BK Precision 2712 DMM

[carloscuev]

Hello, I bought this DMM, which looked one step further from the 2709B that Dave reviewed, it is different in many aspects, pics speak for themselves:



Front:


I prefer Rel button than Hold button, the continuity tester is not as fast as 2709B.

If you are looking to buy this and have a question about the meter please PM me.

[EEVblog]

Nice photos, thanks for sharing.
The choice of Hold function instead of Rel is a big fail IMO.
The Extech EX320 has the same thing, that they swapped on the 330.

Dave.

[House91320]

i think it has a 555 timer in there

[carloscuev]

The choice of Hold function instead of Rel is a big fail IMO.

I agree, completely.

[ejeffrey]

Nice photos, thanks for sharing.
The choice of Hold function instead of Rel is a big fail IMO.

This is an especially weird choice specifically for a high resolution 40000 count meter where you are more likely to want to zero out a background reading!  Makes no sense.

[carloscuev]

Here's the calibration manual provided by BK for this multimeter, sadly I don't have everything needed to calibrate all ranges might be useful to somebody else !

[sorenkir]

Bought one, used but as new, a few days ago (for 30 euros delivered, I should post in 'scores' thread!).
Here is a photo of the PCB:

Cyrustek ES51997 (60000 count ADC, but the meter is "only" 40000 counts!?)
Synctec STM1105 (could not find any data on this microcontroller chip)
Two trimming potentiometers, but they are not described in the calibration procedure!
Input protection and build quality are very good.
Michel.

[David Hess]

Meters like this are on my short list of candidates because of their superior terminal arrangement.

Unfortunately the 2712 like most autoranging meters has variable input resistance.  None of the B&K meters with 10 Megohm inputs are more than 3-1/2 digits. 

[bdunham7]

Meters like this are on my short list of candidates because of their superior terminal arrangement.

Unfortunately the 2712 like most autoranging meters has variable input resistance.  None of the B&K meters with 10 Megohm inputs are more than 3-1/2 digits. 

My BK 391A (manual ranging) is 20000 count and 10M on all ranges and it gets pulled out most often for that specific reason.  Most people looking for a low-priced hand held DMM probably don't care, but obviously voltage dividers and HV probes do! 

[David Hess]

Meters like this are on my short list of candidates because of their superior terminal arrangement.

Unfortunately the 2712 like most autoranging meters has variable input resistance.  None of the B&K meters with 10 Megohm inputs are more than 3-1/2 digits. 

My BK 391A (manual ranging) is 20000 count and 10M on all ranges and it gets pulled out most often for that specific reason.  Most people looking for a low-priced hand held DMM probably don't care, but obviously voltage dividers and HV probes do!

The 391A is not on my short list because it uses the inferior linear terminal arrangement.  When I finally buy a new meter, I may have to just live with that but it opens up lots of options.

Right now the best candidates I know of are the Amprobe 37XR-A and 38XR-A.  They are better than my Beckman RMS225 which I like so much but not better than my Tektronix DMM916 (Fluke 87 clone) except in terminal arrangement.  They are autoranging meters however so I am somewhat dubious of their claims of 10 megohm input resistance.  It is something I need to verify.

[bdunham7]

The 391A is not on my short list because it uses the inferior linear terminal arrangement.  When I finally buy a new meter, I may have to just live with that but it opens up lots of options.

I try not to be too picky because it is easier to get bargains that way.  I only like BK stuff if and when I can get it cheap--I paid $69 for my 391A NOS/NIB.  Why do you dislike the linear terminal arrangement?  I've never had a meter any other way but I don't see what the advantage would be.

[David Hess]

I try not to be too picky because it is easier to get bargains that way.  I only like BK stuff if and when I can get it cheap--I paid $69 for my 391A NOS/NIB.

My current meters still work fine and are just getting old so I can be picky.

Why do you dislike the linear terminal arrangement?  I've never had a meter any other way but I don't see what the advantage would be.

The "star" terminal arrangement with common in the center allows the use of standard 3/4" banana plugs for all inputs.  On my DMM916 with its linear terminal arrangement, I cannot use a 3/4" banana plug in 10 amp mode which is really annoying.

[Ordinaryman1971]

BK meters are really nice, the only thing that bothers me a bit is a beeping sound they make when they about to turn off in auto mode, and the sound they make when switching ranges....
I have 393... not sure if anybody has any experience with this one. I think that is the highest model for the handheld.
Feels solid, like the design... expensive though.

[jfiresto]

I have a B&K 390A, the model below the 393, which is also quite nice. Instead of true-RMS AC, it provides peak reading of voltage and current pulses. Both meters have data logging which I find works well. I have written a command line python utility to capture the data from mine. Its help text suggests the logging's scope:

Read and output measurements from a B&K Precision 390A DMM  (v1.0)

SYNOPSIS
    bk390 [-cDEehTtuv] [device_pattern]

DESCRIPTION
    With decreasing priority, read the character device whose device
    name or port description matches:
        - the regular expression. <device_pattern>, if given,
        - the BK390_PORT environmental variable, if it exists,
        - or 'ttyusb|usb |(serial port|uart)'.
    Ignore upper/lower case and use choice three for an empty pattern.

    Read the device and output its valid measurements to standard out,
    one to a line. If the DMM battery is low, output the measurement
    value with a superfluous leading zero (e.g., 03.4 or -00.7e-3). If
    the input is out of range, output a value of -9999e9 or 9999e9. If
    the DMM is in PEAK mode, suffix each minimum value and prefix each
    maximum value, with a tab, so that they zig-zag down a page.

OPTIONS
    -c  output commas before and between fields, rather than tabs
    -D  output extra, ..-prefixed debugging lines, and a blank line
        after each, valid measurement
    -E  prefix each measurement with the elapsed time in seconds since
        the first measurement, plus a tab:
            0.00    234.5
            0.75    234.6
    -e  prefix each measurement with the elapsed time in seconds since
        the program started, plus a tab:
            0.44    13.54
            1.19    9999e9
    -h  print this help text on the standard output and exit
    -T  prefix each measurement with the date and time in extended
        ISO 8601 format, plus a tab:
            2014-04-15T07:47:22.40  12.70
    -t  prefix each measurement with the time in seconds since the
        epoch, plus a tab:
            1397540842.40   12.70
    -u  suffix each measurement with a tab plus the measurement's units:
            A_AC       V_AC       cont       ohms
            A_DC       V_DC       deg_C      hz
            F          V_diode    deg_F
        If the DMM is autoranging, add a trailing star, '*':
            234.5   V_AC*
            234.6   V_AC*
    -v  output each measurement, again, to standard error

DIAGNOSTICS
    Quietly exit with 0 on a KeyboardInterrupt (SIGINT); noisily exit
    with 1 on an error.

BUGS
    This program ignores the DMM for the first 100 ms -- and the
      possible, stale serial data in the que when the program started.
    pyserial 2.7 does not pass on a Windows Ctrl-C until it has
      received another line from the DMM. It also does not report OSX
      or Windows serial data parity errors.
    Measuring the voltage of an open circuit can make the DMM
      frantically autorange and stop transmitting.

For completeness, here is the inside of a 390A:

[David Hess]

I have a B&K 390A, the model below the 393, which is also quite nice. Instead of true-RMS AC, it provides peak reading of voltage and current pulses.

I would not mind finding a handheld meter which can read the actual peaks with a solid specification but all of the ones I have tested are slow and have vague specifications.  I have pretty much given up on this.

[jfiresto]

I would not mind finding a handheld meter which can read the actual peaks with a solid specification but all of the ones I have tested are slow and have vague specifications.  I have pretty much given up on this.

Here are the specs for the ES51962 DMM chip used in the B&K 390A:

[shakalnokturn]

Mine arrived from the same source as Sorenkir's. Looks like new too, I'll have to check it against a better meter for calibration though.
A few impressions on the thing...

Cute little DMM with a size that fits the hand really well.
The rotary selector has a nice feel, would have been nicer to handle if the pointing end had been a little higher.
The magnet is convenient and just about enough to hold the meter in place, a little stronger would have been better. The iron washer in the tilt stand to keep the magnet stuck to the back of the meter is a fail, just not big enough.
The tilt stand has a good angle but feels flimsy, a hand strap would have been nice.
Overall build quality is towards the minimum, 3 screws to hold the shell together, some torsion on it and you hear the two halves squeaking at each other.

LCD display is very readable although not the best I've seen for viewing sideways, the backlight is good.
It really is a pity for the "Hold" key not being a "Relative" one, any chance of correcting that by unused connections on the PCB?

The PCB layout is more than neat enough to my likings, without looking into routing criticisms.

It seems that BK precision sell rebranded Chung Instrument (CIE) meters, a couple of decades ago I may have hesitated who rebranded who, not any more.
Equivalent to the BK 2712 model is the CIE-3140B top of the family, there is also a CIE-3130B model with 22k counts and a "Relative" key rather than "Hold"...

The 22k count 3130B has better DCV accuracy than the 40k count 3140B. How much sense does that make?

[shakalnokturn]

I got hold of a older BK 2712 that funnily has quite different innards, uses a different ADC and MPU. The MPU confirms that CIE is the manufacturer, it has their name on it...

I went looking for a way to add or modify the "hold" key into a "relative" key on both models, it doesn't look possible.

[David Hess]

I wish BK made a higher performance version of their manual ranging meters.  They are all stuck at 3-1/2 digits.