Tek discontinued the K2002 in June

[Cyclotron]

I hadn't seen it posted and its a significant passing.

https://www.tek.com/en/products/keithley/digital-multimeter/2002-series-8-5-digit-multimeter

Looks like they are promoting the DMM7510 in its stead. No direct replacement offered.

[KungFuJosh]

In a recent conversation with Tek, one of their reps told me:

Unfortunately, I do not believe there are any current plans to create a modern replacement for the 2002. If it is coming, it is far enough out that they are not telling anyone.

[bdunham7]

Well, it is a 30+ year old design and there aren't all that many use cases where it has any advantage over other models.  I suspect Fortive looked at the relatively low volume on precision bench meters plus the cost of reengineering new models and decided that the DMM6500 and DMM7510 pretty much covered most use cases and the Fluke and older Keithley models were redundant.  I can't say I disagree with them for most cases.  I think they discontinue them when they actually run out of parts and there's probably something in the 2002 that turned into unobtanium.  In the case of the Fluke 8846A I'm pretty sure it was the original Altera Cyclones that Fluke ran out of a few years ago.  They must have made an impressive last time buy because they made and sold the meter for more than 5 years after the last shipment of Cyclones.

[KungFuJosh]

Maybe, but I think there is a market for 8.5 digit meters, and it's big money. The HPAK 3458A is old AF but still widely sold, and Fluke has the 8558A and 8588A models with newer designs and huge price tags. Time will tell whether or not Keithley cares about that.

I think the biggest problem with the K2002 is exactly what you said; an old design with old parts becomes problematic.

[bdunham7]

Time will tell whether or not Keithley cares about that.

It's not Keithley, it is Fortive. They have 8.5 digit meters available from Fluke.  They're not going to spend a lot of money designing an all-new high-cost, low-volume product to compete with themselves. 

[Hydron]

I was going to say that Keithley IS Fluke (via Fortive), but apparently the bit of Danaher/Fortive that Tek is in got spun off?
Makes a bit less sense given that change, maybe there are commercial agreements as part of that spin off that discourage it? I do think it's a bit of a shame they aren't looking at a 8.5 digit evolution of the DMM7510.

[tooki]

Time will tell whether or not Keithley cares about that.

It's not Keithley, it is Fortive. They have 8.5 digit meters available from Fluke.  They're not going to spend a lot of money designing an all-new high-cost, low-volume product to compete with themselves.

FYI, it is not Fortive: Some of Fortive’s precision companies were spun off as Ralliant, and Tektronix is among them: https://ralliant.com/our-work#our-brands_id

So Fluke and Tek are no longer under one roof, they’re competitors now.

[tooki]

I was going to say that Keithley IS Fluke, but apparently the bit of Danaher that Tek is in got spun off?
Makes a bit less sense given that change, maybe there are commercial agreements as part of that spin off that discourage it? I do think it's a bit of a shame they aren't looking at a 8.5 digit evolution of the DMM7510.

Danaher (containing Tek and Fluke) spun off Fortive (also containing Tek and Fluke) which spun off Ralliant (containing Tek).

[Hydron]

I was going to say that Keithley IS Fluke, but apparently the bit of Danaher that Tek is in got spun off?
Makes a bit less sense given that change, maybe there are commercial agreements as part of that spin off that discourage it? I do think it's a bit of a shame they aren't looking at a 8.5 digit evolution of the DMM7510.

Danaher (containing Tek and Fluke) spun off Fortive (also containing Tek and Fluke) which spun off Ralliant (containing Tek).

Yeah I was being a bit lazy with Danaher = Fortive, was in the middle of editing to correct that when you replied!

[tooki]

Maybe, but I think there is a market for 8.5 digit meters, and it's big money.

Well that’s the thing: I suspect it’s very much not big money. Given that digital electronics have taken over so many things that used to be done with analog electronics, I would imagine that the need for ultra-precision multimeters is smaller than it once was.

So indeed, chances are the 2002 design reached its limits in terms of manufacturability, and that the new owners concluded that the market simply isn’t big enough to warrant the cost of developing a new one.

And let’s not forget that CERN developed its own open-source 8.5 digit meter, probably precisely to avoid having to buy hundreds of $$$$ 8.5 digit meters from Keithley, Keysight, or Fluke. (They wouldn’t have spent all the money to develop it if they only needed one or two units.) But that design is now out there for anyone who wants it, and it’s precisely the high-volume customers who might make more of those instead of buying COTS.

[KungFuJosh]

All you need to do is look at the price points of what's available, and there's clearly market segments not being served. We can speculate all we want, and time will tell if they care or not. In the meantime, speculate away.

[bdunham7]

So Fluke and Tek are no longer under one roof, they’re competitors now.

Hmmm.  It's hard to keep up!  Next we'll see Fluke instruments with subscriptions!

[nctnico]

And let’s not forget that CERN developed its own open-source 8.5 digit meter, probably precisely to avoid having to buy hundreds of $$$$ 8.5 digit meters from Keithley, Keysight, or Fluke. (They wouldn’t have spent all the money to develop it if they only needed one or two units.) But that design is now out there for anyone who wants it, and it’s precisely the high-volume customers who might make more of those instead of buying COTS.

I don't think you can just go around and built an 8.5 digit DMM as a diy project. Not even in a commercial setting. It takes knowledge and pedigree to hit 8.5 digit accuracy and stability levels. The latter alone takes many years to master. It is like forging Samurai swords. Sure you can make a pointy. sharp object from steel but creating a blade for a real sword is a whole different ball game.

My guess is that Keysight en Fluke will continue with their 8.5 digit meters; historically and pedigree wise, these companies are in better positions compared to Tektronix.

[tooki]

And let’s not forget that CERN developed its own open-source 8.5 digit meter, probably precisely to avoid having to buy hundreds of $$$$ 8.5 digit meters from Keithley, Keysight, or Fluke. (They wouldn’t have spent all the money to develop it if they only needed one or two units.) But that design is now out there for anyone who wants it, and it’s precisely the high-volume customers who might make more of those instead of buying COTS.

I don't think you can just go around and built an 8.5 digit DMM as a diy project. Not even in a commercial setting. It takes knowledge and pedigree to hit 8.5 digit accuracy and stability levels. The latter alone takes many years to master. It is like forging Samurai swords. Sure you can make a pointy. sharp object from steel but creating a blade for a real sword is a whole different ball game.

CERN has a long history of developing custom measurement hardware. (Remember, they do enough hardware that they found it viable to develop an entire EDA program: KiCad.) This isn’t some hobbyist throwing something together. These are experienced engineers with a background in high-tech measurement.

We don’t have to guess about performance: Marco Reps built one and characterized it.

(And yes, it likely took many years to develop, as lots of CERN hardware does. (I don’t have to guess about this: I’ve been peripherally involved in some CERN hardware.))

[TheDefpom]

Well, I wonder if that means they will release the circuit diagrams for it now, seeing as it is "outdated" and no longer made.

[coromonadalix]

i don't know if you can invoke the right to repair to get schematics  ?

[Hydron]

Well, I wonder if that means they will release the circuit diagrams for it now, seeing as it is "outdated" and no longer made.

I think we're in the wrong timeline for nice things like that to happen...

Looks like everything post brown-era (with the exception of a DMM or two that got schematics released, possibly due to US military requirements?) is without schematics, discontinued or not 

[Kleinstein]

The Keithley 2002 meter performance is a bit disapointing for a 8.5 digit meter.
Unless you need the 20 V high impedance range the meter is not that great. In many cases the DMM7510 is a good alternative, though also not perfect for high precision ( a similar issue with the AZ mode firmware - if they would fix that, maybe they could call it 8 digits).
I can perfectly understand the discontinuation. Some hard to get parts (CPLDs, display, relays) can be the final tipping point.

The Cern meter is not a full DMM, but just a limited 10 V range voltmeter with limited protection for use inside test systems. The use a part of test systems is still a good part of the high performance meter market.

[BigBoss]

Nobody cares about it...
All US and European measurement equipment manufacturers will either be sold or disappear or be withdrawn from the business soon.
Including Keysight, R&S, FLUKE, and the others. Chinese companies will run them over.
Siglent. Ceyear, Rigol, and presumably some unknown others. Because the technology shifts to the east. It's not so difficult to design and develop an SA or VNA or high-end DMM nowadays.
While a High-End VNA of Siglent is around 50K euros an equivalent of Keyisght is sold for more than 100K. They request money for every piece of option.
The first victim will be R$$$S ( it's a damn money trap company ). This is the Evolution by Natural Selection ( Charles Darwin )

[nctnico]

And let’s not forget that CERN developed its own open-source 8.5 digit meter, probably precisely to avoid having to buy hundreds of $$$$ 8.5 digit meters from Keithley, Keysight, or Fluke. (They wouldn’t have spent all the money to develop it if they only needed one or two units.) But that design is now out there for anyone who wants it, and it’s precisely the high-volume customers who might make more of those instead of buying COTS.

I don't think you can just go around and built an 8.5 digit DMM as a diy project. Not even in a commercial setting. It takes knowledge and pedigree to hit 8.5 digit accuracy and stability levels. The latter alone takes many years to master. It is like forging Samurai swords. Sure you can make a pointy. sharp object from steel but creating a blade for a real sword is a whole different ball game.

CERN has a long history of developing custom measurement hardware. (Remember, they do enough hardware that they found it viable to develop an entire EDA program: KiCad.) This isn’t some hobbyist throwing something together. These are experienced engineers with a background in high-tech measurement.

But still, CERN projects are typically aimed at use in a very controlled environment. I also have hands-on experience with projects CERN designed / built while taking it to a next level where you can actually have a commercially viable / useable product. This requires a lot of work. Sometimes to the point of starting from scratch with just the basic idea.

[TheSteve]

Well that news just makes my K2002 just a little more special. It also belonged to eevblog user volvo_nut_v70 (Ken) before he passed away. So even more of a keeper.

[pdenisowski]

All US and European measurement equipment manufacturers will either be sold or disappear or be withdrawn from the business soon.
Including Keysight, R&S, FLUKE, and the others. Chinese companies will run them over.

We're growing and just had a record year

https://www.rohde-schwarz.com/us/about-us/news-press/all-news/rohde-schwarz-shows-resilience-in-a-challenging-environment-revenue-exceeds-three-billion-euros-for-the-first-time_229356-1598656.html