3458A - black edition

[notfaded1]

Personally I think that SL935 is the neatest thing!  I watched the video when you were rebuilding the oven for it and reversing the schematic with the trimmed resistors.  Really impressive as long as it's running it doesn't need a bunch of math to compute the TC curve like the SR104 basically because it maintains it's temp in search of the zero slope.  8x8 DMM's o.O!  I can only imagine what you have planned for the ABE!  We can live vicariously through you TiN.

https://www.hpl.hp.com/hpjournal/pdfs/IssuePDFs/1989-04.pdf

Bill

[Wolfgang]

"Same specifications as released in 1989."

Our ancestors did a good job on this one!

Unfortunately, yes. I think they changed the bare minimum to get it thru the ROHS procedure.

All the unpracticalities are still there:
- Spooky and fading VFD
- No USB or LAN connectivity, only GPIB (in 2020!!)
- Antique operation

They are sending me marketing emails every week, but I think buying a well-hung one running for 10 years will have less drift than a new one.
In the meantime I try to survive with my 34470A, and most of the time it works very well (used for thermal drift measurements).

[SilverSolder]

"Same specifications as released in 1989."

Our ancestors did a good job on this one!

Unfortunately, yes. I think they changed the bare minimum to get it thru the ROHS procedure.

All the unpracticalities are still there:
- Spooky and fading VFD
- No USB or LAN connectivity, only GPIB (in 2020!!)
- Antique operation

They are sending me marketing emails every week, but I think buying a well-hung one running for 10 years with have less drift than a new one.
In the meantime I try to survive with my 34470A, and most of the time it works very well (used for thermal drift measurements).

It is a little sad that "our" generation can't seem to move the state of the art forward in a meaningful way.  The same malaise as aerospace and other industries seem to suffer from.

[Wolfgang]

I would not be so pessemistic. You can still buy a Fluke or a Transmillle (both for some extra money), but they are up-to-date and should work fine.
HP is just trying to get away with a minimum investment, I am convinced they could do better if they really want.

[ScoobyDoo]

Hello - there is an article in EDN about the 3458A B.E. :

https://www.eetimes.com/metrology-grade-dmm-gets-a-new-colors-is-that-really-news/?utm_source=newsletter&utm_campaign=ad&utm_medium=EDNTestMeasure-20200120#

Herzliche Grüße/Meilleures salutations/Best regards
ScoobyDoo

[SilverSolder]

Hello - there is an article in EDN about the 3458A B.E. :

https://www.eetimes.com/metrology-grade-dmm-gets-a-new-colors-is-that-really-news/?utm_source=newsletter&utm_campaign=ad&utm_medium=EDNTestMeasure-20200120#

Herzliche Grüße/Meilleures salutations/Best regards
ScoobyDoo

I guess the only way for Keysight to introduce an advancement would be to introduce a whole new meter, that people could slowly "switch to" in their own sweet time.

[Wolfgang]

Hello - there is an article in EDN about the 3458A B.E. :

https://www.eetimes.com/metrology-grade-dmm-gets-a-new-colors-is-that-really-news/?utm_source=newsletter&utm_campaign=ad&utm_medium=EDNTestMeasure-20200120#

Herzliche Grüße/Meilleures salutations/Best regards
ScoobyDoo

I think Keysights conclusion that they could not change anything because the users would never want that is pure marketing bullshit and an excuse for no improvements.

Of course they could:
- officially sell a 2ppm version (and ask a good price for it)
- make a 2020-style LCD like on the 34470A
- incorporate USB and LAN interfaces
- modernize SCPI

it would still be possible to incorporate a "vintage mode" to fully emulate the old instrument so existing users have a drop-in replacement.

I agree that black paint and ROHS PCBs are a lot easier to make. Marketing does the rest if real innovation (once a HP motto) is absent.

[exe]

As far as I understand, we don't know yet what was done and how much effort was put into it. Although, yeah, seems weird no wifi lan in default configuration. Can it be added as an option by replacing a communication card?

I don't think "current generation" cannot do it. To me it seems a matter of demand. I won't be surprised if most sales are in gov/mil areas.

[LazyJack]

I'M pretty sure they did the market research and concluded that none of these would sell as many additional units, that would warrant the development cost. I'm also sure they did the math and figured that making the black edition will recover the costs.
This is such a niche product, that I don't think anyone outside the metrology industry would have any idea on what new features are needed or economical to implement.

[Henrik_V]

Be shure KS asked around in the metrology community about new features for a future 3458B version.
However, as already noted: this community is conservative ( add very,more or most) ...
(VFD is known to work for 20+ years ... how about OLED? , as long as I can get spare parts: no problem*)
On my list: better sync between instruments .. (RefClock IN/out, have 2+ ADC boards in 'perfect' sync) , ethernet


*) I like my V543 Nixi DMM I'm able to read without glasses, even in 20 years

[Wolfgang]

Not sure everything was done so well-planned and deliberately.

Pobably more like this:

1) Sales: Shit. We cannot sell our 3458A anymore to Europe because of this bloody ROHS stuff.
2) Managemeent to engineers: Whats the cheapest way to get around that ?
3) Engineers to management: We can leave all circuits as they are, but we need to make new PCBs and buy lead-free components
4) Management: OK, do that. And paint it black so it at least looks different and new.
5) Engineers: done.
6) Management to Marketing: Make a campaign about the "new" 3458A and explain why "no change" is a lot better than improved specs, a better user interface or long missing interfaces
7) Markting: Groovy ! Almost no investment, no improvement, still looking as new. Lets write a few journal articles that make us look smart.

[dr.diesel]

As funny as that is, and probably more correct than we'd like to admit, there is also don't fix what ain't broke!

If the limited customer base is happy, then just do what needs to be done and move on.

Though I'd really like to have seen ethernet added.

[e61_phil]

I fully agree with Wolfgang, but I don't think the 3458A has a only a small group of customers. Most of the electronic design labs I know have, at least access to, a 3458A. Most of them aren't chasing the last ppm of absolute voltages. They use it to verify drift of circuits or measure extreme values, which are difficult with other meters.
In the time I worked in the university, we had one (new and then uncalibrated) which was used to verify all the 6.5 digit meters. I ask Agilent that time, what they think how often a calibration of the 3458A is needed for that job and the answer was: Never

[TiN]

What does REV? report for the black edition ?

REV 9,2

Does the XYZZY command still work ?

I see no cave here.

[niner_007]

those are small changes, improving the current and AC ranges on the other hand would be a better investment, who cares GPIB or not, disable the VFD display and just fully control it over GPIB 

[PartialDischarge]

The best engineering in the world took place from the 60s to the 80s, no one has the ***** to redesign the hardware of software of the 3458A . Even a organisation the size of Boeing can't design a new plane and get rid of the 737 frame jeez

[bsdphk]

Thanks!

This confirms that they have not touched the software.

To me that makes a lot of sense, in terms of getting to a point where they can sell new meters.

Updating the outboard side would be a major undertaking on its own, involving a fair bit of software archaeology, but it is not impossible.

Commercially it would be stupid to have that delay the RoHS hardware roll-out.

However, I think they will have to deal with it eventually, and therefore I would not be surprised if we see a 'B' upgrade, which only changes the outboard card.

Looking at it, I think you could fit both a USB and Ethernet connector in the GPIB cutout on the backside, thus making the A->B upgrade available to all ages of meters.

(And if they havn't found anybody to do it yet, my company does stuff like that for money :-)

[SilverSolder]

The best engineering in the world took place from the 60s to the 80s, no one has the ***** to redesign the hardware of software of the 3458A . Even a organisation the size of Boeing can't design a new plane and get rid of the 737 frame jeez

I think there is a lot of truth to that.   

 "Boldness has genius, power, and magic in it." -- Goethe

How many modern committee driven cross functional teams achieve boldness, genius, power, and magic?  We lost the plot somewhere along the line.

[e61_phil]

I absolutely don't agree. Your very focussed on that very old DMM. There isn't much room for improvements. The reference one would use today (30 years later) would still be the same. The linearity is also good enough for the needs one would have. It would be very hard to be a factor of two better than the old instrument and it doesn't make much sense to put a lot of effort in something for such a little improvement.

Airplanes are a complete different story. I fully understand to not change more than absolutely needed, to avoid the huge pain with compliance and so on.


There are many enigneering fields which still improve rapidly. Also in the region of measurement technology. RF stuff for example. Or bigger instruments like mass spectrometers and so on.

[Wolfgang]

I think the 3458A improvement potentials are obvious:

- 4ppm / 2ppm option (for a hefty price, probably). They could be painted in gold instead of black 
- modern GUI LCD (like the competitors, Fluke and Transmille)
- USB/LAN

IMHO they will move their ass only when challenged by a competitor with a product doing better for a lot less money.
Not for all, but for a lot of applications, the 34470A is a viable alternative even today.

In the meantime its like mainframe computers; they slowly die out, but you can still make a lot of profit from people than cannot leave the platform.

[iMo]

The next step is to go cryogenic - to keep the entire stuff at around -273K. Something like a single chip meter with all the dividers and shunts put into a small Dewar bottle. While kept cold the noise and TC drift and aging will be almost zero. The entire meter of 34401A box size.  It will come in a few (10-20) years, I bet.

[SilverSolder]

[...]  There isn't much room for improvements. [...]

I would challenge that notion.  -  How about a DMM with a built-in Josephson junction reference, for example.  The work would be to miniaturize it and make it viable.  That's what our ancestors would have done - you know, the guys that went to the moon, the ones that designed the airframes we still use - if they had our technology.  We are completely effete in comparison.

[iMo]

The next step is to go cryogenic - to keep the entire stuff at around -273K. Something like a single chip meter with all the dividers and shunts put into a small Dewar bottle. While kept cold the noise and TC drift and aging will be almost zero. The entire meter of 34401A box size.  It will come in a few (10-20) years, I bet.

Ok, the mankind gets lazy, no willingness to spend XX% of their GDP for the space or arms race..
Thus I should have been more conservative  - "It will come till end of this century, I bet"..

[PartialDischarge]

There are many enigneering fields which still improve rapidly.

I'm not referring to improvements, but to unique feats and leaps forward that won't be repeated again by comparison. Tektronix developed a 250ps pulse generator and a 1GHz analog scope in the 60s, the fastest highest flying plane was developed in the 60s (SR-71, *everything* is unique about this plane), the first high-bypass turbofan engine (same concept as still used today) was developed in the 60s, men were flying out there in the 60s, I could go on an on...  You can put all the ethernets and USB ports on a 3458, or increase a spec analyzer from 3GHz to 30GHz but these are not feats at all.

[Wolfgang]

Disagree. Not all is lost.
If you look at an 110GHz Keysight UXR scope or on a R&S FSWP phase noise analyzer, we *can* make landmark innovations if we want (and the marketing people let us).
The problem seems to be the bureaucracy around the inventors - they slow down everything and kill everything that does not stink like fast money.