Recommend me a bench meter (sorry)

[garytoosweet]

Hello,

I've done quite a bit of research, searched the forum, read the stickies, perused the spreadsheet. But I haven't been able to hone in on exactly what meter I should be looking at.

I'm in Canada if you couldn't tell from the title. Can purchase anywhere in NA.

Primarily do audio work. Lots of vacuum tubes, some transistor. Not really any "modern" needs.

I'm looking for a meter with:
- 1KV rating (frequently measure 700V+)
- mA range
- True RMS
- Continuity mode is a + (understand not many bench meters have this, not a dealbreaker but would be nice)
- Ease of calibration and repairability (Looking to keep for a lifetime, not replace after a few years)
- Smaller form factor is better, space is always an issue. Not a deal breaker if larger though.

I'm looking for something that's all in one, I can use in service everyday, but is also useful for calibrating other equipment. I just aligned my Tektronix 577 with my Fluke 179, and while accurate enough for my needs, more accuracy would have been better. The form factor of a bench meter would be great in my shop. Get it up on the shelf and off of the bench.

I enjoy old stalwart, well made gear that's well designed and has stood the test of time like the HP200 oscillator, for example.

I am flexible on budget but would prefer to keep under $500CAD/$400USD.

Any suggestions as to what I'm looking for?

Thank you in advance!

[bob91343]

How did true rms get to be important?  I can think of better ways to increase meter cost.  And true rms really isn't, if you are dealing with higher frequencies.  Ask yourself why you want this, or is it just because it's become a must have for those who don't understand what it's for.

Unless you need accuracy, analog is usually better than digital.  And most analog meters also measure current.

I find that all in one meters don't do anything that well.  Certainly not everything.  I have an assortment of meters, analog and digital, on the bench at all times and find myself reaching for each one as my need suggests.  I have three oscilloscopes and more than one spectrum analyzer and a few RCL bridges and even a 6-1/2 digit voltmeter on the bench at once.  A clip on dc milliammeter and a clip on ac meter and a little component tester and a few power supplies and signal generators and....  Well you get the idea.

[nctnico]

You could go for something old but still I recommend to look at the Vici VC8145 4.5 digit bench DMM. It is my daily driver for nearly 10 years already. If it breaks I'd buy one again without thinking twice (actually I have two of them). I also have a 6.5 digit bench DMM from Keysight but I rarely use it. For daily use the VC8145 is just more convenient. Compared to older DMMs the VC8145 also has temperature, frequency and capacitance measurement which come in handy as a quick check every now and then.

[nightfire]

Basically you look for some high-end chinese gear  Like GRS (GW Instek, Rigol, Siglent), or some used stuff that is proven over time.

Advantage of old stuff like Fluke 45 or Agilent 34401 is their proven quality and ability to be calibrated by every calibration shop because these pieces of gear are well known.
Disadvantage of that gear is their age, often 20 years, so they will need some maintenance in the future.
I managed to get a Fluke 45 from ebay including calibration for decent money, but I am aware that I probably will have to put some money into it for repairs in the future.

Advantage of new gear is the warranty they have, but with the more recent DMMs of the abovementioned manufacturers there are no real long-time experiences to be shared.
Also calibration could be a bit of a hassle, depending of how well known those are  to calibration centers in your region.

With  a bit of luck you could look for some refurbisher that sells old  (and tested) agilent 34401a or 34410a units, that fit in your budget.


Repairability of modern DMM is probably out of scope for the amateur, as they tend to use more and more complex parts, some of them proprietary and therefore becoming unobtainium nowadays.
I also am aware that if some critical parts of my old (40 yrs) Fluke 8600a will break, it can only serve as a parts donor to other units...

[bdunham7]

- 1KV rating (frequently measure 700V+)

Do you mean DC or AC 1kV?  1 kVDC would be pretty standard, 1kVAC would narrow your options a lot.

- mA range
- True RMS

Pretty standard on bench meters unless you get certain really old models.  Also some older and otherwise excellent meters have limited current measuring capability, although they do have mA--just not uA and 10A ranges.
 Consider one other feature that gets overlooked, high input impedance.  Having >10G input impedance is very handy at times.

- Ease of calibration and repairability (Looking to keep for a lifetime, not replace after a few years)
I enjoy old stalwart, well made gear that's well designed and has stood the test of time like the HP200 oscillator, for example.
I am flexible on budget but would prefer to keep under $500CAD/$400USD.

That's going to rule out anything new and most things that I'm aware of that are less than 20 years old.  If you want a 5+ digit meter, then calibration typically won't be easy because the standards you need aren't cheap or easy to maintain.  And while old-school meters are often very repairable, parts are not becoming any easier to come by. 

Do you want used or new? 

[garytoosweet]

Thanks for all input already.

With my budget and needs, I've been looking at used only.

1kv DC measurement.

High Input impedance would be nice to have.

Strong preference to pro brands like Fluke and Agilent over newer overseas gear.

I hadn't seen the Fluke 45. Need to look into it more but I believe that check all my boxes. Looks like a great value meter. I will look at the 34401/34410 as well. Great suggestions thank you. 

[bdunham7]

OK, look at:

HP 3478A

HPAK 34401A

Fluke 8840A/8842A

All have published schematics and enough of a following that you can get repair help along the way.  The HP3478A is older, has a non-backlit LCD display and high impedance input only to 3V, but it will be the cheapest and is still a surprisingly good meter. (bad advice--see below!) HPAK 34401A is 6.5 digit and is the most advanced of the lot, there is a schematic but it is mostly SMT and has a lot of proprietary parts--it will also be the most expensive.  Fluke 8840A/8842A are 5.5 digit but remarkably accurate, especially the 8842A.  They also have high impedance all the way to 20 volts.  The tough part is finding one in good condition because for whatever reason, a lot of these end up looking like they've been to war. 

[nctnico]

Thanks for all input already.

With my budget and needs, I've been looking at used only.

1kv DC measurement.

High Input impedance would be nice to have.

Strong preference to pro brands like Fluke and Agilent over newer overseas gear.

I have been down that road as well but in the end the old gear has a lot of downsides. Lots of older DMMS have either tiny, hard to read LCD displays or VFD displays which are prone to wear. And many of the higher end bench DMMs have no or slow continuity. My conclusion when looking for a bench DMM is that the old DMMs are just awfull and are only suitable if you need very precise current or voltage measurements. A mistake you often see in these kind of threads is that people think bench DMM = high performance/accuracy DMM. I see several 6.5 digits DMMs suggested already while you are likely better off with a 4.5 digit bench DMM which has similar features as a 4.5 digit hand held.

[tautech]

At $ 389 the SDM3045X is not bad buying:
https://siglentna.com/digital-multimeters/sdm3045x-digital-multimeter/

No cooling fan.

[25 CPS]

OK, look at:

HP 3478A

HPAK 34401A

Fluke 8840A/8842A

All have published schematics and enough of a following that you can get repair help along the way.  The HP3478A is older, has a non-backlit LCD display and high impedance input only to 3V, but it will be the cheapest and is still a surprisingly good meter.  HPAK 34401A is 6.5 digit and is the most advanced of the lot, there is a schematic but it is mostly SMT and has a lot of proprietary parts--it will also be the most expensive.  Fluke 8840A/8842A are 5.5 digit but remarkably accurate, especially the 8842A.  They also have high impedance all the way to 20 volts.  The tough part is finding one in good condition because for whatever reason, a lot of these end up looking like they've been to war.

Careful!  The 3478A won't meet his requirements.  As nice as the 3478A is, and I have two of them that I use frequently, they top out at 300 V and he's listed 1,000 V as a requirement for working on vacuum tube circuits.

The 34401A will meet that requirement and is a great meter, again I have a pair of those that I also use frequently because of working on tube circuits as well, and can probably be squeezed in under the $500 CAD budget listed.

[bdunham7]

Careful!  The 3478A won't meet his requirements.  As nice as the 3478A is, and I have two of them that I use frequently, they top out at 300 V and he's listed 1,000 V as a requirement for working on vacuum tube circuits.

Thanks for the warning!  I wouldn't have even mentioned the 3478A but I actually have one here that I'm to repair and test and it seems a nice old meter.  I didn't know of its limited voltage range and you may have just saved me from blowing it up because I would have just assumed it was good for 1000V.  So scratch that unless he gets an HV probe.

[25 CPS]

Careful!  The 3478A won't meet his requirements.  As nice as the 3478A is, and I have two of them that I use frequently, they top out at 300 V and he's listed 1,000 V as a requirement for working on vacuum tube circuits.

Thanks for the warning!  I wouldn't have even mentioned the 3478A but I actually have one here that I'm to repair and test and it seems a nice old meter.  I didn't know of its limited voltage range and you may have just saved me from blowing it up because I would have just assumed it was good for 1000V.  So scratch that unless he gets an HV probe.

No worries!  I put that out there because I came really close to making the same mistake.  When I started putting my bench together at home, I bought a pair of 3478As and a 3457A for reasonable money with the idea that between those three meters on the bench, I'd be good for measuring voltages and currents and resistances and I could kick the crummy Canadian Tire handheld DMMs out to the car.  I was putting the bench meters in place when I saw the 300V note on the front panel and 350V on the 3457A and thought, ok, well, I'm set up except for tube stuff that's got a B+ worth mentioning.

I found a 34401A somewhat locally for reasonable money and got it shipped in to take care of that.  The other one was one of those NIB US Navy ones being sold on eBay that I got back in January.  That one was a bit more expensive but if the original poster can stretch his budget, that might not be a bad idea to think about if good, used 34401As are hard to find or as expensive as one that's new.

[Cerebus]

Add the Keithley 197 or 197A to the list.

5 1/2 digit, 220,000 count, basic DC accuracy of 0.011%+ 2 counts (1 year), 200 uA range giving nA resolution, 200 mV range giving uV resolution, true RMS, 1000 VDC range, 4-wire ohms, continuity setting, easy to calibrate from the front panel and, unlike modern bench meters, turns on instantly - no waiting for your DMM to boot. You'd probably be able to pick one up for under $200CDN.

I have a 197 and it agrees digit for digit with my Agilent 34461A. The 197A is preferable because it has a backlight. The 197 has a large clear LCD but can be tricky to read without a backlight under some circumstances if the ambient light isn't coming from the right direction.

Edited: Also they are chock full of standard components, no custom parts made of unobtanium to find if you need to fix them and there's a manual with full circuit diagram available.

[guenthert]

     What about acoustic noise?  I'm not kidding.  My HP34401A, probably my most accurate meter, is sadly in storage right now.  My Agilent 34410A's fan noise causes it to be rarely used (otherwise it's a fine meter), while my HP3457A (only goes to 300VDC) is never turned off.

     I'd go with a 5.5 digit meter.  The long scale meters just lure you down into volt-nutterie, which won't end well ...

     Doesn't "true" RMS require a thermal converter?  Not many of those around.  A used Fluke 8506A might be the cheapest one.  Datron 1271 had them as option and can be occasionally found for a reasonable price.  But if highest accuracy isn't all that important, the cheapest "true" RMS might be a HP3400A (with potentially a regular DMM measuring it's DC output).

EDIT: A HP3456A can fairly easily be calibrated one-self, but don't ask a cal-lab for it.  If they do it at all, they'll charge an arm and leg for it (too much work).  I believe calibration of a 34401A can be acquired for less than $100 (a process which can be automated).

[bdunham7]

I have been down that road as well but in the end the old gear has a lot of downsides. Lots of older DMMS have either tiny, hard to read LCD displays or VFD displays which are prone to wear. And many of the higher end bench DMMs have no or slow continuity. My conclusion when looking for a bench DMM is that the old DMMs are just awfull and are only suitable if you need very precise current or voltage measurements. A mistake you often see in these kind of threads is that people think bench DMM = high performance/accuracy DMM. I see several 6.5 digits DMMs suggested already while you are likely better off with a 4.5 digit bench DMM which has similar features as a 4.5 digit hand held.

I won't disagree in the general case as there a a lot of older DMMs that would be a poor choice for your only meter--like the 3478A that I didn't realize was only 300V!  However, the OP had specific wishes about brand and serviceability and that rules a lot of things out, even something like the BK 2831E that I had that was a very handy all-around sort of meter.  The 34401A has fast latching continuity, selectable high-impedance, math, etc and although VFD, the VFDs actually last a very long time.  It's the most modern DMM I can think of that is reasonably serviceable.

[beanflying]

Along with keeping an eye out for the Keithley's Cerebus mentioned also Advantest or Solartron they often go for a bit less then the likes of the 34401A.

On the 34401A if you can stump up the $ then do it  After doing some extended testing is is more stable than the replacement 34461A (still love it for the stats and export functions) and none of the Fan issues of the 34410.

Your other option would be look for a high voltage divider probe to get around the 1kV requirement.

[james_s]

How did true rms get to be important?  I can think of better ways to increase meter cost.  And true rms really isn't, if you are dealing with higher frequencies.  Ask yourself why you want this, or is it just because it's become a must have for those who don't understand what it's for.

Unless you need accuracy, analog is usually better than digital.  And most analog meters also measure current.

I rely on true RMS heavily, I can't imagine going back to an averaging meter. I've used it frequently for measuring current and voltage in discharge lamp circuits which are non-sinusoidal, and measuring current to low power factor loads, again non-sinusoidal, and working on "modified sine wave" inverters and UPS's which produce a square wave with dead time, and switchmode power supplies and PWM drivers. It doesn't really matter anymore anyway, it is not expensive anymore to include true RMS, the only reason anyone even makes meters above the Harbor Freight freebie class that lack it is so they don't cannibalize sales of their better offerings.

Analog meters have some advantages, they're nice for quick relative measurements, but every digital meter I've owned for about the last 25 years has a bargraph under the digits that provides that functionality. Accuracy is important to me, as is being able to read it quickly. I really don't miss trying to interpret the reading on an analog scale with 5 different sets of numbers. I finally gave away my last analog multimeter about 8 years ago when I realized I hadn't used it in years.

[james_s]

     What about acoustic noise?  I'm not kidding.  My HP34401A, probably my most accurate meter, is sadly in storage right now.  My Agilent 34410A's fan noise causes it to be rarely used (otherwise it's a fine meter), while my HP3457A (only goes to 300VDC) is never turned off.

The noise can be dealt with. I installed Noctua fans in my HP frequency counter and function generator which I believe are in the same family as the 34401A. It made a huge difference, they went from an annoying whine to almost inaudible.

[beanflying]

Only the 34410A has the fan the 34401A doesn't. Might be a slight dyslexic model number issue I think I also would be careful playing with case airflow as you might upset the last few PPM of stability/calibration

[wizard69]

You have gotten advice all over the map here and frankly I'm going to mix it up even more.

First consider carefully buying used as others have pointed out the needed ranges may not be there.   More importantly parts for repair might not exists.   I have really mixed feelings about suggesting used, if I did I'd stick with Fluke or HPAK.   You could also consider other units if good documentation is available.

Consider instead a meter actually designed for bench use such as a B&K Precision Model 5491B.   You need to consider the max AC voltage range carefully on this unit as it just barely meets your need.   However it has plenty of functionality and suitable range (so called 4&3/4).   The best part is it can be had for around $400 new with warranty.   This is just one suggestion of what can be had in ≈4 digit multimeters.   I just don't see a good reason to go to 5.5 or 6.5 digits for your usage.   Siglent is another manufacture with a nice low cost bench meter.   

In the end functionality wise the newer bench meters far out class the bench meters of old.   What becomes an issue is that for many of them you are right on the border line of max AC voltage measurements.   On the other hand a lot of the older meters out there didn't even have AC measurements as standard.

Now for RMS measurements, when I first got involved in fixing tube driven RF amps for industrial processes the guy training me insisted that I buy a multimeter (went HP at the time) with a thermal converter.   The only use for it was in setting the filament voltage.   This was due to the power supply supplying current to the filament.   I doubt this is an issue in your case but it highlights that the RMS conversion process can be important.    There are likely engineers on the forum that can better comment on this, but I just wanted to offer it up.

In a nut shell I would not buy used at all unless you get an outstanding deal.   You are likely to pay the same price for a used meter as one of the new 4.5 digit meters and get far more features with the new meter.

One more thing, if you think you need all the digits of an old system meter consider this, many of the tube amp designs are pretty old, the engineers and techs at the time would have been using analog meters.

[AVGresponding]

One limitation of the Fluke 8840 that should be mentioned is that the lowest current range is 200mA, with 10uA resolution. This may not be precise enough for some needs. The 8842 has a 20mA range at least. Also neither have a continuity function though the 4-wire resistance capability compensates for this a bit.

Your price range would seem to allow for a Keithley 2000, which would likely tick all your boxes and a few more besides. A Keithley 2015 might be a bit more than you want to pay, but it has some dedicated audio electronics functions, so shouldn't be ruled out immediately.

[james_s]

Only the 34410A has the fan the 34401A doesn't. Might be a slight dyslexic model number issue I think I also would be careful playing with case airflow as you might upset the last few PPM of stability/calibration

True, with the HP instruments I modified the stability out of the box is terrible to begin with. I added a OCXO reference to the counter which produced a dramatic improvement and I'm hoping someone reproduces the option to add an external reference to the function generator.

For a multimeter I suppose it might be more critical, although you can still cut down the noise to some degree with a quieter fan that moves the same amount of air.

[rsjsouza]

I would add also one that is slightly above your price range if purchased used: the Keithley 2015. It goes to 750V (covers your 700V upper voltage) and has an interesting THD analyzer for audio.

[wizard69]

I would add also one that is slightly above your price range if purchased used: the Keithley 2015. It goes to 750V (covers your 700V upper voltage) and has an interesting THD analyzer for audio.

That is a very interesting possibility for somebody into audio equipment.   I'm still not real thrilled about used but that is a current production machine and maybe more importantly it is two instruments in one for the audio equipment bench.  The THD analyzer might make it easier to justify going over budget.

[HKJ]

I would add also one that is slightly above your price range if purchased used: the Keithley 2015. It goes to 750V (covers your 700V upper voltage) and has an interesting THD analyzer for audio.

That is a very interesting possibility for somebody into audio equipment.   I'm still not real thrilled about used but that is a current production machine and maybe more importantly it is two instruments in one for the audio equipment bench.  The THD analyzer might make it easier to justify going over budget.

Just remember some hearing protection when using the meter, the fan is rather loud.