SDM3045X to SDM3055X-E "Improvement" Thread

[KungFuJosh]

I created this thread so that there was something specifically dedicated to "improving" the SDM3045X DMM.

For those of you that don't already know, the SDM3045X 4.5 digit hardware is identical to the Chinese version SDM3055X-E DMM. With a fairly simple firmware modification, you can improve your DMM to 5.5 digits.

I've been enjoying my 3055X-E for 4 months without any issues.

To make the improvement, check out this post:

patch the original SDM3055X-E_2.01.01.12.ADS with the following 16 bytes.

Then it's ready to flash in a SDM3045X.  As always, do it at your own risk.

https://www.eevblog.com/forum/testgear/siglent-sdm3045x-enough-is-enough/msg3909188/#msg3909188 - more details are in that thread if needed, I think they also covered "unimproving" the device.

After modifying that specific version of the firmware, you can update to any unmodified SDM3055X-E firmware.

Language default will probably be in Chinese. To change it:
Click the Utility button, then System Config should be in English (2nd from right soft button if it isn't),
then Language should also be in English (1st soft button on left if it isn't).
The first language option should be English (but written in Chinese).

[jas]

I tried to update my SDM3045X with a patched SDM3055X-E_2.01.01.12.ADS file, but it failed with "Update Fail" message.

I thought maybe the file is somehow corrupted and tried to update with a SDM3045X_5.01.01.07R1.ADS file which would downgrade the current sw, but that fails too.
The current sw version of my SDM3045X is 5.02.01.09R2, which is a bit strange, because the latest version available on Siglent's website for SDM3045X is 5.01.01.09R2.

I tried with three different USB sticks (all FAT32 formatted) and all failed. Am I missing something, or why the firmware update fails?

[Veteran68]

Not suggesting it's your problem, but Siglent gear is known to be quite finicky about USB flash drives. I've sometimes had to try several different ones to get one to be recognized for firmware updates, and have read countless reports of same from other users. Once I found one that worked, I labeled it for Siglent use. I keep all my flash drives in a zippered case that's bulging at the seams with probably 2 dozen varieties, so I could never remember which ones worked well and which didn't until I labeled one.

[KungFuJosh]

I tried to update my SDM3045X with a patched SDM3055X-E_2.01.01.12.ADS file, but it failed with "Update Fail" message.

I thought maybe the file is somehow corrupted and tried to update with a SDM3045X_5.01.01.07R1.ADS file which would downgrade the current sw, but that fails too.
The current sw version of my SDM3045X is 5.02.01.09R2, which is a bit strange, because the latest version available on Siglent's website for SDM3045X is 5.01.01.09R2.

I tried with three different USB sticks (all FAT32 formatted) and all failed. Am I missing something, or why the firmware update fails?

I would suggest trying to flash an unmodified version of the firmware V5.01.01.09R2 and see if that works or not. If not, then it's possible that either we need hacking instructions for newer firmwares, or your hardware might not be compatible with this mod.

[jas]

I tried with V5.01.01.09R2 and that fails too.
So, either there is something wrong with my USB sticks, or Siglent has prevented this kind of firmware updates with the new V5.02.01.09R2 version.

The USB sticks could be the problem, but I have been using them to update my other Siglent equipment without any trouble, so I don't think there is anything wrong with them, but of course I cannot be 100% sure.

[KungFuJosh]

I tried with V5.01.01.09R2 and that fails too.
So, either there is something wrong with my USB sticks, or Siglent has prevented this kind of firmware updates with the new V5.02.01.09R2 version.

The USB sticks could be the problem, but I have been using them to update my other Siglent equipment without any trouble, so I don't think there is anything wrong with them, but of course I cannot be 100% sure.

I doubt it's the USB stick since you've used them before for the same purpose.

I think it's more likely that you have a hardware variation or a firmware "feature" that prevents using other firmwares. My guess is that you would need to wait for a 5.02.xxxx variant firmware to become available before you can mess around. I don't know for sure though. Any thoughts on this @tautech?

[tautech]

Any thoughts on this @tautech?

Only to try the obvious FW rollback to an earlier FW version and then apply the X-E trick.

[tv84]

The current sw version of my SDM3045X is 5.02.01.09R2, which is a bit strange, because the latest version available on Siglent's website for SDM3045X is 5.01.01.09R2.

Wow. That's a major difference. You should not be able to flash anything available before Siglent releases new FW package.

And, most probably, you shouldn't be able to downgrade...

Nonetheless, I'll try to craft a .2. version for the migration FW to see if it works.

[MarkKn]

Was this discussed before and it was determined that the digits increase, but beyond the accuracy?

[coromonadalix]

there is a thread about the 3 models  

[KungFuJosh]

Was this discussed before and it was determined that the digits increase, but beyond the accuracy?

There's no accuracy issue. The 5.5 digit model for the China market is the same exact hardware. I did a comparison test with an SDM3055, and my X-E version was more stable in the smaller digits. I assume because the X-E model has had a lot more burn in time.

Here's the results of that test:


Note that it's a reference from AliExpress, and while it's a better quality one, I wouldn't bet any money on the cal data. Though it did appear to actually be handwritten.

Within 0.5 mV and more stable is pretty awesome performance.

[Kleinstein]

From the pictures reports around the SDM3048 / 3055E uses a few cheaper parts than the full 3055. AFAIK the main ones are:
1) the input divider with separate resistors instead of a large ceramic divider
2) A REF5025 instead of max6225
3) simpler resistors in the ohm part
4) smaller relays (maybe less tolerant to switching with high voltage present)
5) smaller size shunts - more thermal INL expected
6) no fan : quite, but longer warmup time
So there is a chance to see a little more long term drift and also reference noise. Still even the cheap version it not that bad and well capable of 5.5. digits.

[tautech]

From the pictures reports around the SDM3045 / 3055E uses a few cheaper parts than the full 3055. AFAIK the main ones are:
1) the input divider with separate resistors instead of a large ceramic divider
2) A REF5025 instead of max6225
3) simpler resistors in the ohm part
4) smaller relays (maybe less tolerant to switching with high voltage present)
5) smaller size shunts - more thermal INL expected
6) no fan : quite, but longer warmup time
So there is a chance to see a little more long term drift and also reference noise. Still even the cheap version it not that bad and well capable of 5.5. digits.

Yes we know all this.

Topic is SDM3045X-E HW = SDM3055X-E China version.

[glitchcatcher]

Has anyone tried the Siglent SCD30A shunt after a successful conversion?
Is the "Option Measure" available in current measure mode?

The shunt manual states: "Suggest only working with SIGLENT SDM series 5 ½ or 6 ½ multimeters."
The SDM3055X-E is a 5 1/2 one.

My SDM3045X definitely doesn't have that option.

[KungFuJosh]

The SDM3055X-E is a 5 1/2 one.

My SDM3045X definitely doesn't have that option.

I don't see that button using the SDM3055X-E firmware. Actually, it's not on the standard SDM3055 either.

[Tjuurko]

Has anyone tried the Siglent SCD30A shunt after a successful conversion?

My SDM3065X last FW V3.01.01.12R1 (2022.11.29) definitely doesn't have that option.

[glitchcatcher]

Actually, it's not on the standard SDM3055 either.

My SDM3065X last FW V3.01.01.12R1 (2022.11.29) definitely doesn't have that option.

Ah, what could be is that the shunt first has to be "detected". The manual states: "Inserting SCD30A during the SDM in shutdown state; "
If possible, try to connect a shunt (0.01 Ohm?) to the HI/LO terminals, then turn the device on and switch to DCI.

[KungFuJosh]

Ah, what could be is that the shunt first has to be "detected". The manual states: "Inserting SCD30A during the SDM in shutdown state; "
If possible, try to connect a shunt (0.01 Ohm?) to the HI/LO terminals, then turn the device on and switch to DCI.

I had the same thought. Might be worth a try.

[tautech]

Will try also on return to base. Have shunts in the SHS820X package.

[KungFuJosh]

Will try also on return to base. Have shunts in the SHS820X package.

I tried with a shunt I made that said 0.4Ω in continuity mode. Didn't do anything on either of my DMMs. I inserted it in a powered off state, and then turned it on, and there was no change.

Except I noticed something weird on the real 3055. For some reason, every time I power on, it switches to continuity mode. Power on state is Last, and Last was DCI. 🤷

[tautech]

Will try also on return to base. Have shunts in the SHS820X package.

I tried with a shunt I made that said 0.4Ω in continuity mode. Didn't do anything on either of my DMMs. I inserted it in a powered off state, and then turned it on, and there was no change.

Except I noticed something weird on the real 3055. For some reason, every time I power on, it switches to continuity mode. Power on state is Last, and Last was DCI. 🤷

Hmmm maybe try setting it back to Default (DCV) for a boot or 2 then back to Last. Might somehow have its knickers in a knot. 

[KungFuJosh]

Hmmm maybe try setting it back to Default (DCV) for a boot or 2 then back to Last. Might somehow have its knickers in a knot. 

Is there a time requirement for turning on or off? Like how long it needs to be on before powering off, or how long it needs to be off before powering on to not mess with the settings recall?

[tautech]

Hmmm maybe try setting it back to Default (DCV) for a boot or 2 then back to Last. Might somehow have its knickers in a knot. 

Is there a time requirement for turning on or off? Like how long it needs to be on before powering off, or how long it needs to be off before powering on to not mess with the settings recall?

AFAIK setting changes are not saved instantly like in scopes. Probably 30s is enough to get them captured.
Some while back this was discussed in the SDM3000 thread IIRC.
I’ll check this behaviour also when home and if necessary report it as a new bug. 

[KungFuJosh]

AFAIK setting changes are not saved instantly like in scopes. Probably 30s is enough to get them captured.
Some while back this was discussed in the SDM3000 thread IIRC.
I’ll check this behaviour also when home and if necessary report it as a new bug. 

It was a timing thing, apparently I was turning it on and off too quickly. I didn't time it, but I left it on and off longer, and recall worked correctly.

[glitchcatcher]

AFAIK setting changes are not saved instantly like in scopes. Probably 30s is enough to get them captured.
Some while back this was discussed in the SDM3000 thread IIRC.
I’ll check this behaviour also when home and if necessary report it as a new bug. 

Oh yes, please make them release new FWs as soon as possible, to save us doomed people with the 5.02.01.09R2.