Author Topic: New Keithley DMM6500 and now DAQ6510  (Read 60072 times)

Hexley, cozdas and 2 Guests are viewing this topic.

Online kj7e

  • Frequent Contributor
  • **
  • Posts: 722
  • Country: us
  • Damon Stewart
Re: New Keithley DMM6500 and now DAQ6510
« Reply #75 on: May 23, 2018, 12:37:42 am »
The DMM6500 specs are better matched and in many areas better than to the 34465A.  I love my 34465A, but if I was in the market for another 6.5 digit DMM, I think I would have to go with the Keithley, it just seems to be more meter with better specs for the price.
 

Online Kleinstein

  • Super Contributor
  • ***
  • Posts: 5763
  • Country: de
Re: New Keithley DMM6500 and now DAQ6510
« Reply #76 on: May 23, 2018, 12:52:39 am »
Before getting to excited about the Keithley meter, maybe someone should measure the noise / Allan deviation plot. Some extra low frequency noise like with the DMM7510 could disappoint some users who hope for 7 digit resolution via PC interface.

The DMM6500 looks like a response to the 3446x meters, to also offer the more graphical user interface at a lower price range than the 7510.
 

Online lukier

  • Supporter
  • ****
  • Posts: 596
  • Country: gb
    • Homepage
Re: New Keithley DMM6500 and now DAQ6510
« Reply #77 on: May 23, 2018, 02:49:08 am »
The new DMM6500 has digitizing capabilities (1 MS/s) so it should be compared to Keysight 34465A, which costs more than 34461A (about 250$ USD).
Additionally, DMM6500 base model includes digitizing feature as well as 7 M readings memory; Unfortunately, both these features are paid options for KS 34465A.

So, unless KS revises 34465A pricing, DMM6500 is a real “killer” in the 6.5 digit bench DMM segment.

Ok, but what if you don't need a fast digitiser?

Well the meter is as you've mentioned on par with 34461A price-wise, so you get the digitizer and 7M sample memory for free. Also, what was a big deal for me, not only it supports the scanner cards but the old common ones (2001-TCSCAN). Also, I believe the software (Kickstart vs Benchvue) is cheaper and even free for now (beta version - 90 day trial).

Before getting to excited about the Keithley meter, maybe someone should measure the noise / Allan deviation plot. Some extra low frequency noise like with the DMM7510 could disappoint some users who hope for 7 digit resolution via PC interface.

I should be getting one soon so I can report my experience here. I'll try to capture Allan Deviation as well.
 
The following users thanked this post: cyr, thm_w

Offline sstepane

  • Regular Contributor
  • *
  • Posts: 104
  • Country: ua
Re: New Keithley DMM6500 and now DAQ6510
« Reply #78 on: May 23, 2018, 02:54:57 am »
The new DMM6500 has digitizing capabilities (1 MS/s) so it should be compared to Keysight 34465A, which costs more than 34461A (about 250$ USD).
Additionally, DMM6500 base model includes digitizing feature as well as 7 M readings memory; Unfortunately, both these features are paid options for KS 34465A.

So, unless KS revises 34465A pricing, DMM6500 is a real “killer” in the 6.5 digit bench DMM segment.

Ok, but what if you don't need a fast digitiser?

Well the meter is as you've mentioned on par with 34461A price-wise, so you get the digitizer and 7M sample memory for free. Also, what was a big deal for me, not only it supports the scanner cards but the old common ones (2001-TCSCAN). Also, I believe the software (Kickstart vs Benchvue) is cheaper and even free for now (beta version - 90 day trial).

Before getting to excited about the Keithley meter, maybe someone should measure the noise / Allan deviation plot. Some extra low frequency noise like with the DMM7510 could disappoint some users who hope for 7 digit resolution via PC interface.

I should be getting one soon so I can report my experience here. I'll try to capture Allan Deviation as well.

Also there's web server inside (LXI) with full remote control and data download. Plus quite extensive scripting engine (triggers, etc.) and free script builder software (even though its bulky like 1.5+ Gb).
I have not seen trend/chart feature on 34xxx series, but on dmm6500 it's very handy and quite useful (auto-scaling, touch screen zoom and stuff, triggers).
Only drawback for the moment is somewhat strange capacitance measurement - rarely measures caps in circuit, have no idea why.
 
The following users thanked this post: kj7e

Online kj7e

  • Frequent Contributor
  • **
  • Posts: 722
  • Country: us
  • Damon Stewart
Re: New Keithley DMM6500 and now DAQ6510
« Reply #79 on: May 23, 2018, 03:14:21 am »
The web UI on the 3446x series is almost useless since Java support has been depreciated where as I use the web UI on the 7510 often.  The 34465 trend plot is very nice and well laid out and easier to use than the 7510 IMO, but the 7510 is much more powerful.  I suspect the 6500 and 7510 UI and web UI are nearly the same.

I'm curious what the Factory Calibration costs for the 6500, I know Keithley wants nearly $400 for the 7510.  For comparison I had my 34465A calibrated by Keysight for under $200.
 

Offline pmcouto

  • Supporter
  • ****
  • Posts: 95
  • Country: pt
Re: New Keithley DMM6500 and now DAQ6510
« Reply #80 on: May 23, 2018, 05:40:23 am »
Only just saw this.
Interesting.
The 34461A is a few bucks cheaper here in Oz, but basically the same price. See it's not killing on the price front.
Another spec vs spec and feature vs feature debate, might be a hard to make a general call on this?

The new DMM6500 has digitizing capabilities (1 MS/s) so it should be compared to Keysight 34465A, which costs more than 34461A (about 250$ USD).
Additionally, DMM6500 base model includes digitizing feature as well as 7 M readings memory; Unfortunately, both these features are paid options for KS 34465A.

So, unless KS revises 34465A pricing, DMM6500 is a real “killer” in the 6.5 digit bench DMM segment.

Ok, but what if you don't need a fast digitiser?

You are absolutely right – If fast digitizer is not needed, KS 34461A is a slightly cheaper comparable instrument.
We should also consider that 34461A is the “de facto” replacement for the industry standard 34401A.

However, specs and feature wise, DMM6500 seems to be a superior DMM with an excellent price/features/performance ratio.
Keysight’s DMM product managers are certainly not comfortable with Keithley/Tektronix pricing strategy for this meter...  >:D
 

Offline hwj-d

  • Frequent Contributor
  • **
  • Posts: 554
  • Country: de
  • The economy of capitalism has NO social mission!
Re: New Keithley DMM6500 and now DAQ6510
« Reply #81 on: May 23, 2018, 06:14:43 am »
First things first: which referenz is inside?   :-//
(yes, i assume lm399...)
 

Online kj7e

  • Frequent Contributor
  • **
  • Posts: 722
  • Country: us
  • Damon Stewart
Re: New Keithley DMM6500 and now DAQ6510
« Reply #82 on: May 23, 2018, 06:17:29 am »
First things first: which referenz is inside?   :-//
(yes, i assume lm399...)

Based on specs, I strongly suspect LTFLU-1ACH, but maybe not heated.  Keithley/Tek/Fluke use the LTFLU over the LM399 for this class of spec.

Edit: Bit it could be a well aged and selected LM399.
« Last Edit: May 25, 2018, 12:59:53 am by kj7e »
 
The following users thanked this post: hwj-d

Offline hwj-d

  • Frequent Contributor
  • **
  • Posts: 554
  • Country: de
  • The economy of capitalism has NO social mission!
Re: New Keithley DMM6500 and now DAQ6510
« Reply #83 on: May 23, 2018, 06:33:04 am »
First things first: which referenz is inside?   :-//
(yes, i assume lm399...)

Based on specs, I strongly suspect LTFLU-1ACH, but maybe not heated.  Keithley/Tek/Fluke use the LTFLU over the LM399 for this class of spec.
Yes.  :)

Someone wrote in the forum:
https://www.eevblog.com/forum/metrology/the-ltflu-(aka-sza263)-reference-zener-diode-circuit/msg608846/#msg608846
Quote
I'ts now obvious to me that it will be more complicated to get good results with the LTFLU than with the LTZ1000, so why bother? I'll quote what I heard in an interview with a guy who's interest in life was old English sports cars, on why he liked tinkering with them: "They give you such interesting problems to solve!".
;D
« Last Edit: May 23, 2018, 06:37:26 am by hwj-d »
 

Online lukier

  • Supporter
  • ****
  • Posts: 596
  • Country: gb
    • Homepage
Re: New Keithley DMM6500 and now DAQ6510
« Reply #84 on: May 25, 2018, 12:39:59 am »
OK I got the meter yesterday and so far I find it quite impressive. UI wise it looks & feels like all the more expensive devices from modern Keithley lineup that people reviewed so far. Web interface is very nice as well.

I'm planning to do a mini-teardown soon (maybe this weekend if I have time) and write a short review after some getting more hands-on experience with the device.

For now I attach ~ 1h log of shorted inputs, AZ & line sync enabled, 3 NPLC (in the manual they recommend 1-5 NPLC range for the lowest noise). The meter was warmed up and the temperature should be relatively constant, slightly below the temperature it was calibrated at.

Maybe someone can generate an Allan Deviation plot (I'm playing with that as well at the moment).
 
The following users thanked this post: EEVblog, kj7e, hwj-d, MikeP, adambacon

Online maxwell3e10

  • Frequent Contributor
  • **
  • Posts: 350
  • Country: us
Re: New Keithley DMM6500 and now DAQ6510
« Reply #85 on: May 25, 2018, 12:46:08 am »
Thanks. I can plot in comparison to DMM7510. What voltage range was the data taken on?
 

Online lukier

  • Supporter
  • ****
  • Posts: 596
  • Country: gb
    • Homepage
Re: New Keithley DMM6500 and now DAQ6510
« Reply #86 on: May 25, 2018, 12:50:15 am »
Thanks! It was auto ranged so the lowest - 0.1V
 

Online kj7e

  • Frequent Contributor
  • **
  • Posts: 722
  • Country: us
  • Damon Stewart
Re: New Keithley DMM6500 and now DAQ6510
« Reply #87 on: May 25, 2018, 12:57:43 am »
Thanks! It was auto ranged so the lowest - 0.1V

Could you do one more but set the range to manual 10v.
 

Online maxwell3e10

  • Frequent Contributor
  • **
  • Posts: 350
  • Country: us
Re: New Keithley DMM6500 and now DAQ6510
« Reply #88 on: May 25, 2018, 01:08:50 am »
Here is the plot for 100 mV comparison of DMM7510 and DMM6500. So it seems a bit of a disappointment that the noise with shorted inputs is not the same, as in 34465A vs. 34470A. Also the autozero bump in Allan deviation is still there.
The rms noise on 100 mV setting is 1.1 ppm for 3 nplc, which is still pretty good. For comparison, DMM7510 has about 0.35 ppm and 34465A has about 1.9 ppm. 
« Last Edit: May 25, 2018, 01:28:53 am by maxwell3e10 »
 
The following users thanked this post: EEVblog, chickenHeadKnob, thm_w, lukier, kj7e, hwj-d, MikeP, adambacon

Online lukier

  • Supporter
  • ****
  • Posts: 596
  • Country: gb
    • Homepage
Re: New Keithley DMM6500 and now DAQ6510
« Reply #89 on: May 25, 2018, 01:10:16 am »
Could you do one more but set the range to manual 10v.

OK, I'll try to capture that in the evening.
 

Online TiN

  • Super Contributor
  • ***
  • Posts: 3953
  • Country: tw
  • xDevs.com/live - 24/7 lab feed
    • xDevs.com
Re: New Keithley DMM6500 and now DAQ6510
« Reply #90 on: May 25, 2018, 01:57:14 am »
C'mon, let's crack the hood already and see who lives inside. Perhaps LTZ pimping await for brave souls. :D
YouTube | Chat room | Live-cam | Have documentation to share? Upload here! No size limit, firmware dumps, photos.
 

Online lukier

  • Supporter
  • ****
  • Posts: 596
  • Country: gb
    • Homepage
Re: New Keithley DMM6500 and now DAQ6510
« Reply #91 on: May 25, 2018, 02:03:35 am »
C'mon, let's crack the hood already and see who lives inside. Perhaps LTZ pimping await for brave souls. :D

I'll try to do that over the weekend :) I'll contact you when I'll have the photos.

Regarding the reference, the specs are very similar to other LM399 meters, so that would be my first guess. People say LTFLU-1ACH, but I highly doubt it, it would have better specs and noise-wise too. Also, Fluke 8846A uses LM399 AFAIR, despite being Fluke.

We shall see soon.
 
The following users thanked this post: TiN, hwj-d

Online lukier

  • Supporter
  • ****
  • Posts: 596
  • Country: gb
    • Homepage
Re: New Keithley DMM6500 and now DAQ6510
« Reply #92 on: May 25, 2018, 08:14:15 am »
OK, here's the 1h short on the 10V range. Other settings are the same: 3 NPLC, AZ, Line Sync, no math.

It looks quite good, StdDev of 1.66 uV on 10V is 0.166 ppm.
 
The following users thanked this post: kj7e, maxwell3e10

Online maxwell3e10

  • Frequent Contributor
  • **
  • Posts: 350
  • Country: us
Re: New Keithley DMM6500 and now DAQ6510
« Reply #93 on: May 25, 2018, 11:48:41 am »
Here is the comparison of the noise with other meters on 10V scale. It is OK but not great. They are clearly saving money on the analog hardware. It would be interesting to see what kind of ADC is used.

For comparison 34465A has about 0.075ppm noise on 10V range at 3 nplc and DMM7510 has roughly 0.066ppm noise at 3 nplc.



P.S. It looks like DMM6500 is fairly closely matching in noise the 34461A, see
https://www.eevblog.com/forum/metrology/dmm-adc-noise-comparison-testing-project/msg667540/#msg667540
« Last Edit: May 25, 2018, 01:38:40 pm by maxwell3e10 »
 
The following users thanked this post: lukier, kj7e, hwj-d, adambacon

Online lukier

  • Supporter
  • ****
  • Posts: 596
  • Country: gb
    • Homepage
Re: New Keithley DMM6500 and now DAQ6510
« Reply #94 on: May 25, 2018, 05:37:33 pm »
P.S. It looks like DMM6500 is fairly closely matching in noise the 34461A, see
https://www.eevblog.com/forum/metrology/dmm-adc-noise-comparison-testing-project/msg667540/#msg667540

Interesting, I thought (based on Dave's teardowns) that 344xx are all the same with just some parts added to support extra functions (like the uCurrent range or 10A) and various grades of references or LTZ1000 and of course software crippling.

If the noise on 34461A is very different from 34465A that would suggest a completely different design.

Regarding the DMM6500 I've listed various pros already, mostly modern UI, scripting and interfaces, with all the capabilities they bring, plus the scanner card support.
Now some first cons:
- banana jacks don't look copper! maybe it is some silver plating?,
- also they are quite deep even for this safety-jacks category, I have a bunch of banana plugs with spring loaded safety shroud and they pop out immediately,
- 4.5W standby power and soft power switch (with a standby LED to make it worse),
- transformer hum in standby,
- the probes supplied look and feel quite crappy, PVC cable, not very flexible handles.
 

Online Kleinstein

  • Super Contributor
  • ***
  • Posts: 5763
  • Country: de
Re: New Keithley DMM6500 and now DAQ6510
« Reply #95 on: May 25, 2018, 05:44:20 pm »
The curve for the DMM6500 also seem to include some extra low frequency noise. Not as obvious as with the 7510 but still there.
So the ADC in the 6500 is higher noise than the 34465, even without the extra LF noise. So noise wise the ADC might be more like the 34461 that is at a similar price point.

However the earlier curve for the 100 mV range showed that the input amplifier is rather low noise - not as good as with the 7510, but still better than the 34465, even with the odd extra LF noise. The increased LF noise also in the 100 mV range suggests that the AZ mode also includes the full input amplifier like with most Keysight meters and different from the old Keithley meters (e.g. 2000,2001, 2002), that use a separate buffer upfront.

For the ADC used, I don't think they get that performance from a SD converter chip or similar - so it is likely a multislope converter of some kind likely in combination with a ADC chip (fast SD converter) for the fast data rates (like more than 1000 SPS) and maybe AC conversion.
 

Offline AG7CK

  • Regular Contributor
  • *
  • Posts: 105
  • Country: th
Re: New Keithley DMM6500 and now DAQ6510
« Reply #96 on: May 28, 2018, 07:31:36 am »
I would like to know if anyone has any experience with:

Kickstart 2.0 (priced) vs. older Kickstart (free): Will this instrument run under Version 1.7.0 and 1.9.8 ?

TSP / Test Script Builder - What can be programmed (without buying accessories) and how?

2000-SCAN Card - Would it be possible to DIY anything based on documentation / reverse engineering from the K2000 family.

KTTI-RS232, KTTI-GPIB, KTTI-TSP - Does the instrument have a "Talk-Only" mode so that one could sniff the data for use on an external monitor ? (instead of paying hundreds of dollars for a simple RS232-Card)

"Big Digit" (Large Digits) mode: Can the display be customized / composed by "programming"?




Anything else "new and modern" (apart from the UI and Digitizer)? [I agree with the posters above who emphasizes the "gadget"-style of this box. I have a handful of older (stable and agreeing) 6.5 / 7.5 digit voltmeters / dataloggers, so I am not worried about noise- and/or drift qualities from this meter compared to "old-fashioned" HP-type meters ... ]
 

Online lukier

  • Supporter
  • ****
  • Posts: 596
  • Country: gb
    • Homepage
Re: New Keithley DMM6500 and now DAQ6510
« Reply #97 on: May 28, 2018, 08:51:05 pm »
I would like to know if anyone has any experience with:

Kickstart 2.0 (priced) vs. older Kickstart (free): Will this instrument run under Version 1.7.0 and 1.9.8 ?

Sorry, Linux here so I didn't even bother to download Kickstart.

TSP / Test Script Builder - What can be programmed (without buying accessories) and how?

I just got the meter, so I haven't tried TSP scripts but I thought these are just text files with LUA, so I have no idea about any Test Script Builders or buying accessories for that. TSP scripts are definitely on my list of things to try - like using the digitizer to capture the MCU's current consumption and then calculate an intergral over that to estimate the power usage. I was also thinking that, with the 1 MSPS digitizer, having a FFT function would be nice to see dunno, the noise from simple < 100kHz DC-DC switchers. But without the FFT function built-in I guess implementing it purely in LUA will be a waste of time as it would be very slow I guess.

2000-SCAN Card - Would it be possible to DIY anything based on documentation / reverse engineering from the K2000 family.

I have 2001-TCSCAN card and it works with the meter. The nice thing is that one can program DMM functions, with all their settings, per channel so when switching everything is already set.

Regarding the DIY option it is definitely doable, I was planning to do that, the schematics are there, this article from TiN is helpful as well:
https://xdevs.com/review/kei2001tscan/

and I guess making the non-TCSCAN version is even cheaper as one does not need expensive AD590MH temperature sensor and machine the copper block. The rest is just a bunch of TQ2E-L2-5V relays, terminal blocks, MIC5841 drivers and some other bits and bobs.

In the end, just after I finished redrawing the schematic in KiCAD and before I started doing the PCB layout I managed to score 2001-TCSCAN cards on eBay for around 120 GBP, so the project is abandoned.

KTTI-RS232, KTTI-GPIB, KTTI-TSP - Does the instrument have a "Talk-Only" mode so that one could sniff the data for use on an external monitor ? (instead of paying hundreds of dollars for a simple RS232-Card)

No idea, it has USB and Ethernet by default, so I'm actually glad that the archaic/arcane interfaces are separate paid options, so it doesn't increase the base price of the meter. If somebody needs to replace 34401A or K2000 in a production system then of course they'll have to buy KTTI-GPIB or similar, but for me personally I don't care - Ethernet FTW.

"Big Digit" (Large Digits) mode: Can the display be customized / composed by "programming"?

I haven't found such option yet. What customisation you have in mind?

Anything else "new and modern" (apart from the UI and Digitizer)? [I agree with the posters above who emphasizes the "gadget"-style of this box. I have a handful of older (stable and agreeing) 6.5 / 7.5 digit voltmeters / dataloggers, so I am not worried about noise- and/or drift qualities from this meter compared to "old-fashioned" HP-type meters ... ]

It was similar for me, I have a bunch of other, older meters, some more stable, some with more digits, but I wanted a modern one with all the bells & whistles.
 
The following users thanked this post: AG7CK

Offline cyr

  • Frequent Contributor
  • **
  • Posts: 250
  • Country: se
Re: New Keithley DMM6500 and now DAQ6510
« Reply #98 on: May 28, 2018, 10:38:04 pm »
TSP / Test Script Builder - What can be programmed (without buying accessories) and how?

Have a look at the programming section(s) of the manual:

https://www.tek.com/tektronix-and-keithley-digital-multimeter/dmm6500-manual/model-dmm6500-6-1-2-digit-bench-system-0

Some sample code here:

http://www.farnell.com/datasheets/2607097.pdf?_ga=2.109722629.1255162927.1522665742-1348294533.1519212668

You shouldn't need anything more than a text editor and USB stick or a network connection, although I'm sure "Test script builder" is helpful when developing...
 
The following users thanked this post: AG7CK

Online kj7e

  • Frequent Contributor
  • **
  • Posts: 722
  • Country: us
  • Damon Stewart
Re: New Keithley DMM6500 and now DAQ6510
« Reply #99 on: May 29, 2018, 04:41:12 am »
C'mon, let's crack the hood already and see who lives inside. Perhaps LTZ pimping await for brave souls. :D

I'll try to do that over the weekend :) I'll contact you when I'll have the photos.

Regarding the reference, the specs are very similar to other LM399 meters, so that would be my first guess. People say LTFLU-1ACH, but I highly doubt it, it would have better specs and noise-wise too. Also, Fluke 8846A uses LM399 AFAIR, despite being Fluke.

We shall see soon.

 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf