Picotest M3510A Bench Multimeter

[David_AVD]

Has anyone here bought a Picotest M3510A bench meter?

It seems to be pretty good value for money and has USB interface I need for automated testing.

Picotest is not a brand I've bought before, so just looking for feedback from those that have (or decided not to).

[free_electron]

I have been 'torturing' one for the last four days.

The same machine as a keithley 2100 or Array m3500. Same parent company , just a rebrand.

The good points :
- cheaper than a 34401
- lower noise (volts DC)
- faster ( about 10x. 1NPLC is equivalent to 10NPLC of a 34401 )
- identical SCPI command set of a hp34401.
- the people at picotest are very responsive and i typically get the local guys answer in 15 minutes. if they can't figure it out the guys in taiwan get back to me within one day.

the bad points :
- Communication problems. The machine sometimes locks up under remote control.  Especially when running long command sequences. The machine becomes unresponsive to remote commands and does not answer at all. Only a power cycle exits it from that state. This pretty much kills it for me as i want to drive this under software control. I have done a couple of day long runs where the machine blasts data at full throttle over GPIB and it locks up on averag once every 2 to 3 hours. So this scraps it for me a i need to run acquisition over several days. Same software code ( GPIB command sequence ) works fine on 34401 , 34410 and 34411. Even on a keithley 2002. Picotest locks up. It' like it receives the trigger, does the acquisition ( 100 samples in math mode ) , does the calculations (min max avg and stdev ) and then when i query for the result it simply goes quiet. no answer. any subsequent command after that does nothing apart from 'beep'.

Picotest claims it works better over usb but that is not an option for me as i use some special capabilities of GPIB like group triggers (telling mulitple machines to take a measurement synchronized in time under control of a GPIB bus line. Which cannot be done over USBTMC. ( USBTMC emulates this by sequentially sending commands to the machines in the group but that injects a time uncertainty / delay between the actual execution of the command. I need exact timing )

Note that i am not running the latest firmware but one older stepping.  This was the machine they sent me on demo.
i will wrap up my experiments tomorrow. The Rigol is the next on the torture test.

[David_AVD]

Hmmm.... since I want to use the automation aspect over longer periods, it needs to be reliable.

That said, I would be using the USB port, not GPIB and synchronised triggers are not important for my application.

[David_AVD]

I've been in contact with a supplier and they have some M3510A and M3511A model units that are "New and in stock but cal is over a year old".

They are offering these units at a discounted price (about 25% off).

I have asked about the firmware and they say they are not the latest version.  They did give me a link to the latest files so could upgrade it myself.

The overall functions and PC connection is what I'm after.  I don't need mega high accuracy.  Can anyone see a good reason not to buy one of these discounted units?

[robrenz]

Check on the cost of a full calibration. It could easily exceed that 25%.

[David_AVD]

I wouldn't think the calibration will be out by to a degree that I'd be unhappy with.

The main reason for the model choice is the features, not the absolute accuracy.