? - 1/ 10 Meg
Rt = parallel combination of unknown and internal 10 Meg attenuator
Rx = unknown value
This method keeps the total resistance within the 10 Meg range that can be
measured with reasonable accuracy by the meter but the computation becomes
more and more inaccurate as the unknown resistance goes up in value due to
the meter's ability to resolve the difference between a 10 Meg resistor and
something in parallel with it.? ?Any inaccuracies in the 10 Meg range are
amplified using the extended ohms function.
The 10 Meg attenuator resistor is a? low TEMPCO? tantalum nitride? fine-line
resistor that is not trimmed to precision accuracy? ( 0.5% ) which will
limit the overall accuracy with which a resistor in parallel with it can be
measured by the computation method I described above.
The reason this is done in this manner for a multi-function DMM instrument is
that it is very difficult to make a 100nA current source that would be
needed to inject into a 100 Meg Ohm resistor to limit the voltage across it
so as not to overload the A/D converter input at 10V.? The Ohm's converter
circuitry? would require an internal range resistor on the order of 40 Meg
Ohm in the fine-line resistor network which then would take up a huge amount
of real estate in that network,? not to mention cost.
This extended Ohms function is also available thru the I/O ports on the 3468A
and 3478A meters that puts the 10 Meg attenuator resistance across the puts
terminals and supplies the value of the parallel combination of this with
the unknown over the bus.? Then with a calculator using this value? and the
equation above you can compute the value of the unknown.? There was not
enough room in the memory to code it into the firmware so we left it as an
extended ohms function that can only be accessed thru the I/O? bus.
Conclusion:? get your meter calibrated so the? attenuator 10 Meg resistor is
characterized.
OR in the alternative, obtain a precision 10 MEG or other resistor ( 0.1% or
better ) and put your unknown in parallel with it and measure the parallel
combination using the 10 Meg range of the 3456A and use the equation above
as does the 3456A to compute the unknown value - or use the voltage divider
method suggested earlier by? Stephen? Bell.? ?Good luck with this.
George Hnatiuk, PE
|
?Thank you, George, for this tantalizing information, but I implore you to redraw that schematic and post your sketch even if it's messy freehand.
Regards,
Dave Wise
toggle quoted message
Show quoted text
Hi All
Having now had time to find my notes, by serial prefix 2853A (1988) the fan had changed to a two wire one and the -5.2VDC supply and fan board had changed to part number 08640-60392.
This is a vastly simplified board using a three terminal regulator. The only schematic scan of this board that I have been able to find is of poor quality. I have attached a copy.
George G6HIG Dover UK
On Tuesday, 13 October 2020, 14:46:03 BST, alwyn.seeds1 <a.seeds@...> wrote:
Dear All,
It may be a myth, but a colleague told me that on some instruments the fans had controllers that were designed to avoid microphony problems with the instrument- possible that the HP8640B with its cavity oscillator might be one of
those?
Regards,
Alwyn
_____________________________________________________
Alwyn Seeds, Director
SynOptika Ltd.,
114 Beaufort Street,
London,
SW3 6BU,
England.
SynOptika Ltd., Registered in England and Wales: No. 04606737
Registered Office: 114 Beaufort Street, London, SW3 6BU, United Kingdom.
_____________________________________________________
|
Re: HP-3456A - Question Regarding Resistance Measurement
Barry,
You really need to calibrate your 3456A.? ?The Hewlett Packard / Agilent / Keysight? DMM meters that were not specifically designed for high resistance or low current applications as the 4329A or 34420A measure high resistances by placing the 10 Meg Ohm voltage attenuator (used in the 100V and 1KV range) in parallel with the unknown and computing the unknown value? from? ?
? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?1/ Rx =? ?1/ Rt? - 1/ 10 Meg
Rt = parallel combination of unknown and internal 10 Meg attenuator
Rx = unknown value
This method keeps the total resistance within the 10 Meg range that can be measured with reasonable accuracy by the meter but the computation becomes more and more inaccurate as the unknown resistance goes up in value due to the meter's ability to resolve the difference between a 10 Meg resistor and something in parallel with it.? ?Any inaccuracies in the 10 Meg range are amplified using the extended ohms function.
The 10 Meg attenuator resistor is a? low TEMPCO? tantalum nitride? fine-line resistor that is not trimmed to precision accuracy? ( 0.5% ) which will limit the overall accuracy with which a resistor in parallel with it can be measured by the computation method I described above.
The reason this is done in this manner for a multi-function DMM instrument is that it is very difficult to make a 100nA current source that would be needed to inject into a 100 Meg Ohm resistor to limit the voltage across it so as not to overload the A/D converter input at 10V.? The Ohm's converter circuitry? would require an internal range resistor on the order of 40 Meg Ohm in the fine-line resistor network which then would take up a huge amount of real estate in that network,? not to mention cost.
This extended Ohms function is also available thru the I/O ports on the 3468A and 3478A meters that puts the 10 Meg attenuator resistance across the puts terminals and supplies the value of the parallel combination of this with the unknown over the bus.? Then with a calculator using this value? and the equation above you can compute the value of the unknown.? There was not enough room in the memory to code it into the firmware so we left it as an extended ohms function that can only be accessed thru the I/O? bus.
Conclusion:? get your meter calibrated so the? attenuator 10 Meg resistor is characterized.
OR in the alternative, obtain a precision 10 MEG or other resistor ( 0.1% or better ) and put your unknown in parallel with it and measure the parallel combination using the 10 Meg range of the 3456A and use the equation above as does the 3456A to compute the unknown value - or use the voltage divider method suggested earlier by? Stephen? Bell.? ?Good luck with this.
George Hnatiuk, PE
|
On my 8640B the fan mechanical noise is clearly visible in the
close-in spectrum. Unplugging the fan makes the two little AM spikes
disappear from the screen.
Frank IZ8DWF
toggle quoted message
Show quoted text
On Tue, Oct 13, 2020 at 5:32 PM Dave Wise <david_wise@...> wrote:
Anecdote. I was doing some phase-locked FM work and saw ripple in my discriminator. The fix was lifting the 8640B off the Tektronix 500-series scope it was balanced atop. So yes it's worth an experiment or two.
Dave Wise
________________________________
From: [email protected] <[email protected]> on behalf of alwyn.seeds1 via groups.io <a.seeds@...>
Sent: Tuesday, October 13, 2020 6:45 AM
To: [email protected]
Subject: Re: [HP-Agilent-Keysight-equipment] HP8640B fan
Dear All,
It may be a myth, but a colleague told me that on some instruments the fans had controllers that were designed to avoid microphony problems with the instrument- possible that the HP8640B with its cavity oscillator might be one of those?
Regards,
Alwyn
_____________________________________________________
Alwyn Seeds, Director
SynOptika Ltd.,
114 Beaufort Street,
London,
SW3 6BU,
England.
SynOptika Ltd., Registered in England and Wales: No. 04606737
Registered Office: 114 Beaufort Street, London, SW3 6BU, United Kingdom.
_____________________________________________________
|
?Anecdote. ?I was doing some phase-locked FM work and saw?ripple in my discriminator. ?The fix was lifting?the 8640B off the Tektronix 500-series scope it was balanced atop. ?So yes it's?worth an experiment or two.
Dave Wise
toggle quoted message
Show quoted text
Dear All,
It may be a myth, but a colleague told me that on some instruments the fans had controllers that were designed to avoid microphony problems with the instrument- possible that the HP8640B with its cavity oscillator might be one of those?
Regards,
Alwyn
_____________________________________________________
Alwyn Seeds, Director
SynOptika Ltd.,
114 Beaufort Street,
London,
SW3 6BU,
England.
SynOptika Ltd., Registered in England and Wales: No. 04606737
Registered Office: 114 Beaufort Street, London, SW3 6BU, United Kingdom.
_____________________________________________________
|
Hi All
Having now had time to find my notes, by serial prefix 2853A (1988) the fan had changed to a two wire one and the -5.2VDC supply and fan board had changed to part number 08640-60392.
This is a vastly simplified board using a three terminal regulator. The only schematic scan of this board that I have been able to find is of poor quality. I have attached a copy.
George G6HIG Dover UK On Tuesday, 13 October 2020, 14:46:03 BST, alwyn.seeds1 <a.seeds@...> wrote:
Dear All,
It may be a myth, but a colleague told me that on some instruments the fans had controllers that were designed to avoid microphony problems with the instrument- possible that the HP8640B with its cavity oscillator might be one of those?
Regards,
Alwyn _____________________________________________________
Alwyn Seeds, Director
SynOptika Ltd.,
114 Beaufort Street,
London,
SW3 6BU,
England.
SynOptika Ltd., Registered in England and Wales: No. 04606737
Registered Office: 114 Beaufort Street, London, SW3 6BU, United Kingdom. _____________________________________________________
|
Dear All,
It may be a myth, but a colleague told me that on some instruments the fans had controllers that were designed to avoid microphony problems with the instrument- possible that the HP8640B with its cavity oscillator might be one of those?
Regards,
Alwyn
_____________________________________________________
Alwyn Seeds, Director
SynOptika Ltd.,
114 Beaufort Street,
London,
SW3 6BU,
England.
SynOptika Ltd., Registered in England and Wales: No. 04606737
Registered Office: 114 Beaufort Street, London, SW3 6BU, United Kingdom. _____________________________________________________
|
Re: UNIX files and dd, Re: [HP-Agilent-Keysight-equipment] Agilent 54831M
No need to be so derisive towards windows users.
I'm pretty sure that dd gives final transfer statistics at the
end of a good or bad data transfer. (viz: status=noxfer to suppress
the final transfer statistics, and status=progress to give periodic
transfer statistics updates during the transfer...)
Also, using dd to transfer images to memory cards (CF, SD, thumb drives)
will not always work out like you may think it should. For instance
dd'ing a copy of a bootable CF card onto a data storage type CF card
will not result result in a bootable CF card.
-Chuck Harris
Dave McGuire wrote:
toggle quoted message
Show quoted text
? Another thing is silence in the case of success.? Assume that the computer did what
you told it to do...there is no need for it to say "Hey, you told me to do something,
so I did it!"? That's dumb.? The computer should only say something in an unexpected
situation, i.e. when there was an error.? If you type a command at a UNIX shell and
it responds with another shell prompt, it did what it told you to do.
? And no dancing paperclips.
? Computing for grownups, when they need to get work done.
???????????? -Dave
|
Re: HP HIL keyboard and mouse wanted - in Australia
Hello,
lol - we are working on something similar!
Tam
With best regards
Tam HANNA
Enjoy electronics? Join 15k7 other followers by visiting the Crazy Electronics Lab at
On 2020. 10. 13. 14:57, Anders wrote:
toggle quoted message
Show quoted text
I am working on a PS/2 and/or USB converter to HIL. I have had the
prototype in use with my HP9000/340 since this spring. Contact me
off-list and we shall see what we can do. I guess there is no
shortage of PS/2 or USB keyboards "down under"?
Yes, I developed this because I bought a 9000/340 without keyboard
and spare keyboards were as expensive as the original computer...
|
Some of the 8640Bs had the multi-speed fan and some had single-speed standard square fans. HP changed the design somewhere along the way so there's a mix of both out there.
Thanks,
Barry - N4BUQ
toggle quoted message
Show quoted text
----- Original Message ----- From: "Szabolcs Szigeti" <szigiszabolcs@...>
To: [email protected]
Sent: Tuesday, October 13, 2020 1:42:35 AM
Subject: Re: [HP-Agilent-Keysight-equipment] HP8640B fan
Hi,
As far as I know, the 8640B has a 3 phase motor for the fan, and one of the
power supply boards contains the driver and speed regulator circuits for
it. Unless, of course in your unit it has already been replaced by
something else.
Szabolcs
Henry Newton <EequalsIR@...> ezt ¨ªrta (id?pont: 2020. okt. 12., H,
23:04):
Hello all,
I've just ordered the brass gears from the eBay fellow in India. Many
thanks to all who presented the gear information.
While waiting, I need to replace my noisy fan in my 8640B.
From the red & black leads going to a nearby board, it appears it is a 5V
fan. My fan has a black colored motor.
Any ideas on the CFM? Or a fan source?
Thanks
|
Re: HP HIL keyboard and mouse wanted - in Australia
I am working on a PS/2 and/or USB converter to HIL. I have had the prototype in use with my HP9000/340 since this spring. Contact me off-list and we shall see what we can do. I guess there is no shortage of PS/2 or USB keyboards "down under"?
Yes, I developed this because I bought a 9000/340 without keyboard and spare keyboards were as expensive as the original computer...
|
I'm also not a Linux expert, so I'll let other people speak on that. For the power, I know there is no Molex cable anywhere. They uses industrial PSU and not standard computer one. So of course, you won't have a ready to use cable to plug in any adapter. But, the rear CD-Rom and floppy must have power since it's a 3.5inch IDE cable (without power) that goes to it. I found an old picture of mine where you can see the power cable. It is possible that the last user lost the adapter board to plug the floppy and drive, but the power cable should still be somewhere in the scope. 
|
Hi
Towards the end of production of the HP8640B the fan was changed to a standard two wire one and the power supply board was modified.
George G6HIG Dover UK
On Tuesday, 13 October 2020, 07:43:27 BST, Szabolcs Szigeti <szigiszabolcs@...> wrote:
Hi,
As far as I know, the 8640B has a 3 phase motor for the fan, and one of the power supply boards contains the driver and speed regulator circuits for it. Unless, of course in your unit it has already been replaced by something else.
Szabolcs
Henry Newton < EequalsIR@...> ezt ¨ªrta (id?pont: 2020. okt. 12., H, 23:04):
Hello all,
I've just ordered the brass gears from the eBay fellow in India.? Many thanks to all who presented the gear information.
While waiting, I need to replace my noisy fan in my 8640B.?
From the red & black leads going to a nearby board, it appears it is a 5V fan.? My fan has a black colored motor.
Any ideas on the CFM?? Or a fan source?
Thanks
|
Re: HP-3456A - Question Regarding Resistance Measurement
Hi, may I add one line?
1. get whatever resistors in the 5-10M range you have so they sum up to approx. 100M.
2. solder them all together.
3. measure resistance of each of the resistors and sum up the results.
4. measure resistance of all the resistors connected in series... and tell us what you found out.
G?ran
|
Re: HP-3456A - Question Regarding Resistance Measurement
The following experiment could be interesting:
1. get whatever resistors in the 5-10M range you have so they sum up to approx. 100M.
2. measure resistance of each of the resisitors and sum up the results.
3. measure resistance of all the resistors connected in series... and tell us what you found out.
cheers
Martin
|
Hi,
As far as I know, the 8640B has a 3 phase motor for the fan, and one of the power supply boards contains the driver and speed regulator circuits for it. Unless, of course in your unit it has already been replaced by something else.
Szabolcs
Henry Newton < EequalsIR@...> ezt ¨ªrta (id?pont: 2020. okt. 12., H, 23:04):
Hello all,
I've just ordered the brass gears from the eBay fellow in India.? Many thanks to all who presented the gear information.
While waiting, I need to replace my noisy fan in my 8640B.?
From the red & black leads going to a nearby board, it appears it is a 5V fan.? My fan has a black colored motor.
Any ideas on the CFM?? Or a fan source?
Thanks
|
Re: UNIX files and dd, Re: [HP-Agilent-Keysight-equipment] Agilent 54831M
On 10/13/20 2:20 AM, Manuel Presnitz wrote: I really enjoyed reading your introduction to the Un*x-/dd-World, Dave. I have just two remarks, you may call it nit picking, but maybe worth mentioning for newcomers:
- The dot does have a special meaning in file names, but only at the beginnng of a file name: all files starting with a dot are hidden files, e.g. ".bashrc" Greetings Manuel, I acknowledge your nitpicking and respond with a bigger nitpicking: The leading dot making the file "hidden" is a convention of usage, and is not handled by the filesystem or the core of the OS (the kernel) itself in any way. It is interpreted and treated that way by the "ls" program, which lists a directory. When a program opens a directory (via the OS system calls) and walks down the list of files it contains, they will all be present, with or without leading dots. The "ls" program does this, and looks for the dots, and doesn't print those entries by default. This can be overridden via command line option "-a", for "all files". Nice try, though. ;) - I think the default block size of dd is "1b", which stands for "1 block" and equals to 512 bytes. This hints a the usual usage of dd with socalled block devices like floppy, hard or optical disks. You are absolutely correct about this one, though; the default block size is in fact 512 bytes. I blame the late hour and the soldering fumes, thank you for the correction! I always explicitly specify a blocksize, so I rarely think about it. -Dave -- Dave McGuire, AK4HZ New Kensington, PA
|
Re: UNIX files and dd, Re: [HP-Agilent-Keysight-equipment] Agilent 54831M
I really enjoyed reading your introduction to the Un*x-/dd-World, Dave. I have just two remarks, you may call it nit picking, but maybe worth mentioning for newcomers: - The dot does have a special meaning in file names, but only at the beginnng of a file name: all files starting with a dot are hidden files, e.g. ".bashrc" - I think the default block size of dd is "1b", which stands for "1 block" and equals to 512 bytes. This hints a the usual usage of dd with socalled block devices like floppy, hard or optical disks. Best regards, Manuel, DO1MPY. On Tuesday, 13 October 2020, Dave McGuire wrote: First, it's important not to overcomplicate things. From your
statements I assume you're coming from the Windows world. (...)
|
Re: UNIX files and dd, Re: [HP-Agilent-Keysight-equipment] Agilent 54831M
On 10/13/20 12:44 AM, Sam Reaves wrote: One real word of caution. Make darn sure that you have specified the correct target when using dd to copy files as it is very easy to kill all of the data on the computer hard drive that you are using to do the copies. There is no OOPS command and no warnings about overwriting data. Once the command string is entered and you press enter dd starts the transfer. Excellent advice, and this is something that Windows people might not think of. Central to the UNIX paradigm is the operating system making the assumption that the user actually knows what he/she is doing, and actually means what he/she types, and dutifully and immediately does what the user tells it to do. No "Are you sure?" "Are you REALLY REALLY SURE??" hand-holding. Another thing is silence in the case of success. Assume that the computer did what you told it to do...there is no need for it to say "Hey, you told me to do something, so I did it!" That's dumb. The computer should only say something in an unexpected situation, i.e. when there was an error. If you type a command at a UNIX shell and it responds with another shell prompt, it did what it told you to do. And no dancing paperclips. Computing for grownups, when they need to get work done. -Dave -- Dave McGuire, AK4HZ New Kensington, PA
|
Re: UNIX files and dd, Re: [HP-Agilent-Keysight-equipment] Agilent 54831M
I agree with Dave. The dd program is the bees knees when it comes to byte by byte copies, to and from any source.?
I have used the Windows version of dd to copy files to and from SRAM for installing firmware on older LeCroy scopes. It also comes in handy for making SDCard or PCCard images for industrial machines and the like.
You can get dd for Windows from here: (works just like the Linux version for those that do not have a Linux box.)
The link below is for instructions I wrote years ago for copying binary firmware files to a SRAM care for the LeCroy scopes but it should give you an idea how to use dd. Note that in the case below you do have to have a machine that can install the drivers for SRAM writing under Windows. I think the last version of Windows that can do this is XP but correct me if I am wrong.?
I used to use a program called Memory Card Explorer for doing this but once I discovered dd for free, I now use that. Besides I think MCE won't run on anything newer than XP either. It could be that you could run dd inside a VM under Windows but I don't know if/how that would work. For MCE or dd for SRAM or ATA compliant PCMCIA you need a standard PCMCIA slot not an ExpressCard slot so you have to find an older laptop. I use the Dell D620 or D630 and there are some (but not all) E6400 series that retained that slot.
One real word of caution. Make darn sure that you have specified the correct target when using dd to copy files as it is very easy to kill all of the data on the computer hard drive that you are using to do the copies. There is no OOPS command and no warnings about overwriting data. Once the command string is entered and you press enter dd starts the transfer.
Best to try it out on a test machine that you have a full backup for, or one that you do not care about if you crash it.
73, W3OHM
Sam Reaves
ARS W3OHM
Owner and Moderator of: LeCroy Owners Group on Groups.io (Current and Future Group)
Electronics and Mechanical Hardware Design Engineering Manager
Andritz Rolls Global Research Center (RETIRED)
|
Anders
G?ran Krusell
Martin
Dave McGuire
Sam Reaves