6.5 digit DMM: Keysight 34461A or Rigol DM3068 or... ?

[nctnico]

I may need a bit more precision than my current DMMs can provide. DC accuracy is what I'm after. I have been looking at some old threads but can't find a clear answer whether to buy a Keysight 34461A or Rigol DM3068. The DM3068 seems to have better specs but I'm wary about some software issues reported by some forum members.
Or am I missing a better solution in the same price range?

[blackdog]

Hi nctnico,

If you want some info about the 34461a you can mail me.
And after a email you can talk Dutch to me :-)

I have 2 pieces of the 34461A in my lab and a 3458a in my LAB.
Sorry No Rigol DM3068 to compare, yust 2 pieces of TEK DMM4050 and a 34401A...

Kind regarts,
Blackdog

[dadler]

I would love to see Dave review one of the Rigol DM series meters, in full-honesty mode.

Shahriar/TheSignalPath reviewed the DM3068 here: http://thesignalpath.com/blogs/2012/07/02/rigol-dm3068-6-5-digit-digital-multimeter-review/

I've been trying to compare the two meters you have listed myself, and the historical anecdotal 'evidence' on the forum is "why buy an unknown when the Agilent is not much more expensive and a known good product". A Dave-review would go a long way to assist us in choesing between these products. 

[EEVblog]

I would love to see Dave review one of the Rigol DM series meters, in full-honesty mode.

My guess is it's probably a reasonable meter.
But the Agilent one has better specced, and has a much better screen and that awesome trend plotting feature. And it's got the all-important pedigree.
I think the Agilent is the clear winner in the bench top market.
The Fluke is slightly better spec-wise, so might be more suitable for the voltnuts.
The Rigol would be suitable if you were pinching every last penny.

Shahriar/TheSignalPath reviewed the DM3068 here: http://thesignalpath.com/blogs/2012/07/02/rigol-dm3068-6-5-digit-digital-multimeter-review/, but he is sponsored by Rigol (not really doubting his objectivity, just when you are sponsored it's tough to be fully honest, even if it's subconscious).

I remember watching that and didn't find anything questionable.

and the historical anecdotal 'evidence' on the forum is "why buy an unknown when the Agilent is not much more expensive and a known good product". A Dave-review would go a long way to assist us in choesing between these products. 

I'm not sure it would.
Assuming the Rigol doesn't have any major problems, it's impossible for me to prove that it's going to be a solid long term meter.
The pedigree and trust that Agilent (and Fluke) have earned in their measurement products is a very real thing. And in this case the Agilent 3441A was the industry standard for decades. And there is nothing even remotely to give anyone the heebie-jeebies that the 34461A isn't a solid successor to that throne.

[EEVblog]

The DM3068 seems to have better specs

There are specs, and then there are specs.
You can bet your bottom dollar that Agilent and Fluke have done a LOT more homework and are more conservative in their specs than Rigol will be.
If trust in accuracy is what you are after, go Agilent or Fluke.

[rx8pilot]

For the small difference in cost, I would not even consider the Rigol. Agisight is a company that has always built test gear that meets the need and then price is set. Rigol is a company that is focused on creating the cheapest thing they can get away with. If the Rigol was $300 it may be a consideration, but there is not much difference for a professional.

If you really need that precision, buy the precision instead of hoping for it.

[VK5RC]

As a very happy 34461 owner,  when you are looking at those levels of accuracy,  the ability to take multiple measurements and average and get standard deviations easily is important.  The Agilent /Keysight user interface is excellent,  very intuitive. 

[TooOldForThis]

I have both of these DMMs.   Both of these devices have advantages over the other.  The display on the Agilent is wonderful (and has "Agilent" printed on the case).  The Rigol model has capacitance measurement, more thermal probe options, and dual measurements. 
Both devices (mine anyway) are way within their accuracy specs.   When I connect both units to the same signal, the resulting measurements are so close to each other it's just outright scarey.   

[robrenz]

IF ppm accuracy is your highest priority then the Fluke 8846A/TEK DMM4050 (same unit) is very good.
But the 34461A is a much more user friendly interface and probably accurate enough for most needs.

[EEVblog]

If the Rigol was $300 it may be a consideration, but there is not much difference for a professional.

The Rigol would walk off the shelves at $500 IMO.

[dadler]

and the historical anecdotal 'evidence' on the forum is "why buy an unknown when the Agilent is not much more expensive and a known good product". A Dave-review would go a long way to assist us in choesing between these products. 

I'm not sure it would.
Assuming the Rigol doesn't have any major problems, it's impossible for me to prove that it's going to be a solid long term meter.
The pedigree and trust that Agilent (and Fluke) have earned in their measurement products is a very real thing. And in this case the Agilent 3441A was the industry standard for decades. And there is nothing even remotely to give anyone the heebie-jeebies that the 34461A isn't a solid successor to that throne.

I didn't say it had to be a long review - looks like we just got one. 

[LaurentR]

Very happy user of the 34461A here.

If price is the reason you're looking at the Rigol, you might as well consider a used 34401A. At ~US$400, you have the same specs as what we're considering here and a rock solid meter. A bit old school and not as featured as the Rigol or the 34461A, but plenty of accuracy and confidence that it'll last.

I have nothing against the Rigol. I own several of their products (happy there too), but for DMMs, the current Rigol offering has a ratio of feature/price that's not a slam dunk (as the DS1054Z is), especially after considering the expected quality and reliability of the Agilent offering.

[LabSpokane]

The 34461A certainly isn't perfect and there's several things I'd like to see improved with future firmware releases. For a day to day usable device, I'm very, very happy with it, however. The trending feature is pure gold. I also like the hold function where it will display the last eight consecutive measurements. That's just great for validating a circuit without having to drop the probes for each measurement. The menu system is very intuitive. From a sheer usability standpoint I don't think Agilent has competition right now. They also offer a five year extended warranty for the stupid cheap price of $61. The calibration plans are more costly but not more than a 3rd party shop run by baboons.

[Shim]

I was just in the same boat looking at both. Price may as well be the same, so let's factor that out. Since I couldn't find any good videos or reviews of extended features, I figured I would buy both and return the one I liked less. Picked up the Rigol yesterday, waiting for the Agilent to show up.

[smgvbest]

You will have to forgive me but I'm not hearing or seeing an actually comparison of the two meter.  I'm only reading personal opinions and if you do not compare the two head on then everything is an opinion and personally biased.  It would be nice to see or read hard facts to back up the opinions.  You are probably right in those opinions but come on.  back them up with facts with a real comparisons.   Just my opinion 

[6thimage]

I have a 34461A and the display is what makes it so good (besides accuracy etc.). I haven't used the Rigol meter, so I don't know how well the interface works, but if it is anything like their oscilloscopes, it will work fine but it won't be as good as an Agilent/Keysight.

Unfortunately, I'm not sure how many people will have used both meters. As if you have the money for an Agilent/Keysight, or a Fluke, then you will not get the slightly cheaper Rigol because of trust and the reputation of the manufacturer.

On a side note, I did a review of the 34461A for a website and made a comparison of the main features of 6.5 digit multimeters, which you might find useful - https://docs.google.com/spreadsheets/d/1sm2jnncUdP6m9HRzQfLtX07vIlWBq1gfggIYGI3KiqM/.

[wraper]

Do not buy DM3068. Unlike with Rigol oscilloscopes there is only small price difference. And their service was total crap after it failed on me less than in two weeks (was sent to Germany few times). Took more than 6 months being away in total and a bit of cheating on my side - destroying AC ranges completely to get it finally exchanged. They tried to prove me that it was within spec although last 4 digits on AC voltage/current ranges were useless jumping garbage. Replacement was dirty refub (even hard to call it refub) with PCB burned and one pad coming off under TVS diode which they replaced. And that with my original meter being almost new, just travelling back and forth. As of strange things, there were few bodge wires in my original meter but non such in the replacement unit although replacement was month or two older and with the same PCB revision. Dark stuff on the photo are not flux residues, I cleaned most of them before making photos duh, looked uglier. It is burned PCB with solder mask which came off near the pads (from the beginning, before I removed TVS to see what's under it). Also in the replacement, PCB was really dirty, Like if coated with the dirt form "crap spray". If you see the last photo from original meter, except bodge wires, there are flux residues all over analog stuff which IMO is not acceptable in 6.5 digit DMM. Just remembered, DM3068 input leakage current is really high with 10 Gohm input resistance setting. Around 100 pA IIRC with 50pA typical in the datashet. On other hand on my 34461A it is tiny, one of the advertised features BTW.

[Smith]

You may take a look at the Keithley 2000.  Only drawback in a second hand model is the FVD may have gotten verry dim after years of continuous use.  Although a spare VFD only costs 59euro ex. I have one personal and one at the job. I really like the quallity of keithley stuff.

[bdivi]

One only has to see the fan on the side of the Rigol

Forced airflow in a precision measurement device = terrible idea. Lots of issues with EMF, thermal cycles, noise (acoustic and electromagnetic).

[6thimage]

One only has to see the fan on the side of the Rigol

Forced airflow in a precision measurement device = terrible idea. Lots of issues with EMF, thermal cycles, noise (acoustic and electromagnetic).

The Agilent/Keysight 34461A has a fan in it, with it blowing air directly across the current resistors and towards the voltage reference.

[DJ]

I would love to see Dave review one of the Rigol DM series meters, in full-honesty mode.

Shahriar/TheSignalPath reviewed the DM3068 here: http://thesignalpath.com/blogs/2012/07/02/rigol-dm3068-6-5-digit-digital-multimeter-review/, but he is sponsored by Rigol (not really doubting his objectivity, just when you are sponsored it's tough to be fully honest, even if it's subconscious).

I've been trying to compare the two meters you have listed myself, and the historical anecdotal 'evidence' on the forum is "why buy an unknown when the Agilent is not much more expensive and a known good product". A Dave-review would go a long way to assist us in choesing between these products. 

I'm hoping the 34461A will appear somewhere with even a modest discount (10-15%)

Don't need it at the moment,  but certainly would like to have one.

I think Shahriar has been quite straightforward in his reviews and have not noticed any bias. I believe he is working on an SDM3055 review, which would be a similar though lower resolution alternative to the Agilent.

[LabSpokane]

Don't count on seeing more than a 5% discount any time soon. They priced that meter aggressively and I had to wait in line even though the 34461 has been released for quite some time.

[LaurentR]

You can get the 34461A with the usual EEVBlog discount at TEquipment.net. That's the best price I have been able to get anywhere in the past 6 months.

[nctnico]

Thanks for all the answers (much more than I have expected)!
So far the consensus seems to boil down to: go for the Keysight 34461A because it is based on a solid proven design. Together with what I have read about the DM3068 in other threads I'm convinced the 34461A is the best buy. I also like the build in LAN and webserver so it is more accessible than a 2nd hand older DMM. What also is in favor of the 34461A is that it is available from Farnell for next day delivery so I could postpone buying it until/if the project actually starts.

[dadler]

I'd like to say that I did not mean to convey that I thought the reviewer I referred to was biased. I was just saying that when a product is given to you for free/you have a special relationship with the manufacturer/etc that it can be hard to be entirely unbiased. If you give a product a shitty review, I doubt you would continue to receive special treatment.

I'm more likely to trust a review from someone who bought the product with their own money. And Dave is so high profile that he can be brutally honest, since his opinion is valued so highly.

[tszaboo]

One only has to see the fan on the side of the Rigol

Forced airflow in a precision measurement device = terrible idea. Lots of issues with EMF, thermal cycles, noise (acoustic and electromagnetic).

The Agilent/Keysight 34461A has a fan in it, with it blowing air directly across the current resistors and towards the voltage reference.

Even the HP 3458A has a fan in it...
If I could choose between a Rigol and an Agilent (or whatever it is called this week), I would not think twice about it. BTW doesnt Farnell have this "buy now, pay later" on Agilent stuff?

[orin]

One only has to see the fan on the side of the Rigol

Forced airflow in a precision measurement device = terrible idea. Lots of issues with EMF, thermal cycles, noise (acoustic and electromagnetic).

The Agilent/Keysight 34461A has a fan in it, with it blowing air directly across the current resistors and towards the voltage reference.

Don't sit a 34461A next to a running HP 8568A... the reading will change by several ppm.

I agree that the trend chart is brilliant, but flawed - they need to allow and display one more significant digit for the Y-axis center/range.  Too often I get three identical labels on the Y-axis and have to go to the range settings to find out what the span actually is.

[Noise Floor]

Is 6.5 all you need?  Keithley just introduced a new 7.5 (DMM7510) that looks like it could be snazzy , pretty pricey though. 

If 6.5 floats your boat you could try and win the 34461A here  O0:
http://www.keysight.com/main/campaign.jspx?cc=US&lc=eng&ckey=2546024

[nctnico]

6.5 digits is already a slight overkill. I do expect to do measurements which need to run for a day or two so some longer term stability is definitely a plus.

[Hugoneus]

Shahriar/TheSignalPath reviewed the DM3068 here: http://thesignalpath.com/blogs/2012/07/02/rigol-dm3068-6-5-digit-digital-multimeter-review/, but he is sponsored by Rigol (not really doubting his objectivity, just when you are sponsored it's tough to be fully honest, even if it's subconscious).

How did you come to the conclusion that I am in any way sponsored by Rigol? I review instruments from nearly every manufacturer: Keysight, Rigol, Tektronix, Keithley, Saleae, Siglent just to name a few.

My tests are scientific, my experiments are industry and research relevant. It is up to YOU to make up your mind if the instrument meets your needs. The whole point of doing a > 1 Hour review with several test setups is to stress the instrument and show it in real-life scenarios.

Dave does the same thing, he goes out of his way to try to make an instrument fail. I appreciate that in him and therefore trust his judgment.

No manufacturer has ever paid me a cent. They often donate the equipment which in turn I use for future elaborate tests and educational videos. The Signal Path loses money in fact, out of my own pocket.

[rx8pilot]

The Signal Path loses money in fact, out of my own pocket.

I am thankful you were able to clear that up. The information you have presented in all of your videos is very relevant and useful - much appreciated. I can see the the many, many hours spent prepping for the 1 hour segments. Experiments don't design themselves.

If it ever comes down to it, take donations. Any sponsors will spoil a great thing.

[BillyD]

Removed discussion regarding sponsorship removed, as there is no sponsorship involved.

[LabSpokane]

6.5 digits is already a slight overkill.

Yes, but if you use that old metrology rule of thumb where your device is 10X the accuracy of your measurement, then it's not so silly. 

[EEVblog]

6.5 digits is already a slight overkill.

Not at all.
Even for very basic stuff, a 6.5 digit meter can be very useful.
For example, being able to see the rate of discharge on a battery with minimal drain. a 50,000 count meter won't show a thing, but a 5,000,000 count meter will show a nice trend downwards. You need resolution to do this, not accuracy. But accuracy (and hence cost) usually comes along with resolution.

[EEVblog]

I suggest the poster and/or moderators should edit or delete the untrue posts.

This is not a job for moderators. If it was, no conversations would ever take place in a forum.

[dadler]

Shahriar/TheSignalPath reviewed the DM3068 here: http://thesignalpath.com/blogs/2012/07/02/rigol-dm3068-6-5-digit-digital-multimeter-review/, but he is sponsored by Rigol (not really doubting his objectivity, just when you are sponsored it's tough to be fully honest, even if it's subconscious).

How did you come to the conclusion that I am in any way sponsored by Rigol? I review instruments from nearly every manufacturer: Keysight, Rigol, Tektronix, Keithley, Saleae, Siglent just to name a few.

.....

No manufacturer has ever paid me a cent. They often donate the equipment which in turn I use for future elaborate tests and educational videos. The Signal Path loses money in fact, out of my own pocket.

I was also surprised to see this accusation. Quite nasty of the poster to casually make statements like that without bothering to check into it, especially considering how much it could harm the reputation and credibility of the signal path blog.
I suggest the poster and/or moderators should edit or delete the untrue posts.

I found the "sponsorship" reference online in reference to that particular video review, when researching the same DMMs. I think I assumed it was common knowledge (I am new) that he was *literally* sponsored, in the formal sense. I read multiple posts to this effect and assumed - I apologize.  I am guilty of repeating what I found and thoroughly apologize. I edited my original post because you are correct that I do not have any substantiation.

[dadler]

Here is the reference, for those wondering where I got it:

http://thesignalpath.com/blogs/2012/07/02/rigol-dm3068-6-5-digit-digital-multimeter-review/

Jeremy
July 4, 2012 at 4:23 PM
This review is great. I have a DS1052E 50 MHz DSO, which I like a lot. It seems like Rigol produces some great products, many of which are inexpensively priced.

I know you are sponsored by Rigol, but it would be great to see the downsides of the DM3068 as well as a comparison to the DM3058.

I appreciate the effort you put into creating your videos. They are high quality and well thought out. I would love to see them more frequently, but I understand the amount of effort it takes is very high.

Keep up the good work!

-Jeremy

Emphasis mine. It's on his site, so though it is a comment it has been there uncontested for 2.5 years. It was mentioned so nonchalantly that I believed it to be common knowledge.

[Hugoneus]

Here is the reference, for those wondering where I got it:

http://thesignalpath.com/blogs/2012/07/02/rigol-dm3068-6-5-digit-digital-multimeter-review/

Emphasis mine. It's on his site, so though it is a comment it has been there uncontested for 2.5 years. It was mentioned so nonchalantly that I believed it to be common knowledge.

I had not noticed that comment on my video. I am not sure why this assumption has been made and it is incorrect. I have provided a reply on my site as well to void future confusion.

[dadler]

Here is the reference, for those wondering where I got it:

http://thesignalpath.com/blogs/2012/07/02/rigol-dm3068-6-5-digit-digital-multimeter-review/

Emphasis mine. It's on his site, so though it is a comment it has been there uncontested for 2.5 years. It was mentioned so nonchalantly that I believed it to be common knowledge.

I had not noticed that comment on my video. I am not sure why this assumption has been made and it is incorrect. I have provided a reply on my site as well to void future confusion.

Thanks. I formally apologize to you, I did not intend to besmirch your reputation. I will admit that reading that comment after watching your video made me question your objectivity, and I am glad you have replied and corrected the original commenter. This does not justify my loose-application of what I thought were 'facts', though-I deserve the blame for that.

[Shim]

My 34461A came yesterday. At first glance, there are some significant differences between the two. I am actually slightly leaning toward the Rigol.

Is there any comparisons that anyone would like to see while I have both? Got a bit less than 30 days before one has to go back.

[DJ]

One only has to see the fan on the side of the Rigol

Forced airflow in a precision measurement device = terrible idea. Lots of issues with EMF, thermal cycles, noise (acoustic and electromagnetic).

The Agilent/Keysight 34461A has a fan in it, with it blowing air directly across the current resistors and towards the voltage reference.

Don't sit a 34461A next to a running HP 8568A... the reading will change by several ppm.

I agree that the trend chart is brilliant, but flawed - they need to allow and display one more significant digit for the Y-axis center/range.  Too often I get three identical labels on the Y-axis and have to go to the range settings to find out what the span actually is.

Is the Agilent unusually susceptible, the 8568A noisy* or a test setup acting like an antenna?


*near-field emissions can be a challenge to shield

[DJ]

You can get the 34461A with the usual EEVBlog discount at TEquipment.net. That's the best price I have been able to get anywhere in the past 6 months.

Tax in my state negates the TEquipment discount.

Considering the potential for damage with the Siglent continuity check (8V! What were they thinking?) will eventually bite the bullet and get the Agilent. Maybe will wait for the next Rev. Not a need but more of a want.

[nctnico]

My 34461A came yesterday. At first glance, there are some significant differences between the two. I am actually slightly leaning toward the Rigol.

Is there any comparisons that anyone would like to see while I have both? Got a bit less than 30 days before one has to go back.

For me the major concerns are long term stability, firmware bugs and the PC software. I did not look at the software from Keysight but what I'd like to have is a log spanning 2 or 3 days which I can import into Excel.

[orin]

One only has to see the fan on the side of the Rigol

Forced airflow in a precision measurement device = terrible idea. Lots of issues with EMF, thermal cycles, noise (acoustic and electromagnetic).

The Agilent/Keysight 34461A has a fan in it, with it blowing air directly across the current resistors and towards the voltage reference.

Don't sit a 34461A next to a running HP 8568A... the reading will change by several ppm.

I agree that the trend chart is brilliant, but flawed - they need to allow and display one more significant digit for the Y-axis center/range.  Too often I get three identical labels on the Y-axis and have to go to the range settings to find out what the span actually is.

Is the Agilent unusually susceptible, the 8568A noisy* or a test setup acting like an antenna?


*near-field emissions can be a challenge to shield

Nothing that complicated.  The 8568A blows warm air out of the side vents right at the 34461A which then demonstrates its temperature coefficient of around 1ppm/K.

[blackdog]

Hi,

Just for the fun....

1e, 10V reference in a oven, measured with a fresh calibrated 3458a (+0,45PPM)

HP 34401A, 10V range 100PLC.
Keysight 34461A 10V range 100PLC.
Keysight 34461A 5K thermistor measuring the Temp in my LAB.
The measured time is approximately 24 hours.

The Temp measurement...



10V 34461A after the use of zero key.





The 34401A and 34461A are both very good, they are less than 1PPM / Celsius :-)

Kind regarts,
Bklackdog

[nctnico]

What kind of 10V reference did you use?

[blackdog]

Hi nctnico,

It is a design of mine, about 3 years old and this reference is in a BimBox oven, about 42C.



Kind regarts,
Blackdog

[BillyD]

Many thanks to Dadler for clearing up the confusion about Signal Path. Well done sir!
Billy.

[LaurentR]

Considering the potential for damage with the Siglent continuity check (8V! What were they thinking?) will eventually bite the bullet and get the Agilent. Maybe will wait for the next Rev. Not a need but more of a want.

You'll be disappointed with the 34461A too...
Here is a sampling of what I found around the home:

* Agilent 34461A: 9.4V
* Fluke 87V: 7.3V
* HP 34401A: 6.9V
* Velleman 10ADC ($10 cheapie): 3.0V
* Klein CL1000 clamp (Home Depot grade): 0.4V

There are a few threads about this including: https://www.eevblog.com/forum/testgear/multimeter-with-low-continuity-test-voltage/

[Shim]

Considering the potential for damage with the Siglent continuity check (8V! What were they thinking?) will eventually bite the bullet and get the Agilent. Maybe will wait for the next Rev. Not a need but more of a want.

You'll be disappointed with the 34461A too...
Here is a sampling of what I found around the home:

* Agilent 34461A: 9.4V
* Fluke 87V: 7.3V
* HP 34401A: 6.9V
* Velleman 10ADC ($10 cheapie): 3.0V
* Klein CL1000 clamp (Home Depot grade): 0.4V

There are a few threads about this including: https://www.eevblog.com/forum/testgear/multimeter-with-low-continuity-test-voltage/

Rigol DM3068 (measured with Agilent 34461A) 5.8V

[6thimage]

Considering the potential for damage with the Siglent continuity check (8V! What were they thinking?) will eventually bite the bullet and get the Agilent. Maybe will wait for the next Rev. Not a need but more of a want.

You'll be disappointed with the 34461A too...
Here is a sampling of what I found around the home:

* Agilent 34461A: 9.4V
* Fluke 87V: 7.3V
* HP 34401A: 6.9V
* Velleman 10ADC ($10 cheapie): 3.0V
* Klein CL1000 clamp (Home Depot grade): 0.4V

There are a few threads about this including: https://www.eevblog.com/forum/testgear/multimeter-with-low-continuity-test-voltage/

Rigol DM3068 (measured with Agilent 34461A) 5.8V

Just measured my Keithley 2000 and it's 6.6V

[BloodyCactus]

interesting. my gw instek 8251a bench dmm puts out 2v for continuity according to my handheld dmm (uei dm393)

[Shim]

The more time I spend playing with these two, the harder it gets to choose between them. I will say that right now, I'm only messing around and not really using them in real world scenarios yet.

I did just have a thought, and tested it to confirm, and thought I should post it up.

For those who think the Rigol has an advantage for continuity due to the fact you can adjust the threshold, I don't think so. With the Agilent, you can use the trending feature and the limits feature and get a much wider range. You don't get a continuous beep, but you can get a beep when the set threshold is past. The trend chart on ohms with limits set is kinda cool too.

[blackdog]

Hi,

I do not understand, talking about the continuous test of the multimeters, I have eight bench multimeters and I use this almost never.
When comparing multimeters, this feature in my eyes completely subservient.
I want a multimeter that temperature is stable, good specifications and good support and it is convenient to use, etc. etc.
Choosing a multimeter as is used in the continuous test of the meter use 4.5 or 9,2V: "it's a little silly"

I have two KEYSIGHT 34461A multimeters and what I think is important is the calibration, how well it is done.
I send my DMM here in Europe, to KEYSIGHT in Germany.
I am sure that it is done properly.

My 2 cents, ;-)

Kind regarts,
Blackdog

[KedasProbe]

You can find my DM3068 remarks here:
https://www.eevblog.com/forum/testgear/rigol-dm3068-issues-remarks-improvements-requests/

[nctnico]

I use continuity testing a lot when repairing and/or reverse engineering circuits or just checking whether there is a connection or short between two points. IMHO a general purpose DMM without instant (fast isn't good enough; it must be instant!) isn't useful for that. Also the voltage used for the continuity test doesn't matter. ICs have zener clamps for ESD protection anyway. What is more important is the maximum current a DMM supplies in continuity test. That should be low enough not to cause damage.

Than again: I'd don't see why someone would buy  6.5 digit DMM just for the continuity test  O0

@blackdog: the schematic of your reference looks interesting. Did you use several reference chips in parallel to reduce the noise?

[blackdog]

Hi, nctnico,

I did my best three years ago, in order to keep the noise as low as possible.
I use a stable low noise 8V supply for the LT1021, al lot of filtering and security components, so i wil not blow up the reference.
Now, I would replace the LT1012 opamp by a modern chopper opamp from LT, such as the LTC2057.

The reference is not yet in his cabinet but is located in a polystyrene box, to keep him on as long as possible, until I have a decent housing.
It's a bit of a long-term project, but you have no choice in good voltage references. :-)

Kind regarts,
Blackdog

[SteveyG]

I'm surprised people are recommending the 34461A over the 34410A. I'd much rather have the better specs over the tacky LCD for a DMM.

[blackdog]

Hi SteveyG,

Do you have both instruments, that is 34401A and 34461A?
Have you done comparative measurements with both DMMs?

If you look @ the pictures i posted here, you can see higher resulution and i can tell you, the 34461A is more stable.
AC is better then the 34401A.
But, maybe i do not understand your remark?

Kind regarts,
Blackdog

[Andreas]

Hello,

did you do a ADEV stability calculation over the 2 instruments.
Or do you have the .csv raw data from the comparison measurement?

with best regards

Andreas

[blackdog]

Hi Andreas,

"Ar you talking to me" (Taxidriver)     

Both measurements are done with Keysight BenchVue.

Kind regarts,
Blackdog

[SteveyG]

Hi SteveyG,

Do you have both instruments, that is 34401A and 34461A?
Have you done comparative measurements with both DMMs?

If you look @ the pictures i posted here, you can see higher resulution and i can tell you, the 34461A is more stable.
AC is better then the 34401A.
But, maybe i do not understand your remark?

Kind regarts,
Blackdog

I'm talking about the 34410A, not the 34401A. Yes we have the 34410A and 34461A. There is no successor to the 34410A (yet).

[wraper]

My 34461A came yesterday. At first glance, there are some significant differences between the two. I am actually slightly leaning toward the Rigol.

Is there any comparisons that anyone would like to see while I have both? Got a bit less than 30 days before one has to go back.

Have both, Rigol suck compared to 34461A. Especially on quality and warranty aspect as I said in previous post. Have worse true RMS and larger input leakage current. Not to say Keisight have more bells and whistles. Actually I really regret buying Rigol, 34461A was announced just few weeks after I bought Rigol. On top of that spent half a year to exchange just a few weeks old meter. Just compare how they are built inside to see the difference.

[DJ]

I use continuity testing a lot when repairing and/or reverse engineering circuits or just checking whether there is a connection or short between two points. IMHO a general purpose DMM without instant (fast isn't good enough; it must be instant!) isn't useful for that. Also the voltage used for the continuity test doesn't matter. ICs have zener clamps for ESD protection anyway. What is more important is the maximum current a DMM supplies in continuity test. That should be low enough not to cause damage.

Than again: I'd don't see why someone would buy  6.5 digit DMM just for the continuity test  O0

@blackdog: the schematic of your reference looks interesting. Did you use several reference chips in parallel to reduce the noise?

Input protection varies on IC's. Some are zener-clamped, some are body diode-clamped to the rails, some use dielectric isolation with other means of input protection.  Leakage through input protection diodes can impact amplifier performance. I rarely use continuity testing on bench meters, usually reserving it for cables and such. I have some older unprotected FET's i would imagine would be smoked by excessive gate voltage, hence my surprise that some of the open circuit voltage of continuity testers on bench meters was that high. On a portable Fluke I could see, but not something I would be using on sensitive circuits. Making a mental note to be double-certain to watch its use.

[Andreas]

Hi Andreas,

"Ar you talking to me" (Taxidriver)     

Both measurements are done with Keysight BenchVue.

Kind regarts,
Blackdog

Yes.

Would it possible to get out the .csv values out of the recordings?
A ADEV stability calculation would be interesting between the 2 devices.
see also the KX thread.
https://www.eevblog.com/forum/projects/project-kx-diy-calibrator-reference-sourcemeter/msg592144/#msg592144

With best regards

Andreas

[sotos]

Ok here are mine also.

Fluke 289 continuity or diode =  5.42v
Agilent U3606A continuity or diode = 4.3v.

[blackdog]

Good morning Andreas, :-)

You can play now...
www.bramcam.nl/Diversen/Andreas-Blackdog-10V-Ref-CVS.zip

Almost 58H data in CVS format of the 34461A and the 34401A DMM's

Kind regarts,
Blackdog

[robrenz]

@  nctnico   Jump on this quick!  I guarantee you will not regret this purchase

[nctnico]

@Robrenz: tempting but if only it was the 8846A with the temperature measurement option.

[robrenz]

Its no option, its included in the stock meter using a platinum RTD

BUY IT QUICK!

EDIT:

Sorry I misread the 8845  DONT buy that one

[Andreas]

Hello blackdog,

attached the statistical comparison of the systems:
I think the values are only the offset to the starting value.
(showing the uV deviation over time).

HP34401A:
plotted over time with plotter program shows already discrete voltage jumps.
The standard deviation (8.6 uV) with 200 classes shows that the resolution of the logged data is 1uV.

But all 5uV there is a maximum most probably due to rounding artefacts
 from binary internal resolution to the output at the interface.

All in all its no gaussian distribution but 2 overlapping distributions.

Autokorrelation: (don´t know how to interpret but looks strange).

Allan deviation shows 0.8uV stability.
Averaging of multiple values seems not to improve stability.

HP34461A:

On the time diagram the resolution appears much finer and with more stability.
The standard deviation 3.6uV is factor 2 better than above. No discrete values found.
The distribution looks more gaussian. Although 2 distributions available.

Autokorrelation shows a repeated function all 20000 samples.
Perhaps interference (beat frequency) with mains frequency?

Allan deviation also shows 0.8 uV stability.
Getting a little better when averaging 10-20 values.

I cannot finally state what that means. Usually the Allan deviation and standard deviation should be equal for 1 sample.
And also the nearly same value is not clear to me.
But since it is the whole system which is judged including the reference voltage,
the 0.8uV may also be the reference voltage stability which is limiting.

And finally I wanted to know the resolution of the 34461A and got a big surprise when doing standard deviation with 1000 classes:

It seems that the instrument is switching the range or resolution above +10uV difference.
very strange. Does the instrument cheat ? (and secretly switching ranges)
I would have expected better maths in a instrument in this class.

With best regards

Andreas

[blackdog]

Hi Andreas, :-)


The measurements and outcome are a mix of reference drift, DMM drift, Seebeck effects kabels and tempdrift.

I dit one more measuremt this afthernoon with the 34461A DMM.
Put the range on 10V DC, 100PLC.
And have measured its own drift, input shorted, see the picture.

The drift is about 0.3PPM for 6.5H.
On the Spec. sheet for the 34461A there is for the 10V DC 0.0025 + 4 digit, say 30PPM within Tcal +- 1C
I find my measurements not to bad, if you include all the errors...


Kind regarts,
Blackdog

[Andreas]

Ok, good zero stability,

but it is only the noise of the ADC + the thermal EMF of the wires which you see there.
The reference voltage stability of the 34461A is not part of the offset measurement.

With best regards

Andreas

[Hugoneus]

What temperature do you guys usually keep your lab at?

[blackdog]

Hi Andreas,

I know, but the noise that you see, is already a fair proportion of the variation of the other measurements, but correct me, if you think it is different.

Hi Shahriar,
The temperature in my LAB depends on my emotional state ... 
Normally, the temperature is between 19 and 23C in my LAB.

For my 2x TEK DMM4050 and 2x KEYSIGHT 34461A DMMs I want to make a small box that is temperature controlled.
It wil have a number of Peltier elements to keep the temperature around DMMs fairly stable.
I want to start with the air supply at about 21C, and then measuring with multiple temperature sensors in and outside the multimeters how well the temperature remains stable. This is a project for the coming months.


Again, the schematic that I showed in this topic, is a good reference if you build him well.
The next reference that I'm going to build, contains eight pieces LM399AH, and I will build one, with eight pieces LT1021-5V.
Both reference will be built in an oven, around 42C.
At the moment I test fourteen pieces LT1021-5V, to select out the LT1021 with the largest deviation.

Kind regarts,
Blackdog