Siglent Bench DMMs SDM3065X, SDM3055 and SDM3045X

[Martin72]

I definitely want a DMM with a scanner card. 2 bench DMMs isn't enough for me, and I don't think 3 will be enough either. 😉

I still have a Yokogawa MV202 here, with 20 isolated channels each.

[KungFuJosh]

I definitely want a DMM with a scanner card. 2 bench DMMs isn't enough for me, and I don't think 3 will be enough either. 😉

I still have a Yokogawa MV202 here, with 20 isolated channels each.

Is that the correct model number? google wouldn't show me anything.

[Martin72]

Pardon, MV220...
https://www.ebay.de/itm/115342146666

[KungFuJosh]

Does that data logger have DMM functions? I want the scanner card to setup multiple voltage test points with a single DMM.

[Martin72]

With the successor model (MV2000), we record voltages and temperatures, which can be displayed numerically or graphically.
We use the device for multi-day burn-in tests or heatrun tests.

[KungFuJosh]

I'm considering getting an SDM3065X-SC, and if I do, I want to make a simpler version of pipe2null's breakout box.

Is it possible to use similar OLED's to what he used specifically for showing DC voltages on each external channel? Or will that have a negative effect on accuracy or something else annoying?

If it's not a big issue, then does anybody have a recommendation for displays?

ETA: Another question: is there any reason to not shield the enclosure for the breakout?

Thanks,
Josh

[Kleinstein]

An extra display would be more to the outside of the DMM. It would be a rather separate project, more like a Raspberry doing logging via the LAN / RS232 interface. So the display choice would follow the computer used for logging.
The displays in the pipe2null case are for extra power supplies, not the meter reading.

There would be nothing wrong with also shielding (e.g. connected to the case) the break out box. With so many terminals it is is the question if the classic 4 mm plugs are the right choice - smaller connector(s) or screw terminals could be more suitable. The maximum voltage of the SC version is anyway limited.

[KungFuJosh]

Thanks, that's an interesting idea to setup logging via LAN with a dedicated screen on a breakout box. Certainly more realistic than having voltmeter displays that are accurate and small enough to be built in.

Looks like 2 projects indeed. 😉

Thanks,
Josh

[TheDefpom]

Back to the drift issue mentioned, my testing showed that you need at least 4 hours in a 23 degree C temperature controlled room to get the 3065X stabilised to an acceptable level (still a very small amount of drift) if I was doing something like trying to calibrate the meter I left it overnight so it was on for 8 hours, then it was even better and in my opinion fully stabilised.

The environment absolutely plays a part in the stability of these units, if there is other equipment nearby heating the air that the meter is sucking in, it will cause more drift, they are sensitive to the ambient environment, especially if you are getting down to the uV levels of measurement, if the room is not temperature controlled then the meter will also drift because of the ambient temperature changing throughout the day.

[Martin72]

This is normal behavior for precision measuring instruments.
You have to find a point for yourself where you don't mentally drift away, with all the drifting.

[KungFuJosh]

Damn it. I'm glad I read the datasheet first- the scanner card can't do more than 110VDC per channel. Oh well. Guess I'll probably need to save some money and get another regular bench DMM. 🤷

[Tjuurko]

I'm glad I read the datasheet first- the scanner card can't do more than 110VDC per channel.

A small note: the scanner board is directly connected to the Signal Input Terminals ("200V Max" and "1000V Max").

[tautech]

I'm glad I read the datasheet first- the scanner card can't do more than 110VDC per channel.

A small note: the scanner board is directly connected to the Signal Input Terminals ("200V Max" and "1000V Max").

Yet note the small clearances on its inputs:



There is good reason the recommend input voltages are rated as such.....

[Kean]

The scanner card looks simple enough that you could design your own version with less channels but better rated relays/connectors and greater clearances...

If I didn't already have 3 x Agilent 34970A and 1 x 34972A (and a bunch of modules) then I'd likely buy an SDM3065-SC...

I don't think I have ever measured anything over 100V with my DAQ/DMM mainframes, let alone the 34901A 300V rating.

[kblue]

Looking forward buying the SDM3065X (w/o SC Option as i dont need it). Anything special i should be totally aware of? Any nice modifications like "give the reference some cotton-shield" or whatnot? Or should i wait for the new "6 1/2 gen"?

Another option would be some company / person who would align and test/calibrate my old solartron 7 1/2 digit DMM It does for sure need alignment, i've tested the unit with known references like DMMcheck etc pp and it's a bit out of spec. Sadly it does not have "artifact calibration" or whatever it's called and needs for every-single-range a specific value.

While reading some posts in this thread, i was wondering about the update process. It's not like ... load the latest firmware and go for it? Does it need to be in "order"?

Thanks in advance.

[KungFuJosh]

The scanner card looks simple enough that you could design your own version with less channels but better rated relays/connectors and greater clearances...

That would be a fun. If the regular version supported the SC, then it would be a great project. Since it doesn't, I would need to find a crazy deal on a used SC model, which could eventually happen...but since I only need 3 meters, 1 more meter will do the trick anyway.

However, since it's a potentially fun project, it might be interesting to make the project both for a more robust SC, and also for conversion of a normal SDM to an SC version.

Does anybody have a scanner card they can share super detailed pictures of?  Also images of the inside of the SC model for the missing hardware and other connections?

Thanks,
Josh

[KungFuJosh]

Looking forward buying the SDM3065X (w/o SC Option as i dont need it). Anything special i should be totally aware of? Any nice modifications like "give the reference some cotton-shield" or whatnot? Or should i wait for the new "6 1/2 gen"?

Let the meter run for many hours (500 or more) to age the reference a little. Also review previous posts discussing warmup time. It's a great meter, which is why I want a couple more. 😉

[Martin72]

The scanner card looks simple enough that you could design your own version with less channels but better rated relays/connectors and greater clearances...

If you're unlucky, that won't help.
According to the SC1016 data sheet, the maximum available voltage range is 200V.
It is not entirely unlikely that everything above 200V will then be displayed as an OL.

[KungFuJosh]

If you're unlucky, that won't help.
According to the SC1016 data sheet, the maximum available voltage range is 200V.
It is not entirely unlikely that everything above 200V will then be displayed as an OL.

You're right, it's probably a pointless endeavor. I asked a rep for the schematic (knowing the answer would be a fast no), and he said: "The schematics and the control logic for channel input selection are considered proprietary and not made available."

I doubt it would be terribly hard to reverse engineer the SC board, but if the end result is an OL, certainly not worth it.

Anyway, I'll probably have one next week to take a closer look at. 😉

Thanks,
Josh

[Kean]

If you're unlucky, that won't help.
According to the SC1016 data sheet, the maximum available voltage range is 200V.
It is not entirely unlikely that everything above 200V will then be displayed as an OL.

I had not considered that the firmware might limit range selection when using the SC.
There might be a limit placed by the module plug-in connector clearances.

I am pretty sure there are internal photos on one of the SDM topics here showing the additions for the SC model.

[Kean]

Anyway, I'll probably have one next week to take a closer look at. 😉

Hmmm... if that is the case, let me know how I can make a small donation to the Josh SC reverse engineering fund.

[KungFuJosh]

Minimum donations are $5000 USD. 

[Kean]

Minimum donations are $5000 USD. 

LOL.  OK, I'll just buy my own SDM3065X-SC.  Would be my first bit of Siglent gear.  I also have an SDG2082X in my shopping cart. 

[KungFuJosh]

Minimum donations are $5000 USD. 

LOL.  OK, I'll just buy my own SDM3065X-SC.  Would be my first bit of Siglent gear.  I also have an SDG2082X in my shopping cart. 

For $5k, you should be able to buy 5. 😉

[Johnny B Good]

 Don't waste time trying to reverse engineer this upgrade, it can't be done. The standard SDM3065X doesn't have the interface components installed which is what makes it a factory only option selected when you purchase it.

 Oh yeah, the other point I needed to make is that the voltage reference is fully warmed up and stable some 6 seconds before the meter has finished booting up. There's no benefit to be had in covering it with extra insulation.

 The remaining drift is simply due to the time needed for the rest of the mainboard components to stabilise some 3 or 4 degrees above the current ambient temperature. That "noisy" cooling fan ensures a relatively fast warm up ( in my experience, little more than 20 minutes for the LSD to settle to within a  a couple of adjacent values of the ultimate standstill value reached about half an hour later).