Siglent Bench DMMs SDM3065X, SDM3055 and SDM3045X

[KungFuJosh]

When are we finally going to see a firmware update for these? We're coming up on 3 years since the last update for the SDM3065X.

[Martin72]

There are three different types of updates.
Pure bug fixing, feature enhancements, or both.
If there are no more serious bugs, I don't think there's much point in “investing” in enhancements.
Especially since the new 4000 series is out.
Of course, I could be wrong.

[KungFuJosh]

There are three different types of updates.
Pure bug fixing, feature enhancements, or both.
If there are no more serious bugs, I don't think there's much point in “investing” in enhancements.
Especially since the new 4000 series is out.
Of course, I could be wrong.

There's an update that's been due for a while. At the very least it's supposed to add compatibility for their desktop app.

[Salient666]

I wish at least they would update the 3065x to a version which eliminates the pulsating power LED in standby mode.

[Johnny B Good]

 What "Standby Mode" is this? I have never seen a standby mode option and checking the last two user manual versions I have to hand give no hint of such an option. The power indicator LED, whilst excessively bright (use of a black marker pen cures this), is completely extinguished when the meter is switched off.

P.S. I forgot to mention that my bench meter is the SDM3065X model. This might explain the difference.

 Also, on closer examination of the power button this morning, I see that, despite its emulation of a simple on/off toggle switch like those used by the SDL1020X-E and the SDG2042X, that it is in fact a momentary on/off switch.

 Presumably, if there is a later software update available and I upgrade it, I too might experience the pulsating power led effect (although not quite so annoying in my case on account of the black marker pen 'mod')

[Salient666]

Mine has the latest available firmware (as far as I know of) and when you switch it "off" (which is not actually off, its a soft-button) the power LED will pulsate/"breathe". I find that annoying. My 3055 didn't have that anymore with the latest firmware available for that model (which is I believe a year newer than the latest for the 3065X). I don't know how it is with the 3045X though.

[Peter_O]

My 3055

HW Version 03-00-00-13-00
SW Version 1.02.01.28 (latest)

does show the pulsing LED.
And yes, it would be nice to have that configurable.  But it's not that a point, as my lab runs behind a switched power bar anyway.

[InvisibleMan]

My 3055

HW Version 03-00-00-13-00
SW Version 1.02.01.28 (latest)

does show the pulsing LED.
And yes, it would be nice to have that configurable.  But it's not that a point, as my lab runs behind a switched power bar anyway.

Strange, I have the same SDM3055 with same HW and SW. My LED is off when the DMM is not switched on.

[KungFuJosh]

My SDM605X LED is also off when the DMM is off. My KS34465A and DAQ6510 both have amber LEDs lit when off. They're all connected to a PDU connected to a UPS anyway, so not a big deal here either way.

[TheDefpom]

I just wanted them to add the support for the 30A current shunt to the 3065X so it shows up in current mode like was done for the 3055 for proper readout, although with that said, I have moved the 3065X into storage and replaced it with the Keithley DMM7510, so that is not so important to me any more.

[mawyatt]

From this thread:

https://www.eevblog.com/forum/testgear/new-siglent-multimeter/msg6172347/#msg6172347

We can't seem to get the SDM3065X Histogram to display a typical graph, and in this example using Zero DCI (Input Open Circuit), with plenty of samples. We've tried Auto and Manual modes with various settings but unable to get a respectable Histogram.

The SDM is running the latest firmware all all the meters have over 1hr warmup. Note the DMM and KS display typical Histograms in Auto mode as expected for Zero DCI.

Any guidance, help or comparisons are appreciated.

Best

[mawyatt]

Just powered SDM3065X off and back on. The Histogram Auto Mode shows the same type display where the high-low range is much too high, same as before. However the Manual Histogram Mode settings now seems to display a better result altho shows some "zero bins" within the typical Bell display shape which is puzzling? These are equally spaced which hints that's this a computational error rather than actually not having ADC sample values within these bin ranges.

Edit: Likely the range is too small for the Histogram display for the lowest current range of +-200ua, as 10na is ~20k smaller than the total range.

Best

[Grandchuck]

Yep ... equally spaced.  Rounding error or resolution?

[Salient666]

Yep ... equally spaced.  Rounding error or resolution?

I think indeed rounding errors. I experience the same kind of behaviour when I am logging stuff on my Siglent 3065X vs. the Keysight 34465A and my Picotest M3500A.
Attached graph makes this probably a little more clear. Notice the flat pieces in the trace from the 3065X compared to the other ones.

[KungFuJosh]

Just powered SDM3065X off and back on. The Histogram Auto Mode shows the same type display where the high-low range is much too high, same as before. However the Manual Histogram Mode settings now seems to display a better result altho shows some "zero bins" within the typical Bell display shape which is puzzling? These are equally spaced which hints that's this a computational error rather than actually not having ADC sample values within these bin ranges.

Edit: Likely the range is too small for the Histogram display for the lowest current range of +-200ua, as 10na is ~20k smaller than the total range.

Best

Turn off auto zero and/or reduce the aperture setting.

Comparison in a range with less sensitivity will have less data than running the other devices at higher sensitivity. Keithley/Keysight both set to 100uA, and Siglent set to 200uA in the photo.

Also note that the Keithley has more options for how the histogram is displayed/functions.

Thanks,
Josh

[mawyatt]

Josh,

Here's a result with Auto Zero OFF. This makes the Average magnitude slightly higher (as expected) and SD a little lower, but still displays the empty bins as shown.

Best

[KungFuJosh]

Here's a result with Auto Zero OFF. This makes the Average magnitude slightly higher (as expected) and SD a little lower, but still displays the empty bins as shown.

What NPLC do you have set? I haven't been able to get the empty bins to show, but I get very little data on there if the NPLC isn't set fast like .06 or .006. You must have more interesting noise.

ETA: You should also note along with my previous comment about higher resolution coming with more noise, that the bins are also different. You're showing 400 bins on the KS and 100 on the Siglent. Assuming auto, I also assume that's because of the extra noise data. I would manually set the bins and the ranges closer together, unless the comparison is less interesting than the gaps.

Also, I think it's fairly safe to assume there's always some level of averaging happening on the SDM. Or at least that's what it looks like when tracking data for a long time.

[KungFuJosh]

Here's a couple examples why I think there's averaging (not that we have any choice that I'm aware of to set sample numbers or anything like that).

[mawyatt]

Here's a result with Auto Zero OFF. This makes the Average magnitude slightly higher (as expected) and SD a little lower, but still displays the empty bins as shown.

What NPLC do you have set? I haven't been able to get the empty bins to show, but I get very little data on there if the NPLC isn't set fast like .06 or .006. You must have more interesting noise.

ETA: You should also note along with my previous comment about higher resolution coming with more noise, that the bins are also different. You're showing 400 bins on the KS and 100 on the Siglent. Assuming auto, I also assume that's because of the extra noise data. I would manually set the bins and the ranges closer together, unless the comparison is less interesting than the gaps.

Also, I think it's fairly safe to assume there's always some level of averaging happening on the SDM. Or at least that's what it looks like when tracking data for a long time.

NPLC is 10 in all the examples we've shown, for all meters. The SDM3065X Bins are set to 40, which is what the DMM6500 was also set too. The KS34465A selects the Bin size when in Auto Mode which is what we had shown.

BTW we are more interested in seeing how the SDM3065X compares with the new SDM4000 (why we started this investigation over on the SDM4000 thread), we already know how the SDM3065X compares with the KS34465A and DMM6500 (did such long ago!). The SDM3065X Auto Histogram mode seems to scale the range much too wide and thus only shows a narrow sliver and why we just showed the DMM6500 an KS34465A for reference of what a typical Auto Histogram looks like.

Best

[KungFuJosh]

BTW we are more interested in seeing how the SDM3065X compares with the new SDM4000 (why we started this investigation over on the SDM4000 thread), we already know how the SDM3065X compares with the KS34465A and DMM6500 (did such long ago!). We just showed the DMM6500 an KS34465A for reference of what a typical Histogram looks like for others.

That makes sense. I'm curious about the SDM4065A as well. I'd like to know if it also has some sort of averaging applied like its little brother appears to have. Either way I want one.

[mawyatt]

BTW we are more interested in seeing how the SDM3065X compares with the new SDM4000 (why we started this investigation over on the SDM4000 thread), we already know how the SDM3065X compares with the KS34465A and DMM6500 (did such long ago!). We just showed the DMM6500 an KS34465A for reference of what a typical Histogram looks like for others.

That makes sense. I'm curious about the SDM4065A as well. I'd like to know if it also has some sort of averaging applied like its little brother appears to have. Either way I want one.

We made some edits above, see previous post.

The Display (Trend and Histogram) of the SDM3065X seems limited compared to other lab grade DMMs, and doesn't have the overall ranges of these instruments.  It's an OK instrument and value considering the price tho.

Best

[KungFuJosh]

The Display (Trend and Histogram) of the SDM3065X seems limited compared to other lab grade DMMs, and doesn't have the overall ranges of these instruments.  It's an OK instrument and value considering the price tho.

Indeed. Certainly some room for improvement, but IMO I think it's pretty close to the Keysight. I have not been able to get the gaps you have yet tho.

I think the Keithley destroys them both by comparison. Knowing that Siglent very clearly copied the Keithley for the 4000 series platform, I'm curious if they copied the trend/histogram too.

I'm running a test now with 10VDC to compare with something more reliable than 0 DCI current noise.

[KungFuJosh]

I don't normally use the histogram, so I hadn't really seen the defecit yet. Running a 10VDC test made it very clear.

I stand by my opinion about the Keithley being by far superior in this regard, but the Keysight isn't too bad. I like the visual range/averaging stuff the KS shows, and it also has cursor options. Keithley has a cute peak highlight when it's less obvious where the average is. The Siglent is not keeping up with this feature set at all, and the histogram itself is lacking. Maybe that's what's holding up the next firmware update?

Thanks,
Josh

[KungFuJosh]

One other thing that I think is sad about the 3000 series is the lack of even a basic web server. The SDM3000X series was an obvious copy of the Keysight 3446XA series, so why didn't they copy that too??

I do think it's funny tho that KS didn't bother to update the virtual console at all.

[KungFuJosh]

Histogram looks better in Temp, though they still do a poor job of using the Y axis/vertical space.