EEVblog #731 - Keithley DMM7510 7.5 Digit Multimeter Teardown

[EEVblog]

Inside the new Keithley DMM7510 7.5 Digit Bench Multimeter
What gives this unit it's superb 16ppm/1year accuracy specification?
What construction techniques have they used?
How and why are Fluke so heavily involved?
What considerations are given to the power supply and isolation aspects?
How does it compare with the cheaper Keysight 34770A meter?

The Precision Thin Film Resistor Network video:

Datasheets:
LTC1050: http://www.linear.com/product/LTC1050
ISL28110: http://www.intersil.com/en/products/amplifiers-and-buffers/all-amplifiers/amplifiers/ISL28110.html
LSK389 Ultra low noise matched JFET http://www.linearsystems.com/assets/media/file/datasheets/LSK389.pdf
SSM2212 Dual matched transistor: http://www.analog.com/media/en/technical-documentation/data-sheets/SSM2212.pdf
DG411 Precision analog switch: http://www.vishay.com/docs/70050/dg411.pdf
TPS65251 Triple Switcher: http://www.ti.com/product/tps65251
Freescale MPC5125 Processor: http://www.freescale.com/webapp/sps/site/prod_summary.jsp?code=MPC5125

[Dr. Frank]

Dave,

thank you very much for that great video!!

You can literally feel and see, how much of analogue magic was designed into this instrument..

All these guard rings, isolation slots, and shielding methods, including that really ingenious low noise DC/DC converter..

They put in all of their know how about  Keithley electrometers and Fluke calibrators, obviously..

And one of the former HP3458A engineers works for Keithley since a few years..

So there's really where the brain is inside, not in the Keysight  counterpart.. that's very obvious, if you simply watch both videos.


OK, some remarks about the voltage reference and the thermal shielding..

This LTFLU package is supposed to be the regular one with 4 pins only, (judging from the type designator printing and photos from this device),  so there  is no heater inside the TO99 package, or not used, if there is any heater implemented on the reference chip. (as DiligentMinds always speculates)..

The white (ceramic ?) collar you see around the LTFLU, seems to be the exterior heater, it's named 'H2', obviously.
Its temperature might be controlled by the 2nd OpAmp , maybe.

The plastic or metallic shield around the whole assembly, also the one on the bottom side, also serve as a thermal isolation, so that the whole interior may heat up .. there's a LM35D temperature sensor also, maybe to measure if the whole assembly has been stabilized.

On the bottom side, I assume that there are 4 stable, (leaded?) resistors, which are always needed for the topology of this Reference Amplifier, aka LTFLU / SZA263 / T.I. device from all Fluke calibrators and voltage standards.

Anyhow, very impressive, and maybe, some day, Keithley will replace Keysight as the top notch DMM manufacturer..

Frank

[HighVoltage]

Really great tear down! Thank you for that.

Nice to see some German made WIMA capacitors on the board as well.
Now it would be nice to have a side by side comparison between this one and the Keysight 34470A 7 1/2 digit meter.

[3roomlab]

when pointer tip wriggled with the special DC-DC xfmr ... the magnet wire moved ... shouldnt that be glued and not moving? like the capacitor?

[digital]

Dave thank you for the look at something that I will not be able to afford,what a beautiful piece of design engineering.To be able to look at such engineering goodness was a treat,thank's again for the video.Regards to all.

[TiN]

Money video!

I bet it will put some heavy thinking into voltnut minds who wanted to buy 3458A before.

Few notes I think worth mentioning.

Reason for wonky optoisolation board shape was probably due using same digital processing board between different instruments, not just DMM7510. Not checked, maybe 2450/2460 uses same digital board? Would be expected to unify that piece of design and carry over with different instruments, since it's lot of man-hours spent on doing that bad boy. EDIT: Dooh..yes, video at 44min:09 sec says that

Keithley are good on low-signal magic. During tracing and reversing my Model 2001 units I had a lot of time scratching my head, thinking what are hell is this circuit is doing Even now since we have full schematics for 2001, lots of areas in it are rather cryptic.

Dave mention it's 18-bit ADC, but I remember Keithley marketing materials says 32-bit? Can find source yet, but somehow that stick into my mind.

Given options and free space on board, unpopulated FPGA, empty spot likely for VPG foil near LTFLU reference, wonder if there is something like DMM8510 in works with 8.5 digit somewhere in Keithley.

Interesting that they did not use Z-foil VHP resistor, but rather normal VHP203

R250,R499,R252,R500,R83,R81 around VREF are 99% surely VPG resistors, I have very similar looking VPG foil SMD resistor with same marking style as well

NFP version of Keithley gear usually more expensive than usual ones, btw

Also taking usual Keithley's numbering scheme on PCB boards, I bet million dollars that DMM7510 originally had Model 2040 name, as on analog board labels

[HKJ]

Dave mention it's 18-bit ADC, but I remember Keithley marketing materials says 32-bit? Can find source yet, but somehow that stick into my mind.

It has two ADC, one for high resolution and one for fast sampling.

[Tothwolf]

  Keithley

Nichicon PS http://www.nichicon.co.jp/english/products/pdfs/e-ps.pdf
Nichicon HV http://www.nichicon.co.jp/english/products/pdfs/e-hv.pdf
Nichicon FP http://www.nichicon.co.jp/english/products/fpcap_catalog.html

Panasonic (T-shaped vents) http://industrial.panasonic.com/ww/products/capacitors/aluminum-electrolytic-capacitors/aluminum-electrolytic-capacitors-radial-lead-type
FK series SMD http://industrial.panasonic.com/ww/products/capacitors/aluminum-electrolytic-capacitors/aluminum-electrolytic-capacitors-surface-mount-type

...all 105C rated long life parts
...not "cheaper" 85C rated parts
...and not some cheap generic no-name Shenzhen market special.

Keysight could learn a thing or two here.

[HighVoltage]

Money video!
I bet it will put some heavy thinking into voltnut minds who wanted to buy 3458A before.

You know, that is exactly what I was thinking.
I have been looking at the 3458A for a long time, just have not made a decision to buy.
Who knows, this one might be a good alternative.
May be Dave will also give us a review video.

[TiN]

If that would be me, I would not think a second chance, and get Keithley beast. Yes, 3458A is great and proven, but something bends me in Keithley way.
I wish my hobby funds could allow getting 7510. Trade 2002 to 7510 anyone? 

[Dr. Frank]

No way, guys!

Even though Wayne C. Goeke, designer of the HP3458As ADC, is the person to join Keithley, the 7510s ADC by no means comes near the 3458A!
Lay their specs side by side, and you will promptly see a big quality difference, yet.

Its linearity is mediocre 1..2ppm, again, like most all other 6 1/2 ... 7 1/2 DMMs, and still no other DMM has got this unique , true AUTOCAL function...
Obviously, the true value of this function (i.e., a complete cal lab built inside) is still not clear to everyone...

Frank

[EEVblog]

I have been looking at the 3458A for a long time, just have not made a decision to buy.
Who knows, this one might be a good alternative.
May be Dave will also give us a review video.

They are essentially chalk and cheese.
The 3458A is a metrology grade instrument, used as a transfer standard in cal labs around the world. There is a reason for this.
The 7510 is not. It is a practical 7.5 digit multimeter with a bewildering array of features, orders of magnitude more useful than the 3458A.

IMO, anyone outside of a cal lab or other very specific niche industry requriement who buys the 3458A over the 7510A is (volt)nuts.
I would take the 7510 over the 3458A any day of the week. I'd even take the 34470A over the 3458A. heck, even the 6.5 digit 34461A, there is just no comparison when it comes to practical usage.

[EEVblog]

Dave mention it's 18-bit ADC, but I remember Keithley marketing materials says 32-bit? Can find source yet, but somehow that stick into my mind.

Custom 32bit Keithley special for slow measurements, 18bit AD7982 for faster readings.
I actually spy two AD7982's on the board.

[dr.diesel]

IMO, anyone outside of a cal lab or other very specific niche industry requriement who buys the 3458A over the 7510A is (volt)nuts.e.

Yup, and like mentioned above I'd bet money a K8510 is in the works, and probably cheaper than the old trusty 3458A.

If you're really dying for that 8th digit to stare at, it might be present via SCPI.

[TiN]

Ok, I'm not voltnutty nut enough then, after rethinking what Dr. Frank said, he have a valid point.

1ppm linearity (1ppm range), somehow I expected better, that missed my attention. They could do better, even 2001 is <2ppm maximum, and 2002 is <0.2 max.

But even then, like Dave says, for bench 7510 is great tool.
Keithley's DS does mention noise is "equivalent to or better than many 8½-digit DMMs.", with 1 NPLC.

Also for fan, i think it was put like that intentionally, to have airflow go around case first, and distribute evently, rather than blow directly on components causing turbulence and all uneven gradients.

[Marco]

Even though Wayne C. Goeke, designer of the HP3458As ADC, is the person to join Keithley, the 7510s ADC by no means comes near the 3458A!
Lay their specs side by side, and you will promptly see a big quality difference, yet.

I'd guess the main problem is the ASIC ... the different channel length FET's for each charge path are a bit excessive/expensive.

[Lukas]

A while back, I pulled apart the firmware upgrade for the SMU 2400. Nested zips, then S-Record. IIRC, they used Greenhills Integrity as RTOS. So not your standard Linux / Windows CE job.

[schopi68]

i am currently thinking... about that really interesting ingenious solutions in the ke* 7510 and it's high complexity compared to the ke* 34470A with only a quarter of the amount of PCB and parts in it (and it's smaller outside footprint), less cabling stuff and the overall tidy looking construction.  I am not sure if this correlates directly to the better specs of the 7510.

This reminds me of the HP 3468A (portable brother of the 3478A) Multimeter designed in 1982/83 and the design-philosopy named in the Hewlett-Packard Journal 2/1983 on Page 8. http://www.hpl.hp.com/hpjournal/pdfs/IssuePDFs/1983-02.pdf

[Marco]

I actually spy two AD7982's on the board.

Might use one to determine how to do the multislope run up ... it would be silly overkill for that, but not compared to all the overkill in the thing any way.

[6thimage]

Great video!

I'm a little surprised at what it looks like inside - after Agilent's/Keysight's multimeters, I was expecting the same kind of system design. But it looks like they designed the measurement board and then tacked leftovers from other products together to get the finished device - it's not particularly impressive. I would be interested in seeing the difference between Keithley's special low-noise power supply and a linear supply, to see if there are any differences or whether Keithley are using it as a hacked together method to back themselves out of a corner, that they designed themselves into.

I hoping that the review videos for both the 34470A and the DMM7510 are in the works, as a high-end multimeter comparison would be very interesting. From the few minutes at the beginning of the video, it seems that Keithley aren't using the screen very well - with smaller digits, harder to read colours and harder to see which mode the meter is in. I do wonder about the usability of the multimeter, with the touch screen being the main input method - it looked like changing the range is a little more awkward than having dedicated buttons.

[RupertGo]

That secondary PSU for the analogue board is frankly bizarre. It must be switching a fair amount of current at a fair few MHz - is that really going to be a lower-noise system, even if it's locked to the sample clock,  than a decently-filtered linearly regulated bog standard mains toroid system? I'd love to see some RF EMI readings of that beast, although not as much as I'd like to see the circuit.

[FrankBuss]

Nice multimeter, but I still don't understand why they need the laser trimmed resistors. I understand this for older multimeters, with less processing power, but should be no problem to use just 1%, low tempco resistors and then calibrate it in software. The Freescale microcontrollers on the board have even a double precision floating point unit, and they run at 800 MIPS, so even at 1 MHz samplerate should be no problem to do one more multiplication per sample.

[SeanB]

I would guess that the analogue supply is using a sine wave drive, probably at a fair amplitude and able to give a low distortion output. That will give low noise ( aside from the fundamental) and will provide enough power using the simple 2 turn primary and probably a resonant secondary with some fast switches as synchronous rectifiers on the secondary side. 1MHz as a drive looks about right for the size, and you will get a fair power as well.

I think the 21V rails are used for the relay drive, as they probably want to reduce the current in the coils, and a 24V coil will operate reliably with 21V, and still switch fast enough. There must be some switchers on the supply board though, as getting 5V off 21V with a linear regulator will involve some cooking. The converters are probably on the other side of the board, shielded by the transformer shield, and using those large collection of Nichicon and polymer caps as final filtering.

[lampbus]

"spiral PCB cutouts for vibration isolation..." not.

When the plastic box/lid is in place, it holds it all tightly in place, bridging the gaps in the spiral. Effectively the box hangs directly on four tabs to the main PCB, and the internal circuitry hangs by four tags inside the box.

Thermal isolation. Not seismic isolation.

The plastic boxes look like standard off the shelf enclosures - and cheap ones at that : no gasket or cover interlocking, best can be said of them is the metal thread inserts.

[SeanB]

The boxes are not off the shelf, you can see the diagonals have 2 locating pips, and a corresponding hole in the PCB for them as well, and the bottom sections have the same 2 pips on the other diagonal. Makes it easy to put on the board and align so the screws are going to be in the centre of the holes in the board and will not bind on the GRP material, and positions the box very accurately.

Not cheap styrene or PVC boxes, more like a precision moulded polycarbonate, polypropylene or glass filled nylon, in a rather expensive mould.

[Unixon]

At 20:55 there's a guard trace around Q90 and other circuitry looks like cut/interrupted by an isolation slot.
I wonder if this was intentional or not...
However, at least above Q90 it goes through via to somewhere else, a ground plane perhaps.

Even more cut guard traces at about 24:20. Looks like cutting guard traces with isolation slots is a common practice here...

[thewyliestcoyote]

Never thought I would as about Nyquist sampling on a DMM. With a sample rates as high as 1 MSps has any one thought about doing some testing see the frequency performance of the DC measurement. Haha never thought I would as that about a DMM. 18 bits is almost pushing the limits of 32 bit floating point math for a FFT.

Dave maybe you could do a frequency response testing? Maybe just a frequency sweep from  DC to maybe 2 MHz, or a couple static frequency measurements. This would also show if there is some aliasing possible. Crazy to be thinking about this for a DMM.

Thanks,
Wylie

[nilium]

Excellent video indeed.

Probably a basic question but you mentioned their grounding being improper. Would you please elaborate on that and mention how it would be proper?

Thanks!

NV

[bktemp]

At 20:55 there's a guard trace around Q90 and other circuitry looks like cut/interrupted by an isolation slot.
I wonder if this was intentional or not...
However, at least above Q90 it goes through via to somewhere else, a ground plane perhaps.

Even more cut guard traces at about 24:20. Looks like cutting guard traces with isolation slots is a common practice here...

Looks almost like an afterthought: They did put the guard traces between the pins but then decided the guard traces are not enough and they did cut the slots instead.


I have no idea if it is possible, but could the laser trimming on those resistor networks been used to adjust the temperature coefficient? The traces have different colours, they are probably different materials. Lets say those have slightly different coefficients. By trimming only one material it could be possible to trim the coefficent to zero.

[Marco]

That secondary PSU for the analogue board is frankly bizarre. It must be switching a fair amount of current at a fair few MHz

There's a lot of ferrite in that thing ... would be cool if Dave could measure it's operating frequency.

is that really going to be a lower-noise system

Maybe they want to minimize capacitive coupling to ground as well?

[dr.diesel]

A while back, I pulled apart the firmware upgrade for the SMU 2400. Nested zips, then S-Record. IIRC, they used Greenhills Integrity as RTOS. So not your standard Linux / Windows CE job.

I specifically asked this question when I bought my K2450.  I really confused the sales lady but eventually got to the right guy.  He would only state that is does not run Windows/Linux.

[Unixon]

The U34 has a funny looking long package. What is this IC, what is it doing?

[SeanB]

Guard trace is to reduce leakage current across the board surface. Cutting the board means no more surface, and having the guard trace right to it means no leakage path possible. The slots were probably added to increase both breakdown voltage, reduce stress on the package and ensure that there is no possible leakage path. Cheaper than PTFE standoffs and solid wiring floating in air, and more repeatable and easier to compensate.

[bktemp]

The U34 has a funny looking long package. What is this IC, what is it doing?

https://www.idt.com/document/dst/qs34x2245-datasheet
Looks like it is disconnecting the GPIB ic.

[f4eru]

"Individually Hand selected and tweaked someone with a grey beard or nude virgins, or both"

To attain this kind of precision, it has to be a grey bearded nude virgin.
Very hard to find such a guy, and this makes the cost of these meters...

This free standing vishay resistor is a bit howya doin', could have been fastened somehow.

[mikeselectricstuff]

When the plastic box/lid is in place, it holds it all tightly in place, bridging the gaps in the spiral. Effectively the box hangs directly on four tabs to the main PCB, and the internal circuitry hangs by four tags inside the box.

Thermal isolation. Not seismic isolation.

I suspect some of the cut-outs are also to do with decoupling mechanical stress - on a big PCB, hotspots in various places, as well as the mechanical fixings  are bound to set up some mechanical stresses within the PCB, which you don't want transferred into the leads of sensitive devices.
The slots round the reference for example will minimise the transfer of any board warp or other movement onto the section of PCB containing the reference.

Something else that struck me - all those slots between IC pins - but you still have the potential for surface leakage on the IC body -  is the leakage across the PCB enough that it makes a difference to do cutouts when you still have a short path on the IC body?

[tszaboo]

Nice multimeter, but I still don't understand why they need the laser trimmed resistors. I understand this for older multimeters, with less processing power, but should be no problem to use just 1%, low tempco resistors and then calibrate it in software. The Freescale microcontrollers on the board have even a double precision floating point unit, and they run at 800 MIPS, so even at 1 MHz samplerate should be no problem to do one more multiplication per sample.

I've seen 2% resistors even those were laser trimmed. Things like the 10 Mohm input impedance doesnt tolerate too much difference.

The general feel of the meter it is like it was snapped together. I really dont like the construction,  all those million cables zapping everywhere and boards in every direction... That coax transformer especially. If you need to design a low noise power supply, design it properly. There are options like PCB planar transformers ZCS, and countless methods to generate low noise power. Probably half the guard rings and slots are not necessary. When I say those pins of that dual Jfet in the air  This dual gigahertz CPU is also... I mean cmon, I know that processing power is cheap but do you really need that?
Dave the you cannot appreciate the 3458A until you used it. The first few days I used it I was yaw dropping and smiling, saying that "I can measure stuff that I didnt even thought I can measure". Like potential increase on a solid ground plane after turning on a LED. Probably it even has faster display update than this Keithley.
Probably the end result is still good though. I mean the specs are very good, and it does the job.

[dr.diesel]

This free standing vishay resistor is a bit howya doin', could have been fastened somehow.

It may look funny, but clamping it down is just not required.  You could toss that thing in the back of a log wagon on an epic coast to coast voyage and it would arrive fine.  (well the vishay would be fine anyhow, LOL)

[open loop]

I personally would be very careful  working on that analogue board during a repair. With all of those cut outs I can imagine it would be easy to damage the PCB if one is a little "ham fisted". Can just imagine the repair tech saying, "just broke the PCB going to voltage ref, is that a problem?" 

I wouldn't expect this to break down for at least the next 10 years of careful use. With the meter having a 1 meg sample rate I assume you could record audio 

Is that something Dave would like to have a go at?

[DanielS]

This free standing vishay resistor is a bit howya doin', could have been fastened somehow.

The "Do Not Bend" silkscreen next to the resistor implies that the resistor should not be subjected to any external stresses, likely because it could mess up calibration. Fastening it to something would introduce additional drift from mechanical stresses such as thermal expansion and package compression/deformation from being constrained.

When you are at a resolution level where looking at components sideways can affect readings, many things that do not seem to make sense may become critical.

[lampbus]

The boxes are not off the shelf, you can see the diagonals have 2 locating pips, and a corresponding hole in the PCB for them as well, and the bottom sections have the same 2 pips on the other diagonal. Makes it easy to put on the board and align so the screws are going to be in the centre of the holes in the board and will not bind on the GRP material, and positions the box very accurately.

Not cheap styrene or PVC boxes, more like a precision moulded polycarbonate, polypropylene or glass filled nylon, in a rather expensive mould.

Yes, you are right, I noticed the little pins, but failed to notice that there were no mating holes in the opposing  cover. They were never intended to mate without the PCB in between.

I googled the number in the cover - it is a reference to glass re-enforced nylon. (2040-310) but I don't know what the B means that follows the number.

The mold tool may not be all that expensive - they will not make that many of these parts ? Soft tooling ? They may even use a standard mold tool with different inserts to make the little pips, but I cant see any tool joint lines.

On the other hand, the wall thickness is high to limit thermal conductivity - most enclosures I use have quite thin section so on balance probably all special tooling. I think I would have taken the opportunity to make it look a bit nicer  - they went overboard with the logo on the bail...so why not put one on the internal parts that help make this kit different from the lower accuracy manufacturers ?

[JackOfVA]

There's something odd about the binocular core isolation transformer.  It looks as if it is constructed from two different material types. An outer tube of ferrite (rougher and dark gray) and an inner tube of a smooth light gray material. The light gray color appears to come from an epoxy or parylene coating. Some ferrite material is parylene coated, but I normally associate that with powdered iron material. If the inner core is from MicroMetals, gray corresponds to powdered iron "mix-3" optimized for 20 KHz - 2 MHz for resonant circuits. 


Or, perhaps the outer sleeve is plastic.

If the outer sleeve is ferrite, perhaps it's function is to work as a magnetic shield, although this strikes me as unusual.

[thm_w]

I suspect some of the cut-outs are also to do with decoupling mechanical stress - on a big PCB, hotspots in various places, as well as the mechanical fixings  are bound to set up some mechanical stresses within the PCB, which you don't want transferred into the leads of sensitive devices.
The slots round the reference for example will minimise the transfer of any board warp or other movement onto the section of PCB containing the reference.

Something else that struck me - all those slots between IC pins - but you still have the potential for surface leakage on the IC body -  is the leakage across the PCB enough that it makes a difference to do cutouts when you still have a short path on the IC body?

Epoxy appears to be a few orders of magnitude worse than FR4. The leakage of the IC itself is again a few orders of magnitude worse than the epoxy encapsulation (DG411 = nA's for example). These values were pulled from random datasheets for epoxy/FR4:


FR4:
Volume resistivity: 3x10^13
Thermal expansion: 10-70ppm/K

Epoxy:
Volume resistivity: ~1x10^11
Thermal expansion: 15ppm/mC

http://c-component.ru/sites/default/files/KCC%20Organic%20Material%20EMC%20%28Epoxy%20Mold%20Compound%29.pdf
http://www.leiton.de/formulare/Datasheet%20-%20rigid%20FR4%20PCBs%20-%20Rev%202.2.pdf
http://www.andus.de/_pdf/probimer-77.pdf

I can see there is less chance for dust/contamination to build up between leads. But other than that, not sure, maybe someone else can comment further.

[JoeN]

The U34 has a funny looking long package. What is this IC, what is it doing?

https://www.idt.com/document/dst/qs34x2245-datasheet
Looks like it is disconnecting the GPIB ic.

I just dropped by this thread to ask what that guy is.  Never seen a SSOP-80 before.  They are calling it "MilliPaQ".  Good name.  I want one on my project just because it looks cool.   

Well, at least they are cheap if anyone cares.  Sucky on resistance, though.  Needs an update.  http://www.digikey.com/product-detail/en/QS34X2245Q3G/800-1742-5-ND/1916043

Spoke too soon, they have a much better 3.3V version with Ron of 4 ohm instead of 23 ohm.  http://www.idt.com/document/dst/qs34xvh245-datasheet  http://www.digikey.com/product-detail/en/QS34XVH245Q3G/800-2196-ND/2018992

[Marco]

National did opamps with fA range leakage currents in epoxy packages ... and even then most of that was probably in the circuit.

Worry about drenching them in flux rather than the epoxy.

[victor]

That fan fixed on a huge pcb... 

I'm sure that will act pretty much as a amplifier for the noise. I bet you can attach strings to that board and make a guitar out of a pcb.

[rs20]

I know this question must have an answer, because guard rings are so widely used, but:

What's the point of all these guard rings & isolation slots if the guard rings don't extend, for example, to the plastic on the underside (and top) of the relays & SOIC packages? Are the plastic packages of these components somehow treated to make them far less susceptible to contamination/humidity than the PCB surface? The length of "missing guard trace" on the packages is small, but so is the distance between pins, so it cancels out as far as I can tell from that point of view. For a given level of contamination, there'll be just as much leakage happening on the package surface as on the PCB. Why am I wrong?

Like, look at 26:07. Am I the only one who sees a MASSIVE piece of black plastic completely defeating the clearance and creepage advantages afforded by those intricate slots? 

[rf-design]

When the plastic box/lid is in place, it holds it all tightly in place, bridging the gaps in the spiral. Effectively the box hangs directly on four tabs to the main PCB, and the internal circuitry hangs by four tags inside the box.

Thermal isolation. Not seismic isolation.

I suspect some of the cut-outs are also to do with decoupling mechanical stress - on a big PCB, hotspots in various places, as well as the mechanical fixings  are bound to set up some mechanical stresses within the PCB, which you don't want transferred into the leads of sensitive devices.
The slots round the reference for example will minimise the transfer of any board warp or other movement onto the section of PCB containing the reference.

I think your are right. The box and the board spiral are for both purposes. For thermal and mechanical isolation. The buried zener diode themselve is less sensitive to mechanical stress transfered from the board to the package over die glue to the silicon. But the temperature regulator within the LTFLU could be impacted by mechanical stress. The bandgap voltage of the biplar transistors change with the stress and I think the regulator is based on comparison of the scaled temperature voltage to the bandgap voltage.

[rf-design]

No way, guys!

Even though Wayne C. Goeke, designer of the HP3458As ADC, is the person to join Keithley, the 7510s ADC by no means comes near the 3458A!
Lay their specs side by side, and you will promptly see a big quality difference, yet.

Its linearity is mediocre 1..2ppm, again, like most all other 6 1/2 ... 7 1/2 DMMs, and still no other DMM has got this unique , true AUTOCAL function...
Obviously, the true value of this function (i.e., a complete cal lab built inside) is still not clear to everyone...

Frank

Do you know what is the specific feature which take the linearity from 2ppm to 0.2ppm in the ADC?

[tszaboo]

I think your are right. The box and the board spiral are for both purposes. For thermal and mechanical isolation. The buried zener diode themselve is less sensitive to mechanical stress transfered from the board to the package over die glue to the silicon. But the temperature regulator within the LTFLU could be impacted by mechanical stress. The bandgap voltage of the biplar transistors change with the stress and I think the regulator is based on comparison of the scaled temperature voltage to the bandgap voltage.

Even thoght the LTFLU is less susceptible, the resistors and capacitors connected to it will be affected by bending the board a lot. I dont know all that well the LTFLU, but I could imagine that changing the current going through it is a bad thing.

[rf-design]

I would guess that the analogue supply is using a sine wave drive, probably at a fair amplitude and able to give a low distortion output. That will give low noise ( aside from the fundamental) and will provide enough power using the simple 2 turn primary and probably a resonant secondary with some fast switches as synchronous rectifiers on the secondary side. 1MHz as a drive looks about right for the size, and you will get a fair power as well.

I think the 21V rails are used for the relay drive, as they probably want to reduce the current in the coils, and a 24V coil will operate reliably with 21V, and still switch fast enough. There must be some switchers on the supply board though, as getting 5V off 21V with a linear regulator will involve some cooking. The converters are probably on the other side of the board, shielded by the transformer shield, and using those large collection of Nichicon and polymer caps as final filtering.

The isolation is mainly done by using coaxial on the primary. I will explain that:

You could imagine that a transformer have parasitic caps between primary and secondary. If there is no shielding the parasitic caps are distributed along the primary and secondary wire and couple depending on the details of t he mechanical winding. You can simplify if you assign an equivalent network for the parasitic caps with two coupling caps. If the transform is symetric both caps are equal.
If the transformer is driven by an alternate signal the rectifiying diodes connect alternating the to either one of the parastic caps. If the parasitic caps are equal and the primary driving voltage is also symmetric relative to the primary ground or case earth than the charge balance trough one rectifying period is zero. If either the driving or the parasitic cap is imbalanced you get a effective AC noise current source between the grounds. One method is to shield the driving because a symmetric drive or a balanced coupling is more difficult.

[rf-design]

Never thought I would as about Nyquist sampling on a DMM. With a sample rates as high as 1 MSps has any one thought about doing some testing see the frequency performance of the DC measurement. Haha never thought I would as that about a DMM. 18 bits is almost pushing the limits of 32 bit floating point math for a FFT.

1 second FFT with 1MSps give 20 additional bits in the mantissa which is more than 32Bit IEEE Floating allows. But 64Bit IEEE Floating could handle that. But that I expect that also with 32Bit the numerial errors are much less than the noise from the hardware. For less numerical errors you could also switch to DFT instead of FFT which less sensitive to rounding errors.

[rollatorwieltje]

That 3904 could be used as a temperature sensor. Came across a setup like that when looking at the LTC2990 datasheet, an i2c voltage/current/temperature ic.

[teddy529]

Really great tear down! Thank you for that.

Nice to see some German made WIMA capacitors on the board as well.
Now it would be nice to have a side by side comparison between this one and the Keysight 34470A 7 1/2 digit meter.

If someone is going to do it, that'd be my best birthday gift. Not just tear down, but real applications.

[Dr. Frank]

No way, guys!

Even though Wayne C. Goeke, designer of the HP3458As ADC, is the person to join Keithley, the 7510s ADC by no means comes near the 3458A!
Lay their specs side by side, and you will promptly see a big quality difference, yet.

Its linearity is mediocre 1..2ppm, again, like most all other 6 1/2 ... 7 1/2 DMMs, and still no other DMM has got this unique , true AUTOCAL function...
Obviously, the true value of this function (i.e., a complete cal lab built inside) is still not clear to everyone...

Frank

Do you know what is the specific feature which take the linearity from 2ppm to 0.2ppm in the ADC?

Due to the description in hpj 4/89, the linearity is due to complete compensation of charge injection by the FET switches, which steer the ramping currents.
It's a mix of hardware, i.e. these specially designed multi current stages with adapted FET sizes, plus a special software state machine steered integration pattern.
The 1281/8508 DMMs have also a very linear AD, 0.1ppm, but I can't tell, what's unique about these.

Maybe patents still guard these designs.

Multislope III, IV, and also the 7510 AD look very simple, a few switches, 2 ..3 OpAmps, and that's it..
You can get high resolution out of if it, but w/o further effort as above, linearity stays mediocre

Frank

PS: The 3458A has 0.02ppm typ. linearity, not 0.2ppm!!

[Dr. Frank]

That 3904 could be used as a temperature sensor. Came across a setup like that when looking at the LTC2990 datasheet, an i2c voltage/current/temperature ic.

I didn't pay special attention to these circuit elements in the video, but I recognized also this bunch of discrete transistors, coming in pairs, at it seemed to me.

Reversed bipolar transistor diodes (BC diode, afair) may well serve as ultra low leakage protection diodes for overvoltage..

There's a very uncommon diagram in AoE, about transistor characteristics, showing this low voltage range, with a comment fitting that idea..
I think, the 2N3904 served especially well for that, i.e. leakage in the fA range. (*)
Frank

AoE, p. 294-295, ch. 5.2.2, about protection of OpAmp input stages.
Figure 5.2 shows 10fA forward current at 1mV for the 2N3904 B-C diode, but 10mV for the B-E diode.

reverse currents are even smaller..

great.. no need to use expensive low leakage didoes.

[Marco]

Maybe patents still guard these designs.

Not unless they submarined the hell out of the patents.

I wonder if it wouldn't be smarter of instead of switching in the voltage-resistor combos with JFETs to switch current into the integrator. If you first create the current you can switch in voltages with MOSFETs with negligible on state resistance, no need to be in a hurry. Then you can use a single JFET, with diodes in front of it to dump the current into ground when it's off, to switch the current into the integrator (or maybe that's patented, I haven't looked).

[mikeselectricstuff]

I just dropped by this thread to ask what that guy is.  Never seen a SSOP-80 before.  They are calling it "MilliPaQ".  Good name.  I want one on my project just because it looks cool.   

I'm almost tempted to order one just to x-ray it to see what the leadframe looks like...

[Mickle T.]

I wonder if it wouldn't be smarter of instead of switching in the voltage-resistor combos with JFETs to switch current into the integrator.

Keithley 2002, Prema 6048, Advantest 6581 and many other DMMs do the same thing.

[Marco]

I wonder if it wouldn't be smarter of instead of switching in the voltage-resistor combos with JFETs to switch current into the integrator.

Keithley 2002, Prema 6048, Advantest 6581 and many other DMMs do the same thing.

That diagram still has a high speed switch (ie. non negligible resistance) for switching in the input voltage/resistor. They also have multiple high speed switches for the current sources, all build with different active components.

What I meant is that you could switch voltage/resistor combos into a single virtual ground for a single opamp current source, with slow negligible resistance switches (ie. <100 mOhm MOSFETs). Once the current stabilizes you put it into the integrator at the start of a clock cycle with a single high speed switch, when you want to change the current you take the current off at the end of a clock cycle and rinse and repeat.

[splin]

Dave mention it's 18-bit ADC, but I remember Keithley marketing materials says 32-bit? Can find source yet, but somehow that stick into my mind.

Custom 32bit Keithley special for slow measurements, 18bit AD7982 for faster readings.
I actually spy two AD7982's on the board.

Those 18 bit 1MSPS AD7982s aren't cheap devices at approx $23 each @ 1K. Anyone got the faintest idea why they would have 2 of them?

I can't think of any good reason - they *could* use them to provide full rate 1MSPS digitizing with alternating autozero and gain calibration to avoid drift during long sampling runs but it seems very unlikely. In any case they could have used a single 2MSPS 18 bit AD7641 at $29. Also the manuals/specs don't suggest it does this. 

[eV1Te]

Those 18 bit 1MSPS AD7982s aren't cheap devices at approx $23 each @ 1K. Anyone got the faintest idea why they would have 2 of them?

I can't think of any good reason - they *could* use them to provide full rate 1MSPS digitizing with alternating autozero and gain calibration to avoid drift during long sampling runs but it seems very unlikely. In any case they could have used a single 2MSPS 18 bit AD7641 at $29. Also the manuals/specs don't suggest it does this.

Maybe it uses one for the sense-inputs and one for the standard inputs (e.g. 4-wire resistance)

[Marco]

I can't think of any good reason - they *could* use them to provide full rate 1MSPS digitizing with alternating autozero and gain calibration to avoid drift during long sampling runs but it seems very unlikely. In any case they could have used a single 2MSPS 18 bit AD7641 at $29. Also the manuals/specs don't suggest it does this.

Multislope converters want to know where approximately the slope is, to avoid saturation during run up and keep run down times small ... so they could be using one of them for that.

[Mickle T.]

Multislope converters want to know where approximately the slope is, to avoid saturation during run up and keep run down times small ... so they could be using one of them for that.

All of the eight high-end ADCs in the twelve 8.5-digits DMMs don't want to know where approximately the slope is 

[Marco]

I'm pretty sure the 3458 doesn't actually go to 120 kV ... they at the very least have a limit comparator, which is an ADC.

[Mickle T.]

3458A have a zero-cross comparator only :-(

[Marco]

You're right, they probably just assume a worst case input voltage and swing it through zero more often with more switching than otherwise necessary (and overpower the input voltage on the first slope to make sure it's of known polarity). The lower end HP/Agilent multimeters with schematics available often have a signal running off to an IC with an input called FLASH, which is kinda suggestive. Keithley has a second comparator in 2001 (thus almost certainly in the 2002 as well).

[tszaboo]

Dave mention it's 18-bit ADC, but I remember Keithley marketing materials says 32-bit? Can find source yet, but somehow that stick into my mind.

Custom 32bit Keithley special for slow measurements, 18bit AD7982 for faster readings.
I actually spy two AD7982's on the board.

Those 18 bit 1MSPS AD7982s aren't cheap devices at approx $23 each @ 1K. Anyone got the faintest idea why they would have 2 of them?

I can't think of any good reason - they *could* use them to provide full rate 1MSPS digitizing with alternating autozero and gain calibration to avoid drift during long sampling runs but it seems very unlikely. In any case they could have used a single 2MSPS 18 bit AD7641 at $29. Also the manuals/specs don't suggest it does this.

Probably to have overlapped measurement/calibration cycles. Driving high speed accurate ADCs is a real task, and you cannot switch every other measurement. Settling time of 0.5uS 5V with 18 bits, and to have anti-aliasing filtering... well that is a bunch of high speed amplifiers (fairly precision) after each other.

[splin]

Those 18 bit 1MSPS AD7982s aren't cheap devices at approx $23 each @ 1K. Anyone got the faintest idea why they would have 2 of them?

I can't think of any good reason - they *could* use them to provide full rate 1MSPS digitizing with alternating autozero and gain calibration to avoid drift during long sampling runs but it seems very unlikely. In any case they could have used a single 2MSPS 18 bit AD7641 at $29. Also the manuals/specs don't suggest it does this.

Probably to have overlapped measurement/calibration cycles.

Yes, but as I said, "it seems very unlikely". Having looked at the reference manual I found the statement:

"Digitize function do not support autorange, autozero, or auto delay."

I very much doubt that it will  keep correcting the gain calibration if it doesn't autozero. In any case the AD7982 has a typical gain error temperature drift of only 1uV/K or .2ppm (and a zero drift of .5uV/K). The front end attenuators and amplifiers and filters will add some drift but probably not much. Also the Digitize Voltage accuracy 90-day specs are pretty undemanding at only 110 (measurment) + 75ppm (range) with a (best) temperature coefficient outside of +/- 5C of 10 + 20ppm.

So the question remains as to why they included the second AD7982 given that it probably costs as much, if not more than, all the components in the main multi-slope converter?

Driving high speed accurate ADCs is a real task, and you cannot switch every other measurement. Settling time of 0.5uS 5V with 18 bits, and to have anti-aliasing filtering... well that is a bunch of high speed amplifiers (fairly precision) after each other.

True, but I don't think that's the issue. And the digitize function is not exactly high precision - that 110ppm + 75ppm  (range) accuracy is barely better than 12 bits! I guess that Keithley didn't see accuracy as being particularly important for this feature or, perhaps more likely, the specs are very conservative.

I guess we need someone to reverse engineer the signal paths to the two AD7982s. Or a Keithley insider!

[PolarisAT]

Hi everyone,

first of all I really like Dave's show and the teardowns! Keep up the good work Dave!

But in this teardown I just don't get it how this special transformer/powersupply with the coax cabel works. Can someone please explain me what Dave is talking about and probably make a small drawing for me? Or even better, could Dave make a separate video on this topic? That would be great!

[plesa]

Interesting comparisons Keysight 34470A with Keithley DMM7510 http://imgur.com/gallery/Wbx7q

[SharpEars]

Interesting comparisons Keysight 34470A with Keithley DMM7510 http://imgur.com/gallery/Wbx7q

I wish there were more comparisons of the DMM7510 vs 8 1/2 digit meters for noise levels. I think it would handily beat many of them also!

[Mickle T.]

I think many of the old good nanovoltmeters beats the 7510
EM N31, Tinsley 6045 have a ~100 picovolt p-p noise. How about the 7510?

[splin]

About that coax transformer - for those who haven't already come across this. Some while ago in the "T.C. measurements on precision resistors" thread, Dr Frank mentioned a CERN paper by John Pickering et al: 'Design and Evaluation of a 10-mA DC Current Reference Standard'.

http://cds.cern.ch/record/643294/files/cer-002399331.pdf

Having recently read it took a while before the penny dropped as to where I had come across a transformer just like the one in the paper.

VI. DESIGN FOR ISOLATION

The  back-to-back  calibration  configuration is shown in Fig. 3 and inspection shows that if the power supplies generate common mode current, this will flow through the sense resistor. It, therefore, constitutes an error if it is dc or a disturbance if it is AC. A new dc–dc converter using a patented double-screened transformer construction was chosen, together with a slew-limited low-noise switching controller. This ensures that there will be no difference between use with external power supplied and use with the external supply completely disconnected when under battery power. The double-screened transformer uses two toroidal cores, independently wound, one each for primary and for secondary. Each is fully screened within injection moulded conductive plastic shrouds and coupled with an external “shorted turn” winding around both cores. The resulting coupling capacitance from primary winding to secondary screen is less than 0.1 pF and in use 10 nA of pk-pk AC coupling current is achieved compared with at least 1000 times more in most commercial units.

So that's where Keithley got the design from, and helpfully for them, out of patent by now.

[zlymex]

......
The white (ceramic ?) collar you see around the LTFLU, seems to be the exterior heater, it's named 'H2', obviously.
Its temperature might be controlled by the 2nd OpAmp , maybe.

The plastic or metallic shield around the whole assembly, also the one on the bottom side, also serve as a thermal isolation, so that the whole interior may heat up .. there's a LM35D temperature sensor also, maybe to measure if the whole assembly has been stabilized.

On the bottom side, I assume that there are 4 stable, (leaded?) resistors, which are always needed for the topology of this Reference Amplifier, aka LTFLU / SZA263 / T.I. device from all Fluke calibrators and voltage standards.

Sorry to bring this up, but this is the first time for me seeing this thread and I'm trying to clarify might-be misunderstanding regarding the reference heater.

My guess is, two N95 transistors(FCX495) are the heater drives and also function as heating elements, their emitters are connected to heating resistor R250, R499, R500 and R252 thru thick traces.
LM35D would be the temperature sensor of the oven.
Also, this part of the board/oven is a 7V reference only, so there is no precision resistor pairs to step up to 10V.

[fonograph]

how many measurements per second at 7.5 digit?

[niner_007]

Those 18 bit 1MSPS AD7982s aren't cheap devices at approx $23 each @ 1K. Anyone got the faintest idea why they would have 2 of them?

I can't think of any good reason - they *could* use them to provide full rate 1MSPS digitizing with alternating autozero and gain calibration to avoid drift during long sampling runs but it seems very unlikely. In any case they could have used a single 2MSPS 18 bit AD7641 at $29. Also the manuals/specs don't suggest it does this.

DMM7510 has a secondary measurement function, for example it can measure voltage and current simultaneusly without a relay switch; it might be used for that, or maybe it's for its ACAL function

Maybe it uses one for the sense-inputs and one for the standard inputs (e.g. 4-wire resistance)