Keysight's new 34465A (6.5 digit) and 34470A (7.5 digit) bench multimeters

[hopefully_reliable]

I have just come across these through the 'find a part' section of the Keysight website, in particular
http://www.keysight.com/my/faces/fapHomePage.jspx?productNumber=34465A
http://www.keysight.com/my/faces/fapHomePage.jspx?productNumber=34470A

A quick google later and I found the prices on the Brazilian Newark
http://www.farnellnewark.com.br/searchresults.aspx?page=25&sort=0&idproducttype=0&idproductcolor=0&iddept=0&nupricerangestart=1&nupricerangeend=9999999&type=1&dskeyword=agilent

(conversions using xe.com)
34465A - R$ 11,304.49 - US$ 3,995.37 - £2,588.80
34470A - R$ 23,459.78 - US$ 8,290.73 - £5,372.45

EDIT: As a few people have pointed out, these prices are incredibly high. Expected prices (from Dave's reply - #21) are
34465A - around $1500
34470A - less than $3000

and also a product fact sheet http://www.tequipment.net/assets/1/7/5991-2110EN.pdf - which I've also attached, as I'm guessing it might disappear.

The differences over the 34461A seem to be

• better one year accuracy - for DC volts 30 ppm (65A) and 16 ppm (70A) vs 35 ppm (34461A)
• higher reading rate - up to 50,000 readings/s
• larger memory - up to 2 million readings
• 1 uA (and presumably 10 uA) DC ranges
• 1 GOhm range
• support for thermocouples
• capacitance measurements (1 nF to 100 uF)
• dual line display

The higher measurement speed and the larger memory are both options (which is a bit of a shame).

[quarks]

nice find, have never seen these before

[EEVblog]

Wow, 1uA current range   
And added cap measurement too.
Obviously designed to compete with the new Keithley DMM7510

[Maxlor]

More detailed specs. If anyone needs to make space and throw out their 34461A, I'll take it 

[Dr. Frank]

Strange strategy of the HP successor company...

These new instrument have some slight improvements over the new (budget) 34461A, but especially in relation to these new features real premium prices. (Some more memory, 2 additional current shunts, slightly better volt reference..)
Obviously, they are going back to their roots.. in former times HP stood for "High Price".

The 7 1/2 digit 34470A still has got no LTZ1000 reference inside, very obviously.
This would make absolutely sense at that level of resolution.
And also, it is as costly as the 3458A..

I wonder, which kind of volt reference they use; I doubt that an LM399H can further be selected for 16ppm/yr. stability.

I'm waiting for the complete specs, if these new instruments really make sense at these prices.


OK, there are the specs.

They claim additionally an "AUTOCAL" for these instruments.
But that function is not as powerful at all as in the 3458A!
The INL of the A/D is 2ppm and 1.5ppm only, and therefore not capable of performing a complete range autocal, as in the 3458A.
This may only compensate some T.C. drifts of the components, nothing more.

Disappointing for that price.


And there are some bugs in the specs, concerning memory depth and reading speed:

I assume, they offer optional 50000 rdgs/s maximum , but 50000 or optionally 2Mio readings memory depth (not 2Mio/s reading rate).

The 10µA and 1µA ranges are realized by higher amplification on the same 100µA shunt..as the burden goes down by factors of ten, each. Maybe they copied Daves µCurrent circuitry?

I doubt, that  they can really guarantee the same accuracy on all three ranges... spec is "typical" values for these ranges.


Frank

[kwass]

Wow, 1uA current range   
And added cap measurement too.
Obviously designed to compete with the new Keithley DMM7510

Looks like they're going to release firmware for the 34460A/61A to add capacitance measurement too.  Either that or they made a major mistake in the new spec sheets.

[6thimage]

Wow, 1uA current range   
And added cap measurement too.
Obviously designed to compete with the new Keithley DMM7510

Looks like they're going to release firmware for the 34460A/61A to add capacitance measurement too.  Either that or that made a major mistake in the new spec sheets.

They are due to release a new firmware version for the 34461A on 1st March (version 2.06). I have got a pre-beta version, but they have asked me to keep quiet on any additions they have added. However, they are increasing the version number from 1.10 to 2.06, so you can make what you want of that ...

[6thimage]

I wonder, which kind of volt reference they use; I doubt that an LM399H can further be selected for 16ppm/yr. stability.

Could thermally shielding the LM399, or placing it in an additional oven, significantly help with the long term stability?

The 34461A has a fan at the back which blows air directly across its LM399 voltage reference, which I've always thought was a little odd. Whilst the 34465A and 34470A both have this same fan, might they have boxed it up?

[dr.diesel]

Here's to hoping Dave ends up with one of each in the next few days!   

[robrenz]

34465A - R$ 11,304.49 - US$ 3,995.37 - £2,588.80

•better one year accuracy - for DC volts 30 ppm (65A)


8846A is 24ppm 1 year DC for $1500.00   

[dr.diesel]

34465A - R$ 11,304.49 - US$ 3,995.37 - £2,588.80

Yeah but they changed their name, they've got a couple million new stickers to pay for..     

[rodpp]

This price is in Brazil, normally much more expensive.

In US or Europe probably it´ll be cheaper.

For example, in the same brazilian website the 34461A is R$5,798,34 or US$2,050.

[KA3YAN]

I'm totally fine sticking with my Fluke 8846A.  It might not have the snazzy color screen, but the specs are the same or superior in most aspects.  It doesn't have the extraordinarily fast max reading rate (only a mere 1000 per second) but hey, you can't have EVERYTHING.   

[ivan747]

This price is in Brazil, normally much more expensive.

For some context, iPhones used to be sold officially in Brazil for almost twice the price. Playstation 4 launch price? US$1400 vs. US$400 on Amazon. Brasilians often complain about this here in the EEVBlog forum. It's because of import taxes  It's so bad, actually, that Apple started manufacturing iPhones in Brazil just to stay competitive in that market.

[EEVblog]

And also, it is as costly as the 3458A..

Rumor has it they were working in a 3458A replacement, but this doesn't seem to be it. The same price is a bit crazy though.

The 10µA and 1µA ranges are realized by higher amplification on the same 100µA shunt..as the burden goes down by factors of ten, each. Maybe they copied Daves µCurrent circuitry?

I was thinking it's likely to be an amp as well.

[EEVblog]

Here's to hoping Dave ends up with one of each in the next few days!   

Haven't heard a thing.
I'm getting the new Keithley one though, that will be interesting to look inside.

[Vgkid]

When I saw the price for the 34470A, I thought"mighty close to the pricepoint for the 3458A".
for the 34470A They would need to drop the price by half, in order to match the newest Keithley.

[HighVoltage]

The 34470A looks interesting but for that price I think I really would buy a good working 3458A

[es]

I'm guessing the pricing should be around $1500 and $3100 for the 34465A and 34470A respectively.

If you look at the price of other Keysight gear at the Brazilian site in the first post, the conversion rate is about 8 times.

For example, the MSOX3104T is listed at R$ 118.353,46 while it's $15,400 in the US.

[rodpp]

The conversion rate in that site is, let me say, unreliable.

DSOX3014A - R$12,821.78 - US$3,950
MDO3014 - R$13,949.20 - US$3,990
34461A -R$5,798.34 - US$1.100
Fluke 87V - R$2,490.00 - US$340
U1733C - R$1,099.56 - US$432

US$ prices from tequipment.com

So, it ranges from 2.5 to 8.

My guess is that the conversion rate will be similar to the 34461A, about 5.

[LaurentR]

Is it me or that "peak" function on this screenshot doesn't exist in the current firmware?

There is also the "Time" option instead of just NPLC.

[EEVblog]

I have it on very good authority that those prices are completely wrong.
The 34465A will be around US$1500
The 34470A will be <US$3000
Options will cost extra, in the order of hundreds of dollars.
And yes, I might be able to get one soon for teardown.

[LaurentR]

Another note: the 34465A's DCV specs are given as 30ppm (reading), but only for Cal Temp +/-2C vs. +/-5C. So without ACAL, it actually has worse specs than the 34461A outside of the +/-2C range.

So the improved baseline accuracy of the 34465A vs. 34461A is more creative accounting than anything else. ACAL improves the temp coefficient, but the base spec is basically the same.

The 34470A quotes the DCV spec at +/- 5C.

[TiN]

Dave, since you have bunch of 6.5+ meters already, might consider to have comparison between them in same conditions, using DC standard and foil resistors.
Something like you did with handhelds, roundup of bench meters, since there is no such material on web usually.
Your MV106 should be stable within 10ppm if ambient temperature kept +-2°C.

Interesting to know, that Agile...doh..Keysight offers voltage reference PCB (if PCA stands for PCB Assembly) for $334 USD.
This price is enough to allow using LTZ1000 with few VPG foils

[hopefully_reliable]

I have it on very good authority that those prices are completely wrong.
The 34465A will be around US$1500
The 34470A will be <US$3000
Options will cost extra, in the order of hundreds of dollars.
And yes, I might be able to get one soon for teardown.

I thought they were high, but I didn't check the prices for known products - sorry everyone. But I have changed the first post to include your figures, so no one gets confused.