Getting calibration equipment calibrated


Dave Peterson
 

Hi All,

How much does it cost to get PG506s, SG503/SG504s, and TG501s calibrated?

I can certainly search for local calibration facilities, but having no experience at all: do they tend to be amenable to DIYer's, is the cost in the realm of reasonable, is it worth it?

I have a PG506 that's working beautifully, but I can't be certain that the voltage levels are in spec.

The SG503 seems to work, and I can tell pretty well that it's level above 100MHz, but only on an un-calibrated 7854 scope. So while I feel good about the 100MHz scopes I'm testing with it, no one could call that calibrated.

The TG501 is dead, the SG504 blew a fuse when pressed the "ref" button. So those will need some repair. If I can get them working it would be nice to add them to the calibration set.

It'd be really nice to have all four of these officially calibrated. That'd give me assurance that the scopes I'm repairing are properly tuned up. I don't have any means at hand otherwise to verify level or timing. I could keep working on finding means of establishing standards for level and time. But would it just be more cost effective to pay for calibration?

Thanks,
Dave


Eric
 

Dave e mail me off list I might be able to help you out there. The
calibrations are not hard just a process... the gear you need for the
calibrations especially the sg503 and 504 are substantial you will need a 1
ghz + spec analyser. Calibrating the pg is not bad it is done at DC until
you get to the fast rise pulses you will need a fast scope 2 ghz + or a
sampling system. I take it from the list of gear you are looking to do a
scope?

Eric

On Mon, May 24, 2021, 12:06 AM Dave Peterson via groups.io <davidpinsf=
yahoo.com@groups.io> wrote:

Hi All,

How much does it cost to get PG506s, SG503/SG504s, and TG501s calibrated?

I can certainly search for local calibration facilities, but having no
experience at all: do they tend to be amenable to DIYer's, is the cost in
the realm of reasonable, is it worth it?

I have a PG506 that's working beautifully, but I can't be certain that the
voltage levels are in spec.

The SG503 seems to work, and I can tell pretty well that it's level above
100MHz, but only on an un-calibrated 7854 scope. So while I feel good about
the 100MHz scopes I'm testing with it, no one could call that calibrated.

The TG501 is dead, the SG504 blew a fuse when pressed the "ref" button. So
those will need some repair. If I can get them working it would be nice to
add them to the calibration set.

It'd be really nice to have all four of these officially calibrated.
That'd give me assurance that the scopes I'm repairing are properly tuned
up. I don't have any means at hand otherwise to verify level or timing. I
could keep working on finding means of establishing standards for level and
time. But would it just be more cost effective to pay for calibration?

Thanks,
Dave






Sean Turner
 

Dave,

Time standards are easier than ever now, as there are a plethora of GPSDO products around. That will be orders of magnitude better than needed for aligning cathode ray o-scopes. I adjusted the (admittedly crude) Bulova branded OCXO in my Type 184 to many more decimal places than needed for scope calibration. I do leave the 184 plugged in at all times so the oven stays hot.

For DC voltages, you can find things like this: https://www.ianjohnston.com/index.php/onlineshop/handheld-precision-digital-voltage-source-v2-detail

With a few things like that on hand, you can easily know that a CRO is in good shape. Probably not sufficient for more accurate instruments that require incredibly good calibration accuracy (>=6.5 digit meters, spectrum analyzers, etc), but will serve you fine for oscilloscopes.

As for professional cal services, I do know Keysight will work on personally owned stuff (and a lot more old HP stuff than you might think, a lot of that is still in service in pro labs), but get ready to open your wallet. Especially if you want NIST traceable.

Sean


Eric
 

For timing a good (lots of digits) frequency counter hooked to a GPSDO will get you where you need to know then you time reference Is based on the GPS clocks which are HIGHLY accurate. At least WAY more accurate then a 400 series scope at 3%. I did a TG501 and a 184 in my lab with a DMM, Frequency counter, and 100 Mhz scope. There is one adjustment in the 184 that you have to balance 2 trigger points these are the 2 bright spots the should be level. But the SG503 and 504 need spectral purity characterized as well as the SG503 needs that odd 50 Ohm cable. I have to tare in to my 503 it kind of went nuts on me when I was working on a 465. Sorry for the 2 replies I started this on the small screen and moved to the large one.

Eric

-----Original Message-----
From: TekScopes@groups.io <TekScopes@groups.io> On Behalf Of Dave Peterson via groups.io
Sent: Monday, May 24, 2021 12:06 AM
To: TekScopes@groups.io
Subject: [TekScopes] Getting calibration equipment calibrated

Hi All,

How much does it cost to get PG506s, SG503/SG504s, and TG501s calibrated?

I can certainly search for local calibration facilities, but having no experience at all: do they tend to be amenable to DIYer's, is the cost in the realm of reasonable, is it worth it?

I have a PG506 that's working beautifully, but I can't be certain that the voltage levels are in spec.

The SG503 seems to work, and I can tell pretty well that it's level above 100MHz, but only on an un-calibrated 7854 scope. So while I feel good about the 100MHz scopes I'm testing with it, no one could call that calibrated.

The TG501 is dead, the SG504 blew a fuse when pressed the "ref" button. So those will need some repair. If I can get them working it would be nice to add them to the calibration set.

It'd be really nice to have all four of these officially calibrated. That'd give me assurance that the scopes I'm repairing are properly tuned up. I don't have any means at hand otherwise to verify level or timing. I could keep working on finding means of establishing standards for level and time. But would it just be more cost effective to pay for calibration?



Thanks,
Dave


Dave Peterson
 

Hmm,

Seems kind of a mixed bag then. I would love to get my own level and time references setup. I have only superficially heard about using GPS based clocks/timing. Seems this might be my opportunity to learn up on that. And thanks for the level reference link Sean. I'll follow up on that too.

The spectral requirements of the SG503/504 sounds out of my league. When opening up my wallet for a place like Keysight, are we talking 100s? It certainly can't be more than, say, $500 for calibrating one plug-in? Can it? But paying for professional services does get expensive fast.

On Sunday, May 23, 2021, 9:24:27 PM PDT, Eric <ericsp@gmail.com> wrote:

For timing a good (lots of digits) frequency counter hooked to a GPSDO will get you where you need to know then you time reference Is based on the GPS clocks which are HIGHLY accurate. At least WAY more accurate then a 400 series scope at 3%. I did a TG501 and a 184 in my lab with a DMM, Frequency counter, and 100 Mhz scope. There is one adjustment in the 184 that you have to balance 2 trigger points these are the 2 bright spots the should be level.  But the SG503 and 504 need spectral purity characterized as well as the SG503 needs that odd 50 Ohm cable. I have to tare in to my 503 it kind of went nuts on me when I was working on a 465. Sorry for the 2 replies I started this on the small screen and moved to the large one.

Eric

-----Original Message-----
From: TekScopes@groups.io <TekScopes@groups.io> On Behalf Of Dave Peterson via groups.io
Sent: Monday, May 24, 2021 12:06 AM
To: TekScopes@groups.io
Subject: [TekScopes] Getting calibration equipment calibrated

Hi All,

How much does it cost to get PG506s, SG503/SG504s, and TG501s calibrated?

I can certainly search for local calibration facilities, but having no experience at all: do they tend to be amenable to DIYer's, is the cost in the realm of reasonable, is it worth it?

I have a PG506 that's working beautifully, but I can't be certain that the voltage levels are in spec.

The SG503 seems to work, and I can tell pretty well that it's level above 100MHz, but only on an un-calibrated 7854 scope. So while I feel good about the 100MHz scopes I'm testing with it, no one could call that calibrated.

The TG501 is dead, the SG504 blew a fuse when pressed the "ref" button. So those will need some repair. If I can get them working it would be nice to add them to the calibration set.

It'd be really nice to have all four of these officially calibrated. That'd give me assurance that the scopes I'm repairing are properly tuned up. I don't have any means at hand otherwise to verify level or timing. I could keep working on finding means of establishing standards for level and time. But would it just be more cost effective to pay for calibration?



Thanks,
Dave


Mark Goldberg
 

Keysight and Tektronix will calibrate anything for anyone that brings
money. They are pretty expensive though. There are good third party
calibration labs, and bad ones. Search for ones with NVLAP or A2LA
accreditation or maybe ANAB accreditation. Compare their uncertainties
(which they should be able to provide) to the desired accuracy for what you
want to calibrate.

I keep a subset of my test equipment calibrated and compare the rest. I get
one Scope / Spectrum Analyzer and one Fluke multimeter done. For frequency,
GPSDOs are so good I don't see any reason to pay for a frequency
calibration of anything, at least for my level of usage. I'm set up to
measure frequency very accurately based on a GPSDO and use one as a
reference for all my test equipment.

New equipment with computer control have automated calibration routines and
are way cheaper to do. Older equipment with lots of pots to turn take a lot
of labor and you will pay more. My MDO3024 Scope / Spectrum Analyzer cost
about $140 to calibrate and my Fluke multimeter about $40 at a third party
lab. The quote for my HP 8642A signal generator was about $300 (lots of
pots to turn). Tektronix wanted about $500 for the scope calibration. I
think they provide big discounts to good customers, but not to me.

Regards,

Mark

On Sun, May 23, 2021 at 9:16 PM Sean Turner <sdturne@q.com> wrote:

Dave,

Time standards are easier than ever now, as there are a plethora of GPSDO
products around. That will be orders of magnitude better than needed for
aligning cathode ray o-scopes. I adjusted the (admittedly crude) Bulova
branded OCXO in my Type 184 to many more decimal places than needed for
scope calibration. I do leave the 184 plugged in at all times so the oven
stays hot.

For DC voltages, you can find things like this:
https://www.ianjohnston.com/index.php/onlineshop/handheld-precision-digital-voltage-source-v2-detail

With a few things like that on hand, you can easily know that a CRO is in
good shape. Probably not sufficient for more accurate instruments that
require incredibly good calibration accuracy (>=6.5 digit meters, spectrum
analyzers, etc), but will serve you fine for oscilloscopes.

As for professional cal services, I do know Keysight will work on
personally owned stuff (and a lot more old HP stuff than you might think, a
lot of that is still in service in pro labs), but get ready to open your
wallet. Especially if you want NIST traceable.

Sean







Dave Peterson
 

I might be a bit of an odd bird: I quite enjoy twiddling the variety of pots vs. pure automation. I work in the IC industry, so I get my fill of high level integration, and have spent many hours in the lab with high-end scopes. I could quite easily, and may yet, take in some equipment to check in out at high speed. I'll have to talk to the lab manager. This habby came up after the COVID lock-down. Haven't been in the lab since.

Here's a video of our lab showing our first HBM DDR. I did timing closure on the HBM Phy.

https://www.youtube.com/watch?v=_ug1CLendf8&t=39s


It would be worth seeing, for example, the actual edges coming out of the PG506 as captured on a LeCroy.

After so many years stuck on Si die I quite enjoy the tactile nature of the old scopes. I enjoy tearing them down, giving a good cleaning, fixing their little problems, and making them live again. I'm sure my interests will evolve as I progress.

Thanks for the inputs.

On Sunday, May 23, 2021, 09:38:51 PM PDT, Mark Goldberg <marklgoldberg@gmail.com> wrote:

Keysight and Tektronix will calibrate anything for anyone that brings
money. They are pretty expensive though. There are good third party
calibration labs, and bad ones. Search for ones with NVLAP or A2LA
accreditation or maybe ANAB accreditation. Compare their uncertainties
(which they should be able to provide) to the desired accuracy for what you
want to calibrate.

I keep a subset of my test equipment calibrated and compare the rest. I get
one Scope / Spectrum Analyzer and one Fluke multimeter done. For frequency,
GPSDOs are so good I don't see any reason to pay for a frequency
calibration of anything, at least for my level of usage. I'm set up to
measure frequency very accurately based on a GPSDO and use one as a
reference for all my test equipment.

New equipment with computer control have automated calibration routines and
are way cheaper to do. Older equipment with lots of pots to turn take a lot
of labor and you will pay more. My MDO3024 Scope / Spectrum Analyzer cost
about $140 to calibrate and my Fluke multimeter about $40 at a third party
lab. The quote for my HP 8642A signal generator was about $300 (lots of
pots to turn). Tektronix wanted about $500 for the scope calibration. I
think they provide big discounts to good customers, but not to me.

Regards,

Mark

On Sun, May 23, 2021 at 9:16 PM Sean Turner <sdturne@q.com> wrote:

Dave,

Time standards are easier than ever now, as there are a plethora of GPSDO
products around. That will be orders of magnitude better than needed for
aligning cathode ray o-scopes. I adjusted the (admittedly crude) Bulova
branded OCXO in my Type 184 to many more decimal places than needed for
scope calibration. I do leave the 184 plugged in at all times so the oven
stays hot.

For DC voltages, you can find things like this:
https://www.ianjohnston.com/index.php/onlineshop/handheld-precision-digital-voltage-source-v2-detail

With a few things like that on hand, you can easily know that a CRO is in
good shape. Probably not sufficient for more accurate instruments that
require incredibly good calibration accuracy (>=6.5 digit meters, spectrum
analyzers, etc), but will serve you fine for oscilloscopes.

As for professional cal services, I do know Keysight will work on
personally owned stuff (and a lot more old HP stuff than you might think, a
lot of that is still in service in pro labs), but get ready to open your
wallet. Especially if you want NIST traceable.

Sean







Eric
 

Hi Dave,
Ok that changes a LOT of things. If you have access to that lecroy you should be set to do the calibrations on all the plugins. A 6.5 Digit DMM is plenty for the PG 506 did mine on a keithly 6500. What is important is you know that when you set the controls of the plugins they are outputting that you think they are and have confidence in them. Especially If the LeCroy has FFT functions you can actually get rid of the spec an requirements cause the LeCroy can check the spectral purity of the 503 and the 504. During the cal process that is really just to make sure the harmonics and spurious signals are at an acceptable level. Even if you can get all the way there you will have them checked to the limits of the LeCroy which means you will be good to go for a 400 and even 7000 series scope. Vertical accuracy of a 465 Is only +/- 3% when the PG is checked with a DMM at DC and adjusted, You run at 0.0025% but then you know you are good to tune 3%. If that makes sense. I can take a look at the spec sheet of the LeCroy if you would like but I will need to model number. I expect you will be perfect for the calibrations the only question is can it FFT this might be a "software option" and not base functionality. The fastest a PG506 needs for calibration is to check a 1ns rise time so only 350Mhz I would imagine the LeCroy is 1Ghz+ so doing a final check on the 3 plugins should be well in it wheel house.

Eric

-----Original Message-----
From: TekScopes@groups.io <TekScopes@groups.io> On Behalf Of Dave Peterson via groups.io
Sent: Monday, May 24, 2021 1:18 AM
To: TekScopes@groups.io
Subject: Re: [TekScopes] Getting calibration equipment calibrated

I might be a bit of an odd bird: I quite enjoy twiddling the variety of pots vs. pure automation. I work in the IC industry, so I get my fill of high level integration, and have spent many hours in the lab with high-end scopes. I could quite easily, and may yet, take in some equipment to check in out at high speed. I'll have to talk to the lab manager. This habby came up after the COVID lock-down. Haven't been in the lab since.

Here's a video of our lab showing our first HBM DDR. I did timing closure on the HBM Phy.

https://www.youtube.com/watch?v=_ug1CLendf8&t=39s


It would be worth seeing, for example, the actual edges coming out of the PG506 as captured on a LeCroy.

After so many years stuck on Si die I quite enjoy the tactile nature of the old scopes. I enjoy tearing them down, giving a good cleaning, fixing their little problems, and making them live again. I'm sure my interests will evolve as I progress.

Thanks for the inputs.


On Sunday, May 23, 2021, 09:38:51 PM PDT, Mark Goldberg <marklgoldberg@gmail.com> wrote:

Keysight and Tektronix will calibrate anything for anyone that brings money. They are pretty expensive though. There are good third party calibration labs, and bad ones. Search for ones with NVLAP or A2LA accreditation or maybe ANAB accreditation. Compare their uncertainties (which they should be able to provide) to the desired accuracy for what you want to calibrate.

I keep a subset of my test equipment calibrated and compare the rest. I get one Scope / Spectrum Analyzer and one Fluke multimeter done. For frequency, GPSDOs are so good I don't see any reason to pay for a frequency calibration of anything, at least for my level of usage. I'm set up to measure frequency very accurately based on a GPSDO and use one as a reference for all my test equipment.

New equipment with computer control have automated calibration routines and are way cheaper to do. Older equipment with lots of pots to turn take a lot of labor and you will pay more. My MDO3024 Scope / Spectrum Analyzer cost about $140 to calibrate and my Fluke multimeter about $40 at a third party lab. The quote for my HP 8642A signal generator was about $300 (lots of pots to turn). Tektronix wanted about $500 for the scope calibration. I think they provide big discounts to good customers, but not to me.

Regards,

Mark

On Sun, May 23, 2021 at 9:16 PM Sean Turner <sdturne@q.com> wrote:

Dave,

Time standards are easier than ever now, as there are a plethora of
GPSDO products around. That will be orders of magnitude better than
needed for aligning cathode ray o-scopes. I adjusted the (admittedly
crude) Bulova branded OCXO in my Type 184 to many more decimal places
than needed for scope calibration. I do leave the 184 plugged in at
all times so the oven stays hot.

For DC voltages, you can find things like this:
https://www.ianjohnston.com/index.php/onlineshop/handheld-precision-di
gital-voltage-source-v2-detail

With a few things like that on hand, you can easily know that a CRO is
in good shape. Probably not sufficient for more accurate instruments
that require incredibly good calibration accuracy (>=6.5 digit meters,
spectrum analyzers, etc), but will serve you fine for oscilloscopes.

As for professional cal services, I do know Keysight will work on
personally owned stuff (and a lot more old HP stuff than you might
think, a lot of that is still in service in pro labs), but get ready
to open your wallet. Especially if you want NIST traceable.

Sean







 

Outside calibration is complex situation, especially since it is understood differently by the
providers of the service, and their customers. this disconnect can be severe, and often leads to
some significant unhappiness on the part of equipment owners.

I documents all olf this in detail when I was writing for the AEA (Aircraft Electronics Association),
you can grab those articles HERE on our site:
https://www.sphere.bc.ca/test/data.html

look halfway down the left hand yellow area where you see the AEA logo. "understanding calibration"
be aware that you will likely get your gear back just as you sent it, with no improvement at all.

I love metrology, and the whole idea of chasing precision to wind up with things as accurate as possible.
I have a GPS, three rubidium standards, and three high end OCXOs I cross check just to get a clean
known 10.000000Mhz value for doing counter calibration. This path is NOT for everyone, and no single standard,
without regular external validation is of any value in the calibration world. Every lab should have some known
references as a sanity check, even if it's just a few precision resistors and known frequency. Known DC voltage
can be harder, and known AC voltage harder still.

One of my best friends runs a metrology lab here, and I watch what he has to go through each year to
have traceable accredited standards, it is not for the faint of heart. I am eternally grateful he helps me
with my cals, or I could never really be sure of anything.

all the best,
walter

--
Walter Shawlee 2
Sphere Research Corp. 3394 Sunnyside Rd.
West Kelowna, BC, V1Z 2V4 CANADA
Phone: +1 (250-769-1834 -:- http://www.sphere.bc.ca
+We're all in one boat, no matter how it looks to you. (WS2)
+All you need is love. (John Lennon)
+But, that doesn't mean other things don't come in handy. (WS2)
+Nature is trying very hard to make us succeed, but nature does not depend on us.
We are not the only experiment. (R. Buckminster Fuller)


Harvey White
 

There's a lot of stuff you don't need to do unless you want to.

Remember that the analog scopes you're checking, unless they have digital counters (7D15, et al), you're deciding whether or not you read to the edge of the trace, the middle of the trace, or the other edge of the trace.  Check the specs on the scopes to see what accuracy you *can* read them to.  3%?  You don't need a six figures accurate counter or DMM to set that.

For frequency, you can easily take any available counter and check it to a packaged oscillator.  Then find one that's calibrated (you do have access to that) and see what it reads on that one.  Or if you don't, you're getting close.  Ditto with the voltmeters.

Rise time, well, that is a possible problem, and if you're going after that, you do need bandwidth.

Harmonic Distortion?  Either a good distortion meter at that frequency, or a spectrum analyzer, or a scope with an FFT.

Flatness?  bring on the high bandwidth scope.

However, how much do you need those particular measurements? Most people calibrate the timebases, vertical and horizontal sensitivity, look at the risetime, look at the bandwidth, and let it go at that.

For digital scopes, I'm not even sure you *can* adjust any of the timing.

Harvey

On 5/24/2021 12:31 AM, Dave Peterson via groups.io wrote:
Hmm,

Seems kind of a mixed bag then. I would love to get my own level and time references setup. I have only superficially heard about using GPS based clocks/timing. Seems this might be my opportunity to learn up on that. And thanks for the level reference link Sean. I'll follow up on that too.

The spectral requirements of the SG503/504 sounds out of my league. When opening up my wallet for a place like Keysight, are we talking 100s? It certainly can't be more than, say, $500 for calibrating one plug-in? Can it? But paying for professional services does get expensive fast.



On Sunday, May 23, 2021, 9:24:27 PM PDT, Eric <ericsp@gmail.com> wrote:
For timing a good (lots of digits) frequency counter hooked to a GPSDO will get you where you need to know then you time reference Is based on the GPS clocks which are HIGHLY accurate. At least WAY more accurate then a 400 series scope at 3%. I did a TG501 and a 184 in my lab with a DMM, Frequency counter, and 100 Mhz scope. There is one adjustment in the 184 that you have to balance 2 trigger points these are the 2 bright spots the should be level.  But the SG503 and 504 need spectral purity characterized as well as the SG503 needs that odd 50 Ohm cable. I have to tare in to my 503 it kind of went nuts on me when I was working on a 465. Sorry for the 2 replies I started this on the small screen and moved to the large one.

Eric

-----Original Message-----
From: TekScopes@groups.io <TekScopes@groups.io> On Behalf Of Dave Peterson via groups.io
Sent: Monday, May 24, 2021 12:06 AM
To: TekScopes@groups.io
Subject: [TekScopes] Getting calibration equipment calibrated

Hi All,

How much does it cost to get PG506s, SG503/SG504s, and TG501s calibrated?

I can certainly search for local calibration facilities, but having no experience at all: do they tend to be amenable to DIYer's, is the cost in the realm of reasonable, is it worth it?

I have a PG506 that's working beautifully, but I can't be certain that the voltage levels are in spec.

The SG503 seems to work, and I can tell pretty well that it's level above 100MHz, but only on an un-calibrated 7854 scope. So while I feel good about the 100MHz scopes I'm testing with it, no one could call that calibrated.

The TG501 is dead, the SG504 blew a fuse when pressed the "ref" button. So those will need some repair. If I can get them working it would be nice to add them to the calibration set.

It'd be really nice to have all four of these officially calibrated. That'd give me assurance that the scopes I'm repairing are properly tuned up. I don't have any means at hand otherwise to verify level or timing. I could keep working on finding means of establishing standards for level and time. But would it just be more cost effective to pay for calibration?



Thanks,
Dave
















Stephen Hanselman
 

Walter,

I second your comments. We had a “cal-light” lab. Two GPSDO’s and a HP3458 and a set of Julie labs precision resistors. The 3458 got the stds lab cal at keysight ($1800) and then everything was aligned to that unit.

Our customers did not consider us a cal lab but a repair station. They were required to use specific labs which caused problems. We received MANY pieces of equipment that “didn’t pass cal” that merely required doing the entire calibration procedure carefully. Even with our loose definition of “cal to zero” we never had a unit rejected by the same lab that failed it originally.

Be careful with who you choose for your cal facility, there are a few good ones and many “Lick and Stick” operations.

Regards,

Stephen Hanselman
Datagate Systems, LLC

On May 24, 2021, at 07:58, walter shawlee <walter2@sphere.bc.ca> wrote:

Outside calibration is complex situation, especially since it is understood differently by the
providers of the service, and their customers. this disconnect can be severe, and often leads to
some significant unhappiness on the part of equipment owners.

I documents all olf this in detail when I was writing for the AEA (Aircraft Electronics Association),
you can grab those articles HERE on our site:
https://www.sphere.bc.ca/test/data.html

look halfway down the left hand yellow area where you see the AEA logo. "understanding calibration"
be aware that you will likely get your gear back just as you sent it, with no improvement at all.

I love metrology, and the whole idea of chasing precision to wind up with things as accurate as possible.
I have a GPS, three rubidium standards, and three high end OCXOs I cross check just to get a clean
known 10.000000Mhz value for doing counter calibration. This path is NOT for everyone, and no single standard,
without regular external validation is of any value in the calibration world. Every lab should have some known
references as a sanity check, even if it's just a few precision resistors and known frequency. Known DC voltage
can be harder, and known AC voltage harder still.

One of my best friends runs a metrology lab here, and I watch what he has to go through each year to
have traceable accredited standards, it is not for the faint of heart. I am eternally grateful he helps me
with my cals, or I could never really be sure of anything.

all the best,
walter

--
Walter Shawlee 2
Sphere Research Corp. 3394 Sunnyside Rd.
West Kelowna, BC, V1Z 2V4 CANADA
Phone: +1 (250-769-1834 -:- http://www.sphere.bc.ca
+We're all in one boat, no matter how it looks to you. (WS2)
+All you need is love. (John Lennon)
+But, that doesn't mean other things don't come in handy. (WS2)
+Nature is trying very hard to make us succeed, but nature does not depend on us.
We are not the only experiment. (R. Buckminster Fuller)






 

Dave,

for a voltage reference you can get something quite economically (less than $25) on Amazon or eBay that is based off an AD584:

https://www.amazon.com/AD584KH-Precision-4-Channel-Voltage-Reference/dp/B07SR9KTV6/ref=sr_1_1_sspa

https://www.ebay.com/itm/254166950181

This should let you get a good bead on your meters. Of course, if you have more than a couple meters you can just take the average of their readings and call it good enough for government work (or for DIY calibration).

They say that the man with several multimeters never knows the correct value of a measurement, but you can just average them all together and get a pretty good bracket. Also, if you are measuring a "known value" (e.g. the voltage on a new battery) you can get a good idea of 1) how close your meters are to true, 2) the variance is across your meters, and 3) which meters read low or high and by how much.

Frequency standards can be had almost as cheaply:

https://www.amazon.com/WSDMAVIS-Frequency-Reference-Oscillator-Interface/dp/B08R6VQRC3/ref=sr_1_2

everything I saw on eBay was more expensive, but also looked like it was higher quality:

https://www.ebay.com/itm/383586682637

or you can buy a brand new GPSDO from Leo Bodnar for $140 + s&h:

http://www.leobodnar.com/shop/index.php?main_page=product_info&cPath=107&products_id=301

but I know that you already have a tinySA, which will probably serve as a frequency reference too (maybe also as a voltage reference?). Throw in a couple of random TCXOs, average across several different frequency counters, and again, you're good enough for government work.

-- Jeff Dutky


Dave Daniel
 

I use one of these. And one can send it back for recalibration.

https://dmmcheckplus.com/

DaveD

On May 24, 2021, at 12:22, Jeff Dutky <jeff.dutky@gmail.com> wrote:

Dave,

for a voltage reference you can get something quite economically (less than $25) on Amazon or eBay that is based off an AD584:

https://www.amazon.com/AD584KH-Precision-4-Channel-Voltage-Reference/dp/B07SR9KTV6/ref=sr_1_1_sspa

https://www.ebay.com/itm/254166950181

This should let you get a good bead on your meters. Of course, if you have more than a couple meters you can just take the average of their readings and call it good enough for government work (or for DIY calibration).

They say that the man with several multimeters never knows the correct value of a measurement, but you can just average them all together and get a pretty good bracket. Also, if you are measuring a "known value" (e.g. the voltage on a new battery) you can get a good idea of 1) how close your meters are to true, 2) the variance is across your meters, and 3) which meters read low or high and by how much.

Frequency standards can be had almost as cheaply:

https://www.amazon.com/WSDMAVIS-Frequency-Reference-Oscillator-Interface/dp/B08R6VQRC3/ref=sr_1_2

everything I saw on eBay was more expensive, but also looked like it was higher quality:

https://www.ebay.com/itm/383586682637

or you can buy a brand new GPSDO from Leo Bodnar for $140 + s&h:

http://www.leobodnar.com/shop/index.php?main_page=product_info&cPath=107&products_id=301

but I know that you already have a tinySA, which will probably serve as a frequency reference too (maybe also as a voltage reference?). Throw in a couple of random TCXOs, average across several different frequency counters, and again, you're good enough for government work.

-- Jeff Dutky





Dave Peterson
 

For fun last night I characterized my bench DMM with my PG506 in DC mode. I was able to use the deflection error to determine that the entirety of the DMMs range held to within 0.2%. Up to the 100v output of the PG506.


So, having compared my oranges to be the same, I am now ready to make apple pie!

I'm at least getting lots of great references to research on my own. I think the conclusion is, for what I'm doing, no, paying a cal shop isn't really worth the money. I'm not splitting atoms here. I should be able to at least say I'm using crab apples to make apple pie.

On Monday, May 24, 2021, 9:41:00 AM PDT, Dave Daniel <kc0wjn@gmail.com> wrote:

I use one of these. And one can send it back for recalibration.

https://dmmcheckplus.com/

DaveD


On May 24, 2021, at 12:22, Jeff Dutky <jeff.dutky@gmail.com> wrote:

Dave,

for a voltage reference you can get something quite economically (less than $25) on Amazon or eBay that is based off  an AD584:

https://www.amazon.com/AD584KH-Precision-4-Channel-Voltage-Reference/dp/B07SR9KTV6/ref=sr_1_1_sspa

https://www.ebay.com/itm/254166950181

This should let you get a good bead on your meters. Of course, if you have more than a couple meters you can just take the average of their readings and call it good enough for government work (or for DIY calibration).

They say that the man with several multimeters never knows the correct value of a measurement, but you can just average them all together and get a pretty good bracket. Also, if you are measuring a "known value" (e.g. the voltage on a new battery) you can get a good idea of 1) how close your meters are to true, 2) the variance is across your meters, and 3) which meters read low or high and by how much.

Frequency standards can be had almost as cheaply:

https://www.amazon.com/WSDMAVIS-Frequency-Reference-Oscillator-Interface/dp/B08R6VQRC3/ref=sr_1_2

everything I saw on eBay was more expensive, but also looked like it was higher quality:

https://www.ebay.com/itm/383586682637

or you can buy a brand new GPSDO from Leo Bodnar for $140 + s&h:

http://www.leobodnar.com/shop/index.php?main_page=product_info&cPath=107&products_id=301

but I know that you already have a tinySA, which will probably serve as a frequency reference too (maybe also as a voltage reference?). Throw in a couple of random TCXOs, average across several different frequency counters, and again, you're good enough for government work.

-- Jeff Dutky





 

Crab Apples can actually make some damn good apple pie. As with cherry pie, you really want something very sour for the filling: gives it better flavor.

-- Jeff Dutky (who grew up on 2 acres with 5 apple trees, 3 sour cherry trees, and all the pie you could eat)


Ken Eckert
 

A really handy bench reference is the Fluke 515A, does DC, AC, and Ohms. No
current though.

On Monday, May 24, 2021, Jeff Dutky <jeff.dutky@gmail.com> wrote:

Crab Apples can actually make some damn good apple pie. As with cherry
pie, you really want something very sour for the filling: gives it better
flavor.

-- Jeff Dutky (who grew up on 2 acres with 5 apple trees, 3 sour cherry
trees, and all the pie you could eat)






Dave Daniel
 

Where does one get a Fluke 515A calibrated, and at what cost?

DaveD

On May 24, 2021, at 14:19, Ken Eckert <eckertkp@gmail.com> wrote:

A really handy bench reference is the Fluke 515A, does DC, AC, and Ohms. No
current though.

On Monday, May 24, 2021, Jeff Dutky <jeff.dutky@gmail.com> wrote:

Crab Apples can actually make some damn good apple pie. As with cherry
pie, you really want something very sour for the filling: gives it better
flavor.

-- Jeff Dutky (who grew up on 2 acres with 5 apple trees, 3 sour cherry
trees, and all the pie you could eat)









 

I agree, a fluke 515A is an excellent calibrator (if working and calibrated) for any home lab, maybe ditch the batteries, they are nothing but trouble. They show up regularly, but again, the issue is how do you know it's still good, especially if you buy it used? Being defective is the usual reason they get sold. Others have mentioned some very good DC references that can be used to cross-check it that are not expensive, so you can get to the goal line if you are careful. Also, make friends with somebody close by that has a better bit of gear than you do so you can do a quick check or cal if needed. Outside validation is essential.

I use a regularly cal'd Fluke 5100B to do my DMMs and similar items (I had several 515As, but got rid of them for this), plus I cross-check to a cal'd hp 34401A (I just can't afford a 3458A, but it is the magic weapon if you can afford to cal it!). I also built a super stable 10.0000VDC reference I use as a secondary cross-check, along with a selected super stable 10k resistor.

all the best,
walter (walter2 -at- sphere.bc.ca)
sphere research corp.


Stephen Hanselman
 

5 standard cells, I think Leeds and Northrop did a set in a very nice Oak box.

Seriously my final standard was a a 3458A with a 3245A, what a set of equipment.

Regards,

Stephen Hanselman
Datagate Systems, LLC

On May 24, 2021, at 14:34, walter shawlee <walter2@sphere.bc.ca> wrote:

I agree, a fluke 515A is an excellent calibrator (if working and calibrated) for any home lab, maybe ditch the batteries, they are nothing but trouble. They show up regularly, but again, the issue is how do you know it's still good, especially if you buy it used? Being defective is the usual reason they get sold. Others have mentioned some very good DC references that can be used to cross-check it that are not expensive, so you can get to the goal line if you are careful. Also, make friends with somebody close by that has a better bit of gear than you do so you can do a quick check or cal if needed. Outside validation is essential.

I use a regularly cal'd Fluke 5100B to do my DMMs and similar items (I had several 515As, but got rid of them for this), plus I cross-check to a cal'd hp 34401A (I just can't afford a 3458A, but it is the magic weapon if you can afford to cal it!). I also built a super stable 10.0000VDC reference I use as a secondary cross-check, along with a selected super stable 10k resistor.

all the best,
walter (walter2 -at- sphere.bc.ca)
sphere research corp.







garp66
 

Of course, primary calibration labs at NIST (US), NPL (UK) and NRC (Canada) all function nicely.

The Canadian NRC Metrology labs may have a faster turn-around time and somewhat of a lower cost( CDN$ to US$ exchange rate ~0.8x) , depending on the
calibrations required, and are fully equivalent to NIST...

https://nrc.canada.ca/en/research-development/research-collaboration/research-centres/metrology-research-centre

fun fun fun !