Siglent Bench DMMs SDM3065X, SDM3055 and SDM3045X

[Muttley Snickers]

Martin Lorton has made a couple of videos on the Siglent SDM-3055 bench multimeter if anybody is interested.

Part 1

Part 2

[verbatone]

I'll have to agree with most of Martin's points in the video.  Though I think I'll give my unit a proper review of some of the issues he mentions.

The continuity meter delay is worse at lower thresholds...this is somewhat intuitive, setting it back to 50 Ohms makes it more responsive than Martin's test shows.  I think my old Fluke uses 40 Ohms for the threshold and rarely do I fault it's continuity check.  If I really want to know how many Ohms it is, I'll flip the unit to resistance to see what it really is.

Again, I don't agree that you can fault the meter for having a slow update when you want 1s integration time.  I suppose there are ways to have a running average rather than a non-overlap average and still maintain a 1s integration time, but since the display is updating 1 time/sec, you can expect the display to update 2 times/sec.

One other thing I notices, even in the later firmware is that the dual reading between ACV and freq is dog slow.  These two measurements should not require relay switching, but still, the time to make each measurement really bogs down the machine.  Also for dual measurements, beyond the current disconnect issue point out both by Martin and the Signal Path, I can understand the latency in measurements for current/voltage measurements being slow as the relay probably requires some settling time...though I hope they can improve this in the future and not require an open circuit period.

And if Siglent can do another couple rounds (or just one good one) of firmware updates with the same improvements as R12->R13, that would be great.

Oh yeah, and Siglent, just publish a change log, it would take 30 minutes of time to make.  I'm sure you're not fixing 100's of bugs per releases.

[Armxnian]

I don't think display update rate has anything to do with integration time. NPLC only affects the internal sampling speed. I can have a device take 100 readings per second and have a display update rate of 1Hz, so I would see the 100th reading every second if no mean functionality was implemented. On the other hand, I could have a device take 1 reading per second and have a screen update rate of 100Hz, meaning the screen would show the same value 100 times for 1 second.

[verbatone]

Armxnian, while it's true what you're saying...those measurement methods could be done, I don't really see the point.  In the end, you're trying to find a settled value and the integration time has a direct correlation to the settled value of a specified period of time.  Siglent says they do 5 sample per sec on slow mode, but they only update the display about once a sec.  So there is clearly a mismatch.  If this is true, they are either averaging about 5 reading to get 1 update per display, or they're doing as you mention and only showing one value of the 5...which means they cannot resolve more than about .2s.

On other topic from Martin's video.  I don't have the voltage settling issue he notes.  I added a very large RC on the output of my supply and let it settle for a long time, then hooked up the SDM3055...it goes almost directly to the final result, no settling.  There does seem to be bit of a jump in DCV when the filter is applied, but it doesn't jump much.

In looking at some other things, I used the Easy DMM software to capture some data.  It doesn't seem to matter what capture rate I use, it's always about 50 samples per second.  Not sure what that's about, but the display on the unit and the display on the Easy DMM shows pretty much the same thing.  Strangly there is a weird bobble when the system goes through a zero crossing in slow mode, on both sides of zero.  This can be rectified by going to either middle or fast, or raising the DC offset of the system.  See the attached picture.

[Armxnian]

Since slow mode takes 5 samples/s, I don't see much reason to average the readings. Why not just display all 5 readings in 1 second? I would rather see the 5 individual "live" readings. Averaging readings to display on the screen would be more useful for faster sampling rates where displaying all of the readings wouldn't make sense, or be impossible, like in fast mode (150 samples/s).

Furthermore, fast mode introduces noise, whether it be from power line or internal components, and can be viewed in the trend plot. You also lose a digit of resolution. Increasing update speed and potentially getting a less "stable" reading is a worthy trade off, since you would be using slow mode for the most accurate and precise measurements anyway.

Most hand held meters on the market have 3-4 updates/s, and even have bar graphs with 60 updates/s. Offerings from Keysight or Keithley are capable of much higher. 1-2 screen updates for a bench meter like the SDM3055 is IMO unacceptable, especially since it is capable of sampling at 150x/s. Increasing the screen update rate to something even just a little higher like 5x/s would make the meter much more effective and a pleasure to use.

[verbatone]

Hmm, I guess we can go back and forth on this.  I still feel like there are some concepts slipping through the crack, but in general, I agree with you.  I wish the meter gave a bit more control over the update rate and the effective LPF effect.  Ideally they would have both controls exposed, but even the 34401 doesn't have this feature.

In reality, I'm less concerned that the slow mode is slow update, and more concerned that the middle and fast have basically the same screen update rate.  That means the fast function become useless unless you're logging data.

[tautech]

@Armxnian and verbatone

Thanks for your ongoing discussion, I've linked Siglent R&D to this thread.
Between you (and others), if you can belt out a wish list of improvements needed it would be appreciated. 
SDM3055 owners please.

Thanks and please continue.
 

[sanman]

I was on the fence about buying this meter. I wanted 5 1/2 digits and I liked the screen and features but, after seeing the reviews, I was worried about speed/update.

After reading that Siglent recently released updated firmware that addresses speed/update I went ahead and ordered from TEQUIPMENT. Should arrive sometime next week! 

[tautech]

Latest  R15.1 FW
http://www.siglentamerica.com/USA_website_2014/Firmware&Software/firmware/SDM3055_(V100R001B01D01P15R1).rar

[Armxnian]

Flashed my unit. With the increased update rate, it's like a completely different meter. You don't have to wait forever to get a reading on the slow option. "Middle" speed is now very useful as it gives a balance between a stable reading and high update rate. Fast mode is insane, I can rotate the dial on my dp832 and see the values change live. Looks like the feedback of others and mine paid off. 

I did notice one thing though. When you change the display mode to bar graph, trend or histogram, the meter update rate significantly slows down, back to where it was before the new firmware. I'm guessing this is due to processing limitations.

[tautech]

Flashed my unit. With the increased update rate, it's like a completely different meter. You don't have to wait forever to get a reading on the slow option. "Middle" speed is now very useful as it gives a balance between a stable reading and high update rate. Fast mode is insane, I can rotate the dial on my dp832 and see the values change live. Looks like the feedback of others and mine paid off. 

I did notice one thing though. When you change the display mode to bar graph, trend or histogram, the meter update rate significantly slows down, back to where it was before the new firmware. I'm guessing this is due to processing limitations.

Thanks for this feedback. 

More please as you find it.....
 

[Armxnian]

More please as you find it.....
 

If there are listeners I'll keep talking 

'Power on last' seems to work randomly. I can't find much of a pattern to when it works or when it doesn't other than when it does, and you don't change anything, it continues to work for a while until you shut it off with different settings.

On fast mode, the reading seems to jump around by millivolts. However, if you enable the AC filter, it settles down. There shouldn't really be any measurable (at least on a meter) ac ripple coming from a dp832. Power line noise could be a culprit since increasing the sampling rate, as in the case of fast mode, reduces normal mode rejection. But would enabling an ac filter located at the input block noise injected from the outputs? Sounds weird.

[Eric-H]

I see two strange things on my SDM3055:

1) When measuring VDC in slow mode I get different values when the filter is on or off. E.g. with the input shorted, I measure a value of 0.000 mV when the filter is off, and a value of 0.007 mV when the filter is active. I checked using the histogram display and see two sets of values, one clustered around 0.000 mV when the filter is off and another set around 0.007 mV when the filter is on.

2) When using a K-type thermocouple I measure a stable temperature of e.g. 20.0 deg C. When I then switch to VDC mode and directly back to temperature mode, I get a value that can differ between 0.5 to 3 deg C from the previous value. This occurs when the temperature mode was active for  a few (10+) minutes before switching to VDC and back. It seems as the SDM3050 does some compensations on enter temperature mode that are not repeated while the temperature mode is active. After switching from VDC and back to temperature the compensation is done again causing the measured temperature to jump a few degrees.

Has anyone else seen these effects on their units?

I hope Siglent can fix these bugs in a next firmware release. It would give me more confidence in the measured values.

[Siglent America]

Hi Eric.
Were you using the latest FW (15R1)?
It is available on the Siglent websites.

If this occured using the latest FW then I will report this to Engineering.
Thank you.

[Eric-H]

Hi,
Yes I'm using the latest firmware.
 
Thanks for the fast response!

[tautech]

I see two strange things on my SDM3055:

1) When measuring VDC in slow mode I get different values when the filter is on or off. E.g. with the input shorted, I measure a value of 0.000 mV when the filter is off, and a value of 0.007 mV when the filter is active. I checked using the histogram display and see two sets of values, one clustered around 0.000 mV when the filter is off and another set around 0.007 mV when the filter is on.

When this is obsevered do you have the Relative measurements setting On or Off?

[Eric-H]

I see two strange things on my SDM3055:

1) When measuring VDC in slow mode I get different values when the filter is on or off. E.g. with the input shorted, I measure a value of 0.000 mV when the filter is off, and a value of 0.007 mV when the filter is active. I checked using the histogram display and see two sets of values, one clustered around 0.000 mV when the filter is off and another set around 0.007 mV when the filter is on.

When this is obsevered do you have the Relative measurements setting On or Off?

Relative measurement is turned off.

[tautech]

I see two strange things on my SDM3055:

1) When measuring VDC in slow mode I get different values when the filter is on or off. E.g. with the input shorted, I measure a value of 0.000 mV when the filter is off, and a value of 0.007 mV when the filter is active. I checked using the histogram display and see two sets of values, one clustered around 0.000 mV when the filter is off and another set around 0.007 mV when the filter is on.

When this is obsevered do you have the Relative measurements setting On or Off?

Relative measurement is turned off.

As much info as possible is needed for R&D to replicate faults/bugs and initiate fixes.
Just trying to squeeze a little more from you. 

[Eric-H]

I see two strange things on my SDM3055:

1) When measuring VDC in slow mode I get different values when the filter is on or off. E.g. with the input shorted, I measure a value of 0.000 mV when the filter is off, and a value of 0.007 mV when the filter is active. I checked using the histogram display and see two sets of values, one clustered around 0.000 mV when the filter is off and another set around 0.007 mV when the filter is on.

When this is obsevered do you have the Relative measurements setting On or Off?

Relative measurement is turned off.

As much info as possible is needed for R&D to replicate faults/bugs and initiate fixes.
Just trying to squeeze a little more from you. 

To show what I found I've attached a screenshot of the effect. It shows the data gathered first with the filter off and then with the filter on. No other changes where made to the settings of the SDM3055 or the hardware setup. The data around 0 .000 mV was captured with the filter off, the data around 0.007 mV with the filter off.

To eliminate noise I've used a 4 mm to BNC converter with a 50 Ohm BNC terminator attached.

[rf-loop]

I see two strange things on my SDM3055:

1) When measuring VDC in slow mode I get different values when the filter is on or off. E.g. with the input shorted, I measure a value of 0.000 mV when the filter is off, and a value of 0.007 mV when the filter is active. I checked using the histogram display and see two sets of values, one clustered around 0.000 mV when the filter is off and another set around 0.007 mV when the filter is on.

When this is obsevered do you have the Relative measurements setting On or Off?

Relative measurement is turned off.

As much info as possible is needed for R&D to replicate faults/bugs and initiate fixes.
Just trying to squeeze a little more from you. 

To show what I found I've attached a screenshot of the effect. It shows the data gathered first with the filter off and then with the filter on. No other changes where made to the settings of the SDM3055 or the hardware setup. The data around 0 .000 mV was captured with the filter off, the data around 0.007 mV with the filter off.

To eliminate noise I've used a 4 mm to BNC converter with a 50 Ohm BNC terminator attached.

Still inside specifications.      (this do not mean it is ok)

[Armxnian]

1) When measuring VDC in slow mode I get different values when the filter is on or off. E.g. with the input shorted, I measure a value of 0.000 mV when the filter is off, and a value of 0.007 mV when the filter is active. I checked using the histogram display and see two sets of values, one clustered around 0.000 mV when the filter is off and another set around 0.007 mV when the filter is on.


Has anyone else seen these effects on their units?

I get the exact opposite. Probes shorted: with the filter off, 6uV, with the filter on, 0V. When feeding a 2Vrms AC signal with 0dcV offset, I get the normal noise of the meter with the filter off. But with the filter on I get 6-7mV. Doesn't really make much sense.

[tautech]

1) When measuring VDC in slow mode I get different values when the filter is on or off. E.g. with the input shorted, I measure a value of 0.000 mV when the filter is off, and a value of 0.007 mV when the filter is active. I checked using the histogram display and see two sets of values, one clustered around 0.000 mV when the filter is off and another set around 0.007 mV when the filter is on.


Has anyone else seen these effects on their units?

I get the exact opposite. Probes shorted: with the filter off, 6uV, with the filter on, 0V. When feeding a 2Vrms AC signal with 0dcV offset, I get the normal noise of the meter with the filter off. But with the filter on I get 6-7mV. Doesn't really make much sense.

Don't you mean uV?

[Armxnian]

Don't you mean uV?

Nope.

[tautech]

Continuity test

I wanted to quantify the continuity test of the SDM3055 with the lastest 15R1 firmware.
Previously I  too thought performance was slow and this was confirmed in Sharihars review but it has improved with this latest firmware.

So how to do this?
A recent thread by member joeqsmith shows IMHO a very good method:
https://www.eevblog.com/forum/testgear/using-an-arb-to-test-meter's-continuity-feature/



Results:
Continuity triggering stable to 15 Hz. (0-4V pk squarewave 50% duty)
Minimum pulse width reliably triggered is 15 ms. (@ 1Hz rate)

Equipment used:
Siglent SDG1010, output set to 50 Ohms impedence.

[Muttley Snickers]

You might also be able to use a AWG to gauge what the display update rate is more accurately by configuring either a ramp with increasing amplitude or even a basic high low square wave, the values don't really matter it's the meters ability to display the transition that counts.

Obviously some meters can output a reading much faster to PC than they can display it, but again as a reliable and accurate tool to calculate the update rate an AWG and perhaps in conjunction a scope to document these figures is a better solution to some video content producers simply declaring that "I think it's about one or two a second but it's a bit hard to tell'.

Also as most cameras will only do 25 or 30 and some better units 50 or 60 frames per second so recording of the meters display update rate during this type of reproducible test is something that I have not seen yet by people doing reviews on equipment and for sure as some equipment is so ridiculously fast that even some high speed cameras may struggle to get a grip.

Just some thoughts.


Muttley