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

[6thimage]

First impression of the 34465A is that it's more compact than my Keithley 2000, and the display is bright, clear, and has a wide viewing angle. Nice.

That sounds exactly like the 34461A.

Auto-ranging, on the other hand, is so much slower it's not even funny. One of the things I really like about the K2000 is that I can apply a signal, and by the time I've looked up at the meter, it's already found the correct range and showed a measurement. The 34465A is much more like a handheld meter, really quite slow, and that's disappointing. Maybe there's a setting that'll speed it up?

Really? I find that the 61A has a very similar auto-ranging speed to the 2000, I can't see why it would take longer on the 65A.

Edit: Autoranging will take longer if you are on the larger apertures (as the measurements take longer). For normal use (probing different signals) there is no need to use the higher apertures (e.g. 100 PLC).

There's also an annoying relay that clicks away if the meter is left in high-Z voltage mode. As the inputs charge and discharge, the meter switches range and back to accommodate the instantaneous voltage. Maybe I'll just have to leave it in 10M mode instead. It's a shame, but the K2000 doesn't have this quirk either.

You shouldn't leave it in the high-z mode - from the help menu (press and hold the high-z soft button), the 10 MOhm is what they expect you to use all the time, with the high-z being there if you need it for a particular purpose (which is rare). But having said that, the charge accumulation (and the relay dumping it when it gets to -12V) will not happen when you are making a reading, it only happens when the inputs are floating. Also, for what its worth, turning off auto-zero decreases the charging time.

Edit: It might be worth resetting your settings to the default (press shift, then the run-stop button).

[6thimage]

Any one know if the software update for the '61A will include the ability to log direct to a USB stick?  That is one feature that is surprisingly missing.

Unfortunately the 34461A's new firmware will not allow it to log to USB memory or the internal memory - these are the extra features that have been added to the 65/70A. Whether Keysight will release an option in the future that brings the digitising functionality, all be it in a limited way compared to the newer models, is another thing. The firmware is shared between all 4 devices (60/61/65/70A), so it should be possible but whether they will ever do it is another thing.

If we get enough people emailing Keysight's contact centre asking for data logging functionality in the 61A, they might add it at some point as an option.

[LaurentR]

Auto-ranging, on the other hand, is so much slower it's not even funny. One of the things I really like about the K2000 is that I can apply a signal, and by the time I've looked up at the meter, it's already found the correct range and showed a measurement. The 34465A is much more like a handheld meter, really quite slow, and that's disappointing. Maybe there's a setting that'll speed it up?

Really? I find that the 61A has a very similar auto-ranging speed to the 2000, I can't see why it would take longer on the 65A.

I find this weird to. Which measurement are you autoranging? Even with relay activation, I found the 34461A immensely faster at autoranging than any HH DMM I have. And the 34465A is supposed to be even faster then the 34461A. Per the datasheet, the autoranging time in DCV is <30ms for the 34461A and the autoranging time for all measurements is < 5ms (!) for the 34465A.

Note that the measurement itself may be slow. The reset value for all 3446x series is NPLC10 and autozero ON, so by default you get a measurement time of 0.33/0.4 sec, and that's noticeable (it certainly dwarfs the autoranging time). You can put the 34465A in NPLC 1 or .2 or .02, still get near-full accuracy and blindingly fast readings.

[6thimage]

Auto-ranging, on the other hand, is so much slower it's not even funny. One of the things I really like about the K2000 is that I can apply a signal, and by the time I've looked up at the meter, it's already found the correct range and showed a measurement. The 34465A is much more like a handheld meter, really quite slow, and that's disappointing. Maybe there's a setting that'll speed it up?

Really? I find that the 61A has a very similar auto-ranging speed to the 2000, I can't see why it would take longer on the 65A.

I find this weird to. Which measurement are you autoranging? Even with relay activation, I found the 34461A immensely faster at autoranging than any HH DMM I have. And the 34465A is supposed to be even faster then the 34461A. Per the datasheet, the autoranging time in DCV is <30ms for the 34461A and the autoranging time for all measurements is < 5ms (!) for the 34465A.

Note that the measurement itself may be slow. The reset value for all 3446x series is NPLC10 and autozero ON, so by default you get a measurement time of 0.33/0.4 sec, and that's noticeable (it certainly dwarfs the autoranging time). You can put the 34465A in NPLC 1 or .2 or .02, still get near-full accuracy and blindingly fast readings.

I think you have hit the nail on head there. If you have it in 100 PLC it will take considerably longer to autorange. By the sounds of it AndyC has been playing a lot with the settings, so it might be best to do a reset to the defaults (press shift then the run-stop button).

[AndyC_772]

Guys... long day at work followed by the arrival of a new toy. Of course I've been playing!

[6thimage]

Guys... long day at work followed by the arrival of a new toy. Of course I've been playing!

We don't blame you one bit, just remember you don't have to tweak all the settings in the same night

If you do want to just play with it, remember you can always reset it to the defaults when your done, or even set the multimeter to start up in its default state.

On a bit of side note here, you can save the multimeter's state and tell it to use it at start up, which I find really useful as you can change the trend chart so that it automatically fits the data.

[AndyC_772]

Had some more time to play now, and it's all good... mostly

Auto-ranging on the 2W resistance range does take a few steps to go from open circuit to correctly reading shorted probes, especially if they don't make really good contact straight away, and the meter tries some intermediate ranges before settling on the 100 Ohm range. It's OK if the aperture is set to 1 PLC, but on 10 PLC (the next available step up), it takes a while.

I spent a while yesterday looking for the red grabber, which I was sure should have been included in the test lead kit. I was going to give Keysight a call today to say it was missing, but having looked at the included accessories on their web site, it turns out you only get a black one (WTF?!)

So hooking the meter up to, say, log current for a period of time unattended is impossible unless you have a substantial enough clip already. And guess who wanted to leave it logging the charging process of a 12V lead-acid battery overnight 

(Seriously Keysight, that's just being cheap, even for you!)

On the plus side, setting it up to log to USB was dead easy and worked first time. I'm at a bit of a loss as to why anyone would pay for the MEM upgrade, when the 50,000 point limit of the standard meter doesn't apply if you're logging to a file on a USB stick. I checked this morning; 70,000 points taken at 3msec intervals, no problem at all.

Has anyone else noticed the noise from the fan? It's quiet, but quite high pitched and... whiney. Maybe it's inevitable, what with the fan being so small, but it's irritating nonetheless. Fortunately the radio drowns it out easily.

The probe tips needed a good clean in order to give consistent resistance readings; a good wipe with a paper towel soaked in Flux-off rosin flux remover worked wonders. They still feel a bit low rent compared to the TL175 set that came with my Fluke 289, though, and really hoped for better with a £900 multimeter. Some Kelvin clips for the 4W resistance range would have been nice too; you don't get anything at all to perform 4W measurements.

Amazingly, you do get a USB cable, but no Ethernet cable, and no printed manual. I guess I shouldn't be surprised.

Overall impression? From a usability point, a bloody good meter, as expected. I've no way to check its accuracy, though I've no reason to doubt it's anything other than superb. It's a shame that the accessories aren't more plentiful, but that's much easier to fix than any deficiency in the meter itself.

Also nice to see is that my Agilent 6632B set to 5V outputs +5.00008V, and in CC mode @ 1A, I get 0.99968A. Good enough for a general purpose bench supply 

[6thimage]

I spent a while yesterday looking for the red grabber, which I was sure should have been included in the test lead kit. I was going to give Keysight a call today to say it was missing, but having looked at the included accessories on their web site, it turns out you only get a black one (WTF?!)

Yeah they include a black grabber, and a red and green mini-grabber. I always assumed that the black grabber was meant for a ground connection. The mini-grabbers will hold onto quite a lot, so it might be worth trying them. I've never found the black grabber all that useful, as it only has a 90 degree bend in it. It would be more useful if they provided some grabbing hooks.

On the plus side, setting it up to log to USB was dead easy and worked first time. I'm at a bit of a loss as to why anyone would pay for the MEM upgrade, when the 50,000 point limit of the standard meter doesn't apply if you're logging to a file on a USB stick. I checked this morning; 70,000 points taken at 3msec intervals, no problem at all.

From what I've read, logging to USB is limited to 100 hours (which I'm told is a maximum of 360 million readings). I have been reliably informed that once you go above the 50,000 point limit, the trend chart reverts to the recent/all mode - essentially making it so you can't use the cursors or zooming controls. Also, the digitising mode only works with the internal memory, due to its speed, so people who use that might want the extra memory.

Has anyone else noticed the noise from the fan? It's quiet, but quite high pitched and... whiney. Maybe it's inevitable, what with the fan being so small, but it's irritating nonetheless. Fortunately the radio drowns it out easily.

The fan in the 61A isn't too bad, it's a little noisy, but I wouldn't say high pitched. You could always swap it out with another small fan if you really wanted to.

Is there any chance of you taking the cover off and snapping a few pictures? I'm sure there are a lot of people (like me) who want a glimpse of the reference and the changes they've made.

[LabSpokane]

Save yourself a lot of frustration and shitcan those supplied grabbers now. They are just flimsy, malformed crap. 

[AndyC_772]

I'm inclined to agree; I've seen worse, but they certainly don't feel like the quality items I'd expected.

The black one was a good fit for the 12V battery I charged last night. The mini-grabbers didn't look up to passing any current, so I used the Keysight to log voltage instead, and recorded current with my Fluke 289 instead. The Fluke's leads and accessories are much nicer.

[LaurentR]

Sold my Keysight lead set on EBay. Agree that the various Fluke sets are much nicer. The one thing the Keysight have for them is that they are seriously thick (16ga!), which make for a 15A rating and lower resistance than anything Fluke I have.

[radioFlash]

Here are some photos of my 34465A. It's using an LM399 voltage reference as expected.

[radioFlash]

And some more photos: Full size of the top side of the PCB and detail of of the top of the PCB towards the rear.

The PCB is already on Rev4. It looks a little messy around the input selector switch--doesn't look like they bothered to clean the flux.

[radioFlash]

And a photo of the voltage reference.

[6thimage]

Thanks radioFlash they are very useful - I'm guessing you've removed the shield which sits above the voltage reference, is there also a fan shield of some sorts?

I'm a little surprised to see the LM399, as I was under the impression that the 34465A and 34470A had the same reference.

Your board's revision is 4 the same as what is used in the 34461A, although it has a different model number. I originally thought the only difference would be the population of components, but yours has some extra footprints and some changed parts.

I've combined one of your images with one from Dave's teardown.

But the changes I can see are:

• populated relay at K201
• resistor at R324 (looks like a crystal, left of the LM399)
• U202 - AD8638 (auto-zero rail-to-rail op-amp)
• U203 - TL071C (JFET op-amp)
• sot-23 at U111
• resistor (R116) on the right hand side of U110
• resistor R318 left of K201
• change of package (& possibly precision) for R231 (in between relays)
• relays changed to Omron brand

Is there any chance you could take a picture of the back of the front panel?

[Dr. Frank]

radio Flash:

Thanks for these great photos!

Please make a side shot of R324, showing its imprints on both sides.
it's assumed to be a Vishay metal foil resistor, which would tell something about (better) Ohm stability and maybe the "Autocal" feature.

(Ohm stability 34461A: 80ppm/yr. // 34465 + 34470: 20ppm/yr.)

Frank

[HighVoltage]

On all my newer 34401A I see yellow transparent tape on the larger electrolytic capacitors (heat protection during production may be?).
Interesting, that we do not see this tape in either the 34460A or the 34465A boars.
Thanks for the great pictures.
 

[radioFlash]

Here's a photo with the shield in place. This is the only shield in the meter-there's not a separate one for the fan/regulator.

[KedasProbe]

On the plus side, setting it up to log to USB was dead easy and worked first time. I'm at a bit of a loss as to why anyone would pay for the MEM upgrade, when the 50,000 point limit of the standard meter doesn't apply if you're logging to a file on a USB stick. I checked this morning; 70,000 points taken at 3msec intervals, no problem at all.

Can you share a log file?
How many significant digits are there? 
(like this +3.31785468E+00,+3.61554515E+00)

Data log mode
The maximum reading rate is 1000 readings/s  (max. 100 hours)

"Digitizing" from what I read they mean fast reading like 5000 readings/s

[AndyC_772]

Here's a 10 sample log. Probes are connected to a 9V battery. Meter is set to the 10V range, acquisition time is set to 200us, and sample interval is set to 2ms, both of which are as short as I can get. If anyone knows how to sample at 1ms intervals, I'd be interested to know what I've missed.

Code: [Select]

Start date:,03/10/2015,Start time:,15:09:42.910
Sample interval:,0.002000
Reading #,Reading
1,+8.80413134E+00
2,+8.80414814E+00
3,+8.80423098E+00
4,+8.80415278E+00
5,+8.80414582E+00
6,+8.80419101E+00
7,+8.80415741E+00
8,+8.80417653E+00
9,+8.80417885E+00
10,+8.80416958E+00

Edit: here's another log with the meter set to 1 PLC acquisition time and 0.5 sec sample interval:

Code: [Select]

Start date:,03/10/2015,Start time:,15:18:38.130
Sample interval:,0.500000
Reading #,Reading
1,+8.80401932E+00
2,+8.80401939E+00
3,+8.80402039E+00
4,+8.80401911E+00
5,+8.80401959E+00
6,+8.80401877E+00
7,+8.80401866E+00
8,+8.80401888E+00
9,+8.80402165E+00
10,+8.80401964E+00


[6thimage]

Here's a photo of with the shield in place. This is the only shield in the meter-there's not a separate one for the fan/regulator.

That's not what I was expecting. The metal shield is part number 5041-5231 and is used on all four meters (34460/61/65/70A), but the Keysight part website mentions a fan shield (5041-5261) that is used on the 65/70A. I thought that this fan shield might be made from plastic to direct the airflow away from the voltage reference, but if it isn't fitted it makes me wonder why they list it as a part.

[6thimage]

Here's a 10 sample log. Probes are connected to a 9V battery. Meter is set to the 10V range, acquisition time is set to 200us, and sample interval is set to 2ms, both of which are as short as I can get. If anyone knows how to sample at 1ms intervals, I'd be interested to know what I've missed.

Code: [Select]

Start date:,03/10/2015,Start time:,15:09:42.910
Sample interval:,0.002000
Reading #,Reading
1,+8.80413134E+00
2,+8.80414814E+00
3,+8.80423098E+00
4,+8.80415278E+00
5,+8.80414582E+00
6,+8.80419101E+00
7,+8.80415741E+00
8,+8.80417653E+00
9,+8.80417885E+00
10,+8.80416958E+00


8 decimal places - this is the same as the CSV output from the 34460/61A. I'm not sure why you can't get 1 ms sample intervals - I can't find a reference to it in the manual.

[6thimage]

Here's a 10 sample log. Probes are connected to a 9V battery. Meter is set to the 10V range, acquisition time is set to 200us, and sample interval is set to 2ms, both of which are as short as I can get. If anyone knows how to sample at 1ms intervals, I'd be interested to know what I've missed.

I have just been informed that you need to change your settings a little to get to 1 ms intervals. If you having auto-ranging enabled, the minimum sample interval is increased so the meter can autorange and perform a measurement without over-running into the next sample period. Additionally, the auto-zeroing increases the time required for the measurement as well.

So to get a 1 ms sample interval either change to a fixed range, disable auto-zeroing or change to a smaller aperture.

[AndyC_772]

Auto zero doesn't seem to make any difference, but manual range selection does. Set the meter to 10V manually and I get 1ms intervals.

Code: [Select]

Start date:,03/10/2015,Start time:,16:29:47.790
Sample interval:,0.001000
Reading #,Reading
1,+8.80342110E+00
2,+8.80344253E+00
3,+8.80340430E+00
4,+8.80342110E+00
5,+8.80344485E+00
6,+8.80348772E+00
7,+8.80350684E+00
8,+8.80343326E+00
9,+8.80334289E+00
10,+8.80338808E+00


[Sparky]

Did anyone get the Digitize (DIG) option? I would love to hear what maximum sustained sample rate is possible when simultaneously transferring data to a PC with USB and LAN connections.