Finally 7.5 Digit - My "new" Keithley DMM7510

[kmoonwalker]

also adjustment of such tiny values is often giving not very good results due to fact that you need to keep really constant temperature and this usually gives +/- ppms here and there. I remember that during my practice time in nearby metrology company they gaved me a job to test and cal some of things they manufactured and it was crazy how presumably good adjustment for more accuracy in T=5C was "killing" pretty good results in over 30C test later (It was a specialized carry on tester in a case to use outside all year round).

[Martin72]

That's why temperature is your biggest “enemy,” especially in the ppm range.

[Scottjd]

Second time is a charm, I’m happy…..
So I went with someone that has a stock of these meters in the US. After some readership on his company I decided to contact him and make the purchase. Price was negotiated and although a little higher than what I originally paid with import tax/tariffs it was only another $300, but well worth the price.
This was a DMM-7510 RACK unit, and looks like it spent its life racked from only the few scratches on the top rear. Although it’s a RACK system the seller has included new bumpers, handle, quick start guide, USB cable and probes. It also arrived with a power cord.
It was foam injected packed, very well protected for shipping and in near new condition. He did mention they are from Europe, not sure exactly were in Europe but the time was +7.5 from my time zone and they were acquired before my day light savings change last week. So I’m sure I could figure it out, but wasn’t that interested, well not yet.

He mentioned me has a lot more, if your unstressed DM me I can provide this contact information. He has no objections with me giving his information out, but I just don’t know was the forum policy is on listed emails and phone number, that and he would probably get spammed by bots scrapping the information.
 
It has 4 self calibrations notes and one calibration adjustment that was done by KeySight, I know right. The KeySight calibration was done December 22, 2021, so about 8 to 9 months after it was put into operation. The last self calibration was on the same day as the adjusted calibration. It was left in TSP language and not SCPI, that makes sense since to make adjustment for the calibration it need to be in TSP.

It’s very clean, like clean room clean.
And seems to be in spec with everything I tested so far.
After warmup everything seems to be functioning and in spec.
Resistance down to 1mΩ, 10mΩ and 100mΩ spot on (using 4 wire current sensing precision resistors) and all the way up through every range to 1GΩ with 2 wire and 4 wire, both are good.
DC Voltage, dead on. Surprisingly dead on.
I still need to set up my rig again for low current testing, but I did test the 3A and 10A ranges, front and read jacks and all was good.

Checking against the PDVS 2 Mini V2 gave me great results. Also checked out perfect with the DMMCheck Plus r8.
PDVS calibrated at 29.9C. The unit internal temp was at 30.9C when this test was ran.
Ambient room temperature for PDVS calibration was 23.7C, my ambient temperature was at 24.44C
With this picture, I think it wavered the last two digits from 59 to 62 and I watched it for 10 minutes.

[TheDefpom]

Well, that looks like it turned out very well in the end, you got yourself a great meter by the looks of it, you must be thrilled!

My 7510 was out about by 200uv at 10V (not calibrated since 2017 when it left the factory, it had never even had an ACAL done !), I spent this morning improving that situation.

Something I did find was that I might have a relay playing up, as I could get slightly different voltage readings when I first power up, and if I then switch to frequency and then back to DC it is about 100 counts higher, before switching the voltage would drift a bit too, and after switch it did not, so that does lead me to suspect a relay contact is not in the best shape.. I am going to need to keep an eye on that in case that is something i need to dig into further.

I fired up my Fluke 731B reference, PDVS2mini V1, Datron 4700 calibrator, Fluke 5450 etc yesterday to give them plenty of warm up time, and spent this morning cross checking and verifying that everything made sense before adjusting the DCV mode of the 7510, part of the DC adjustment is a 10K resistor, and so I cross checked the 10.000140K in the Datron calibrator against a 10.00054K in the Fluke 5450A, and a 9.999852K standard resistor on my Advantest R6581T, again a sanity check before doing that DC adjustment, DCV and resistance is now bang on, I also briefly checked the lower voltage ranges by using my Fluke 720A to scale the 10V down to 1V, 100mV, 10mV etc and those all checked out too.

It is very easy to adjust the DMM7510 yourself if you have suitable references AND sufficient confidence in your references, I was hesitant to adjust my unit in case it was more correct than all of my other references, but after spending a few hours cross checking everything I was confident that it was in need of adjustment.

You can do the adjustment from the web interface by sending the TSP commands from there, you don't need any special software.

I also had a look at the AC mode but have not adjusted it as I can tell that my 4700 calibrator is not flat enough across the frequencies between 100Hz and 50KHz, I verified this using my Fluke 540B thermal transfer standard, that is something I should look at at some point, of course using the 540B would have allowed me to correctly level the calibrator output across the frequencies and voltages to get it correct, but it is only out by about 1mv across the entire frequency range at 10V,

[Scottjd]

Well, that looks like it turned out very well in the end, you got yourself a great meter by the looks of it, you must be thrilled!

I am thrilled. It will be so nice another having to grab another meter for miliohm measurements and just use one instrument.
If the current side checks out (I suspect it will) then I can retire the Keithley 480 PicoAmmeter. For something form 1978 its suited my needs but you can’t breath or move when taking a measurement. It’s just a little sensitive. The extra money in cost ended up being the same cost as if I was to have it calibrated with a local lab in my area. So I guess it all the same.

My 7510 was out about by 200uv at 10V (not calibrated since 2017 when it left the factory, it had never even had an ACAL done !), I spent this morning improving that situation.

I was wondering about your meter.  After all you got me started on this mission with your unboxing video. And then I blame Tony for the other videos I watched as I followed up after yours.
I did make sure to check that the cal password was not changed. Was getting one error of a stale temp, but then realized it was from Node 2. They must have had them clustered on a network. So a reset and firmware update fixed all that.

Calibrating the 10V range isn’t bad from what I was reading in the manual, it even suggests using a different low drift temp resistor if available since the fluke calibrator they list is typically lower then 10K.

So I am curious, did you try the auto cal first before manually making the adjustments?
I’ve left mine for now, it’s only ready 0.2C difference from the last location it had the auto cal done.
 

[TheDefpom]

Yes I had done the ACAL which did initially get the values closer but didn’t solve the aging shift.

[exe]

I decided not to calibrate my unit for now. There are several reasons:

- I'm going to replace fan, after which I need to verify performance. It doesn't make sense to calibrate before that
- I want an adjustment, and I don't want to pay twice (calibration -> adjustment -> another calibration)
- I don't really need precision, I only need stable readings and good linearity.
- it's getting cold and I don't want to thermally shock my unit.

Anyway, my calibration adventure is this. So far found only two companies that were willing to work with me: fluke and messtechniknord.de. The latter didn't reply when I asked if they can adjust first and then only do DC-calibration. I wanted only DC calibration as I don't use other functions. They didn't answer. The price of calibration was 421euro + shipping.

Fluke said they cannot do calibration in NL (but they could do it for my K2000), and they would send the unit to Cologne, Germany. The price is 568euro.

TBH, I'd just send my resistor decade box, asked to calibrate it at 20C, and that would solve all my issues.

[exe]

Guys, just a little update, you might be interested. I had a ~5uV difference between +10V and -10V ranges. I replaced the fan with a different one, and the difference has gone. The original fan on my unit was very electrically noisy, I suspect that could influence readings. Also, voltage readings of my TEN399 are within ~1uV with another DMM7510 it was calibrated against.

PS messtechniknord.de replied to me. They said that doing partial calibration is potentially even more expensive as they'd need to make a custom protocol for me. I think I'll pass.

[ITArchitect]

What fan did you use and did it help with the noise?

[exe]

What fan did you use and did it help with the noise?

Please follow this thread to see pros, cons and part numbers used: https://www.eevblog.com/forum/testgear/all-about-keithley-dmm7510-bugs-and-features-recipes-advice-notes/msg5930842/#msg5930842

BTW, could you guys please report what print(localnode.internaltemp) (TSP) or :SYSTem:TEMPerature:INTernal? says? After full warmup (90mins), along with ambient temperature. I need this to understand efficiency of my fan. My numbers are: internal temp 35.9C, room temp ~20.1C. This is 15.5C temperature rise on my unit. I expect stock fan to be more efficient, with rise I estimate about 11-12C. It's important to wait at least whole 90 mins as temperature rises very slowly.

[jjoonathan]

My vanilla-bean never been modified DMM7510 that has been running for the last 7 hours doing nothing but an occasional voltage or continuity reading:

Code: [Select]

print(localnode.internaltemp)
39.344598732

[Martin72]

PS messtechniknord.de replied to me. They said that doing partial calibration is potentially even more expensive as they'd need to make a custom protocol for me. I think I'll pass.

The idea is nice, that if you only have part of it calibrated, you only pay for that.
However, I can well imagine that it could even be more expensive, as you can't measure automatically, but only by hand.
I didn't think the cost of a full calibration including a report was too expensive; at least it was significantly cheaper than Tek.
And with the new calibration, I had paid a fraction of what the meter would have cost me new.
But everyone has to decide for themselves what is worth what to them.

[KungFuJosh]

In case any of you new DMM7510 owners want to improve the continuity test, you can download the script here: https://www.eevblog.com/forum/testgear/new-keithley-dmm6500/msg5893332/#msg5893332

Thanks,
Josh

[exe]

My vanilla-bean never been modified DMM7510 that has been running for the last 7 hours doing nothing but an occasional voltage or continuity reading:

Code: [Select]

print(localnode.internaltemp)
39.344598732

Thanks! What's the ambient temperature? My guess is ~28.4C 

I didn't think the cost of a full calibration including a report was too expensive; at least it was significantly cheaper than Tek.

Absolutely, I'll go with the when I need it. So far the best price. Sorry if it appeared I'm bashing them, I'm not.


PS just in case, after getting internal temperature, the device enters "remote command" mode. To quit remote command mode one can use command "logout". To enable continuous measurement again, one can press "trigger" button for 2 seconds, then choose "continuous measurement".

[alex-kv]

A bit late, but I had my 7510 calibrated in October 2024 for €410 (excluding VAT) at https://1acal.com in Germany. Getting a quotation is easy via the form available under the “Cal-Request” tab.

[Itanium7000]

It’s exciting to come across this topic here, so I registered an account just to post this. I recently bought a used DMM7510 that was verified with a Fluke 732B and Fluke 5720A. Everything seems fine and within specs; the 10V DC range is even at a remarkable 1.5ppm.
However, there is a terrible flaw in its keypad design. Over time, most of the buttons became paralyzed, with only the Home and Menu buttons still working, and even then, they required significant force. I decided to take it apart for repair—it was a bit risky, but I had no other choice. After half a day of hard work, I finally succeeded!

[Martin72]

Congratulations—and how did you do it?

[jjoonathan]

Ooo, thanks for the high res photos! I made a video of this process a while back and I also included a cutout template if you want to make a perfect adhesive layer on a vinyl cutter (but don't worry, if you just slap some double sided tape on there and call it a day I won't tell).



I just scrubbed the surfaces and after a couple of years the buttons are misbehaving again, so I ordered some replacement conductive rubber pads but I haven't actually made the time to go and swap them out.

[BillyO]

There still seem to be piles of these available and going for as little as $1,600 shipped.

I'm trying to control myself...

https://www.ebay.ca/itm/127571831035

[Martin72]

Ooohhhh....
Unfortunately, I need every penny for my move at the moment, otherwise I would definitely give in again.

Actually, you should buy two, then wait until the “glut” from China subsides.
Then offer one of them for sale, and you're sure to recoup the price for both.

[Itanium7000]

Oh, if only I had seen your video sooner, it would have saved me so much trouble! I actually redrew the adhesive profile myself, bought some 3M 300LSE double-sided tape, printed the template onto it, and then hand-cut it using a carving knife.

However, I must say the most grueling part was separating the screen/keypad panel from the PCB and cleaning off the old adhesive residue. After that, I used Isopropyl Alcohol (IPA) to clean the conductive pads under the buttons and the contact points on the PCB. It’s a bit discouraging to think that I might have to do this all over again in the future, because with this design, the buttons are bound to fail again sooner or later.

Before starting the repair, I bought a sheet of conductive pads intending to stick them onto the buttons. But once I got it open, I realized each button has two contact points, and their diameter is quite small—only about 2mm. The pads I bought were 6mm in diameter. Since everything seemed to work fine after cleaning, I decided not to use them, as I would have had to trim them down to 2mm. If the keypad fails again in the future, I’ll probably have to go through with applying those conductive pads.
I really wonder why Keithley doesn't release a firmware update to allow full control via the touchscreen as an alternative to the physical keys. That way, we wouldn't even need to repair the keypad when it fails, avoiding the risk of damaging the panel during disassembly.

[KungFuJosh]

Hmmm. I feel like I should make some replacement adhesive cutouts with my laser and sell them for $1M on feeBay.

[Scottjd]

Hmmm. I feel like I should make some replacement adhesive cutouts with my laser and sell them for $1M on feeBay.

I was all on until I saw that price 😂

[jjoonathan]

Go ahead! Anyone and everyone who is willing to hold inventory and handle shipments has my permission to use the template.

https://www.dropbox.com/s/g3gghe2odsm8igm/dmm7510%20adhesive%20outline.svg?dl=0

I thought about doing it myself, didn't care enough, and here we are: no DMM7510 adhesive templates on ebay. If you list for $1M it will probably piss me off enough to print some out and undercut you, though 

[KungFuJosh]

Hmmm. I feel like I should make some replacement adhesive cutouts with my laser and sell them for $1M on feeBay.

I was all on until I saw that price 😂

Only $250k on EEVblog direct. 🤣

Go ahead! Anyone and everyone who is willing to hold inventory and handle shipments has my permission to use the template.

https://www.dropbox.com/s/g3gghe2odsm8igm/dmm7510%20adhesive%20outline.svg?dl=0

I thought about doing it myself, didn't care enough, and here we are: no DMM7510 adhesive templates on ebay. If you list for $1M it will probably piss me off enough to print some out and undercut you, though 

Thanks! That helps a lot. Do you have an exact overall dimensions for the outline? I don't have a DMM7510 to confirm the size with.

All kidding aside, what would be a good price for these if I decide to make and stock them? Good price as in what people would pay for them. Then I can figure out the cost of production and see if it makes sense or not.

Thanks,
Josh