Meterman 38XR

[DJPhil]

After much debate with myself over how best to spend my small chunk of disposable income I picked up a 38XR. It smells great!

After Dave's review of the 37XR, I figured this was worth a careful look when I saw it selling cheap. I reviewed the user manuals online (37XR and 38XR) and here's a short version of the comparison.

37XR	38XR
Inductance	Temperature
TTL/CMOS Logic Tester	4-20mA Loop Current
Improved Accuracy	RS232 IR for datalogging
	AC+DC Measurement Mode

For the detailed differences I copied the specs to text files and recorded the differences. Below is a reformatted summary of the differences between those features shared by both meters. I know it's big, but I included it to save anyone who needs it the trouble of the pdf file tennis match I did comparing specs.

37XR	38XR
DC VOLTS
Accuracy: ±(0.1 % rdg + 5 dgts)	Accuracy: ±(0.25 % rdg + 5 dgts)
AC VOLTS
Minimum reading on 1000mV range: 14 mV	No such entry (weirdness)
DC CURRENT
Resolution: 0.01 µA	Resolution: 0.1 µA
...ceramic fuse 10×38 mm on 10A input	...ceramic fuse 10×38mm on 20A input (just a typo)
Burden Voltage
µA Range: 1 mV/ 1 µA	µA Range: 1 mV/ 1 µA
mA Range: 10 mV/ 1 mA	mA Range: 1 mV/ 1 mA*
A Range: 35 mV/ 1 A	10A Range: 30 mV/ 1 A
	*Included with meter is an errata notice stating this should read 11mV/mA, which sucks
AC CURRENT
±(1.5% rdg +10 dgts) on 100µA to 100mA ranges	True rms/±(1.5 % rdg + 20 dgts) on 100 µA to 100 mA ranges
±(2.0% rdg +10 dgts) on 400mA range	True rms/±(2.0 % rdg + 10 dgts) on 400mA range
±(2.5% rdg + 20 dgts) on 10A range	True rms/±(2.5 % rdg + 20 dgts) on 10A range
Peak Hold accuracy: ±(3.0 % + 200 dgts) 100µA range unspecified	Peak Hold accuracy: ±(3.0 % + 200 dgts)
	AC coupled true RMS specified from 10 % to 100 % of range
FREQUENCY
Resolution: 0.01 Hz	Resolution: 0.1 Hz
DIODE TEST
Test current: 1.0 mA (approximate)	Test current: 0.5 mA (approximate)
DUTY CYCLE
Frequency range:	Frequency range:
0% to 10% (40 Hz to 990 Hz)	0% to 10% (40 Hz to 20 kHz)(Typo?!)
10% to 90% (40 Hz to 20 kHz)	10% to 90% (40 Hz to 990 Hz)(I hope so!)

The manual for the IR communication addon is available for the Meterman version (Meterman 38SW), but I haven't found one for the Amprobe 38SW-A. I assume it's the same, I was told by an Amprobe rep that it was compatible, and retails for around $25US. I definitely plan on tinkering with the RS232 out before shelling out for the kit. When I took it apart (of course) I didn't see a receiver in the meter, so I'm guessing it's one way IR, and the book says 9600 8,N,1 (maybe odd parity too), so I can't imagine there's much in that RS232 adapter they're selling. There's a post on the LabView forums where someone's worked out the serial output codes, so I think I can figure something out.

I ordered it from these guys off of ebay after emailing them to see how low I could get their 'Make An Offer' price. The nice lady I spoke with in email said the lowest they could go was $65US and free UPS Ground, so I jumped on it. It just arrived today after being shipped through two blizzards and an earthquake (in Chicago of all places), and I think that could be good or bad.
I spoke with an Amprobe rep on the phone (they bought Meterman some time ago) and was told that Amprobe won't honor the warranty if it wasn't purchased from an authorized dealer. At this deep a discount I was willing to take a calculated risk, but be advised.

Sorry, I know it's disjointed, but I don't organize well. 
Questions welcome. If anyone's interested I'll dig out a camera and take pictures of the guts.

Hope this helps any who might be agonizing over the midrange meters a bit.

[rossmoffett]

Have you messed with the 4-20 mA loop current?  What's its resolution?

That's an extremely useful feature for Instrumentation and PLC troubleshooting.  I'd love to have a meter that could do loop power for $60! 

[EEVblog]

After much debate with myself over how best to spend my small chunk of disposable income I picked up a 38XR. It smells great!

After Dave's review of the 37XR, I figured this was worth a careful look when I saw it selling cheap. I reviewed the user manuals online (37XR and 38XR) and here's a short version of the comparison.

37XR	38XR
Inductance	Temperature
TTL/CMOS Logic Tester	4-20mA Loop Current
Improved Accuracy	RS232 IR for datalogging
	AC+DC Measurement Mode

For the detailed differences I copied the specs to text files and recorded the differences. Below is a reformatted summary of the differences between those features shared by both meters. I know it's big, but I included it to save anyone who needs it the trouble of the pdf file tennis match I did comparing specs.

37XR	38XR
DC VOLTS
Accuracy: ±(0.1 % rdg + 5 dgts)	Accuracy: ±(0.25 % rdg + 5 dgts)
AC VOLTS
Minimum reading on 1000mV range: 14 mV	No such entry (weirdness)
DC CURRENT
Resolution: 0.01 µA	Resolution: 0.1 µA
...ceramic fuse 10×38 mm on 10A input	...ceramic fuse 10×38mm on 20A input (just a typo)
Burden Voltage
µA Range: 1 mV/ 1 µA	µA Range: 1 mV/ 1 µA
mA Range: 10 mV/ 1 mA	mA Range: 1 mV/ 1 mA*
A Range: 35 mV/ 1 A	10A Range: 30 mV/ 1 A
	*Included with meter is an errata notice stating this should read 11mV/mA, which sucks
AC CURRENT
±(1.5% rdg +10 dgts) on 100µA to 100mA ranges	True rms/±(1.5 % rdg + 20 dgts) on 100 µA to 100 mA ranges
±(2.0% rdg +10 dgts) on 400mA range	True rms/±(2.0 % rdg + 10 dgts) on 400mA range
±(2.5% rdg + 20 dgts) on 10A range	True rms/±(2.5 % rdg + 20 dgts) on 10A range
Peak Hold accuracy: ±(3.0 % + 200 dgts) 100µA range unspecified	Peak Hold accuracy: ±(3.0 % + 200 dgts)
	AC coupled true RMS specified from 10 % to 100 % of range
FREQUENCY
Resolution: 0.01 Hz	Resolution: 0.1 Hz
DIODE TEST
Test current: 1.0 mA (approximate)	Test current: 0.5 mA (approximate)
DUTY CYCLE
Frequency range:	Frequency range:
0% to 10% (40 Hz to 990 Hz)	0% to 10% (40 Hz to 20 kHz)(Typo?!)
10% to 90% (40 Hz to 20 kHz)	10% to 90% (40 Hz to 990 Hz)(I hope so!)

The manual for the IR communication addon is available for the Meterman version (Meterman 38SW), but I haven't found one for the Amprobe 38SW-A. I assume it's the same, I was told by an Amprobe rep that it was compatible, and retails for around $25US. I definitely plan on tinkering with the RS232 out before shelling out for the kit. When I took it apart (of course) I didn't see a receiver in the meter, so I'm guessing it's one way IR, and the book says 9600 8,N,1 (maybe odd parity too), so I can't imagine there's much in that RS232 adapter they're selling. There's a post on the LabView forums where someone's worked out the serial output codes, so I think I can figure something out.

I ordered it from these guys off of ebay after emailing them to see how low I could get their 'Make An Offer' price. The nice lady I spoke with in email said the lowest they could go was $65US and free UPS Ground, so I jumped on it. It just arrived today after being shipped through two blizzards and an earthquake (in Chicago of all places), and I think that could be good or bad.
I spoke with an Amprobe rep on the phone (they bought Meterman some time ago) and was told that Amprobe won't honor the warranty if it wasn't purchased from an authorized dealer. At this deep a discount I was willing to take a calculated risk, but be advised.

Sorry, I know it's disjointed, but I don't organize well. 
Questions welcome. If anyone's interested I'll dig out a camera and take pictures of the guts.

Hope this helps any who might be agonizing over the midrange meters a bit.

Crikey, at $65 I recon that's gotta be close to one of the best value meters under say $150 on the market.

I've been eyeing off sgi_direct for a while now, they sell Amprobe stuff in bulk at insanely low prices, shame about shipping prices to Australia, otherwise I'd be reselling the damn things myself!

Thanks for sharing the info.

Dave.

[DJPhil]

Have you messed with the 4-20 mA loop current?  What's its resolution?

That's an extremely useful feature for Instrumentation and PLC troubleshooting.  I'd love to have a meter that could do loop power for $60! 

I must say that I'd never heard of this feature before doing this research. I haven't used it myself (yet) so all I can do is parrot the specs as listed in the manual:

DC CURRENT (4 to 20mA)
Range: 0 to 100 %
Accuracy: ±0.5 % + 5 dgt
Resolution: 0.01 %
Burden voltage: 1 mV/mA*
Input protection: 0.5A/1000V fast blow ceramic fuse 6.3×32mm on ?A/mA input
* Errata sheet included with meter says this should read 11mV/mA (Same as with the basic DC Current reading)

I didn't include the eratta on this mode before as it was a feature only of this meter. There were two errata sheets included with the meter, one with the two burden voltage changes (DC Current and 4-20mA DC Current), and one applying to all 3?XR series meters. The second refers to the General Specifications > Operating Environment section, and notes that in the 10A ranges the operating temp should be 0-40C @<70% RH instead of 0-50C @<70%. This is one of those things I should have covered in the original post, but I'm scatterbrained.

Test proceedure:
Measuring 4-20 mA Loop Current
1. Set the Function Switch to the loop-current function.
2. Connect the Test Leads: Red to mA, Black to COM.
3. Turn off power to the circuit being measured.
4. Open the test circuit (-X-) to establish measurement points.
5. Connect the Test Probes in series with the load (to the measurement points).
6. Turn on power to the circuit being measured.
7. Read the display. 0 % = 4 mA, 100 % = 20 mA.

It seems that they essentially measure the current with the same method and accuracy as the DC Current measurement and reformat the data as a percent of accepted range. The manuals have a few odd typos, so I think there was a lot of copy/pasting going on when they put them together. I suspect that the listed accuracy percent and count refer to the mA measurement and not the displayed percent, as it is identical with the basic DC current listed accuracy (See Edit).

I'm just starting to look in to this, so I don't know if this represents a valuable feature or if it lacks in some way that would make a dedicated tester more useful.

Edit: Just rigged up a simple circuit with a target current draw within the 4-20mA range to see what it looks like. I'm using all salvage parts, so tolerances are iffy. Using min/max/avg I get 8.452mA/8.455mA/8.454mA for a minute or so of letting it ride, which is easily explained by the (15 year old AT) power supply and the meter's accuracy, among other things. I flip it over to 4-20mA mode and I get 27.82%/27.84%/27.83%. Math alone yields 27.825%/27.84375%/27.8375% for min/max/avg.

I know this is quick and dirty, but I hope it communicates what the meter actually appears to be doing.

Crikey, at $65 I recon that's gotta be close to one of the best value meters under say $150 on the market.

I've been eyeing off sgi_direct for a while now, they sell Amprobe stuff in bulk at insanely low prices, shame about shipping prices to Australia, otherwise I'd be reselling the damn things myself!

Thanks for sharing the info.

Dave.

Your welcome! Glad to share.

[rossmoffett]

Hey thanks!

That feature is for industrial electronics/instrumentation.  4-20 mA is the dominant signal type pretty much worldwide.  If you want to open a pneumatic valve with a PLC (programmable logic controller) you use a 4-20 mA output card connected to a current/pressure transducer and it will regulate the air flow to the valve to open/close it incrementally.  Pressure transmitters use it for power and signal - they burn at most 4 mA in operation and have an internal resistance that increases with pressure.  So you give them 24 VDC, then watch the current to see what their pressure is.  If it's a 0-100 psi transmitter, then 4 mA is 0 psi and 20 mA is 100 psi.  That's where the percentage is useful.. 25% would be 25 psi, it's just a convenience.

I misunderstood that it does loop power, which would make it a competitor with the Fluke 887 and 888 (which cost about as much as their model numbers).  That would allow me to disconnect power from a device, like a level transmitter or pressure transmitter, and just probe it to instantly power it up and find its current.  But like Dave said, with all of the other features alone that already makes this an awesome buy! 

Thanks a lot for going to the trouble to pull those specs for me.

[DJPhil]

Ah, I understand now. Looking this up online had the effect of filling in half the picture with tiny pieces, but this makes it plain.

Thanks a lot for going to the trouble to pull those specs for me.

Any time.

[dfnr2]

Just chiming in to add a datapoint.  When I started building up my lab in 1999-2000, I couldn't afford Flukes, I had good experience with other Wavetek equipment, so I snagged a few Wavetek meters inexpensively on Ebay.  I sent them all to Fluke (who had bought the line) to calibrate, and they lost one of the 85XT's.  They offered to replace it with the Wavetek Meterman 37XR, which was autoranging, so I said OK.  I liked that it was autoranging, had a good case and stand, nifty features, etc.  Because it was my best autoranger, I used it as my main meter, but used the wavetek 85XT or 27XT to double check.

Recently, I'd noticed differences between readings on the 37XR and the others, and I sent it in for early cal along with the rest of my DMMs.  The 37XR was far out of spec on the lower voltage ranges.  The sad thing is that there is no way to adjust this meter!  I asked the tech if this is unusual, and he that this is not too uncommon with the Ampbobes; one engineering department fitted with these meters recently brought a bunch of these scopes in for cal (that were in spec on their last cal), and half the meters were off.

Every other DMM I have had, including a wavetek 85XT, wavetek 27xt, a couple of cheap radio-shack 22-805s', and a cheap AW SPERRY DM-11, not to mention the HP 4368a and HP34401a, have always been received and returned in spec.  As far as I'm concerned, I'll never buy another one of these.  I decided to replace it with a used fluke 87v off ebay; I bought it from one of the dealers that calibrates it before shipping--well worth the money, since even a new meter will need a cal.  I really love the 87V.  Wish I'd bought it years ago.

This experience reinforces some old rules of thumb about meters:

1) If your measurements are important, always double check them with other equipment.  In fact, it's nice to check with a scope, just to make sure you're measuring what you think you are measuring.
2) If you're using the scopes professionally, or otherwise care about the accuracy, don't forget to cal your workhorse meters WHEN YOU GET THEM, and EVERY YEAR.
3) If you want a great reliable handheld, get a fluke.  If you want a great reliable benchtop meter, get an HP (or a fluke).
4) You can never have too many meters, especially workhorses you can use to check the others.  I use the HP34401a as my reference.  It's easy to use up all your good meters in one setup.

Dave F

[armandas]

Just got an Amprobe 37XR-A meter I had bought on ebay a couple of days ago.. for £18 plus shipping. It looks like new, didn't come with papers, but I got probes + alligator clips and a magnetic hanger. That's got to be the best deal I have ever had on ebay  

P.S. price may have something to with auction title ("LARGE MULTIMETER")

[peters]

I have added a driver to Sigrok for the Meterman 38XR multimeter. Go to https://sigrok.org/wiki/Meterman_38XR to read about it. To try it out  download the Sigrok nightly build  https://sigrok.org/wiki/Downloads, or compile from source if that's the way you roll.

[Zero999]

Just got an Amprobe 37XR-A meter I had bought on ebay a couple of days ago.. for £18 plus shipping. It looks like new, didn't come with papers, but I got probes + alligator clips and a magnetic hanger. That's got to be the best deal I have ever had on ebay 

P.S. price may have something to with auction title ("LARGE MULTIMETER")

I believe Amprobe bought Meterman awhile ago.

I also have the 37XR-A. It's a good meter. The build quality isn't the best. The switch began to go a bit, after five years of regular use, but it didn't stop me from buying a new one.