Siglent SDM 3055 problem, sparky noise in power off mode ??

[oz2cpu]

anyone else seen / heard this ??

Siglent SDM 3055 is 5month old, it power up fine, and works fine,
but when powered down, i hear some odd sparky sounds from it, on semdom and random occasions,
it sounds like a fly fell into it, or sparks, or a relay at too fast switching frequency with too little voltage..
when it happens, the duration is only 3-5 sec, then it stops for many more hrs.

there are like 3-6-12 hrs between this, so it gave up quite a lot of times trying to video this phenomenon.
the unit is sealed , and on warranty, so i did not open to see what is the cause,
I even got a super good thermal camera, so i am sure i would be able to track the cause and find a good solution,
maybe other users could be happy to hear about it.

233 V mains, 50Hz, and the unit uses 3W in power off mode.
ambient is 25C and 52 % RH

[2N3055]

Hello Thomas,

don't get offended, but did you try a different IEC power cord? I had encountered exactly that problem, inside power plug on a cord, metal would loose elasticity and would have loose contact inside.
Symptoms were exactly like that.

Regards,
Sinisa

[oz2cpu]

thanks a lot , it is a super good suggestion, I am open to try out anything.
but the sound came out of the ventilator holes on the side,
so i did not check the IEC cable, for that reason,
i got a lot of them, so it is a great idea to try a new cable.
right now the unit has been standing all by it self, with the original cable, i only heard it one time today,
but the repeat time is really that low normally, so i can't call it a result yet.
I will update if i find the cause,
or send it back incase it continue with a new cable.

PS: i also got a friendly suggestion : try to flip the voltage selector a few times,
I do that too, while cable is out of course,
lets try to rule out all the easy things first.

[oz2cpu]

NOW it is going on again !!
and this time i finally managed to record it

this happens only a few sec pr day, (3 short times during this recording)
so to record it, it is insanely difficult to do,
sorry about my girl who came to say hello, did not know it was the only sec a day, I needed total quiet for a recording ..
so there is something wrong with the standby supply.

https://youtu.be/hCBbnuBiqwc

[2N3055]

That actually sounds a lot like a fast chatter of relay.... Really unusual sound.

[tautech]

That actually sounds a lot like a fast chatter of relay.... Really unusual sound.

Whereas it sounds like a loose or dry joint to me. 
Siglent outsource their SMPS so maybe one missed QC.

So yes it needs checking out by a service center or official authorization to open it for further investigation.

[oz2cpu]

there are 5 relays inside arcording to the eevblog teardown video,
3 of them are bistable type, so i dont think they can do this.
it sure sound like fast random drive signals to a relay.

i did ask my supplier for permission to open, and track down the problem for them,
(NON void warranty of course, hopefully they say yes.)
since it is very rare to see this, i feel they will never be able to find the problem or the cure,
and also we need to know ?? is it just a simple defect in my unit ? or a general design issue ?
will other similar units do this too ? same age, serial range ?
please check, if you own one, let it stay with mains, but power off on the front, put it where you sit and sleep
so you detect the weak audio sound

it seems like it happens a little bit more often, the first hr after mains is applied,
or like in the video, i just put it down on the ground, i could feel cold air from garden door,
so it might be affected by temperature or humidity

my plan : solder wires to the two standard relays coils,
connect to scope, set trigger, and memory mode, let it run a few days, and then i will have all the coil pulses stored for research,
then i also want to monitor supply voltage of the standby system,
and maybe also mains voltage, just to isolate all input variables.

[tautech]

my plan : solder wires to the two standard relays coils,
connect to scope, set trigger, and memory mode, let it run a few days, and then i will have all the coil pulses stored for research,
then i also want to monitor supply voltage of the standby system,
and maybe also mains voltage, just to isolate all input variables.

Please don't do this as if it's still under warranty the service center may void it.
Open it if approved to and check for an obvious and easy/simple fix but otherwise send it in for repair.

[oz2cpu]

of course i will only open and investigate, if approved by siglent and my unit is still under waranty.
it is a rather expensive unit, and i am very happy about it,
i just feel i need to know, if it is a common design error, or not.

[tautech]

i just feel i need to know, if it is a common design error, or not.

First I've heard of such a thing with this model in this thread.

Sold a few here in NZ but not large #'s and never see them again......tautech touches wood.

[2N3055]

Thomas,

I just said it sounds like relay chatering. It doesn't have to be relay, it can be a magnetostriction of a PSU transformer or common mode choke while sparking from a cold joint somewhere in a standby circuit, like Tautech said...

Point is, I understand the curiosity to find out  what it is... 
That is what why we do what we do... 

But if it is in warranty, I would return it just to be sure.

Unless they approve the "surgery" in which case we expect details... 

All the best..

[oz2cpu]

> in which case we expect details... 

exactly, i also feel i am the only one competent to find this issue,
since i am the only one who have seen it, and is patient enough to wait for it to happen.

i keep you updated,
so far i am happy you can now see it is really there.. not something i make up :-)
 

[oz2cpu]

latest news : my brand new 3055 unit is send back to siglent for repair,
this means they want to find the problem them self, fix it, and send it back to me,
meanwhile my reseller was so kind to send me a borrow unit so i can continue my daily work,
i dont expect i will get any detailed technical response of the problem or solution.

[klausES]

It reminds me very much about the noise of my 3055 with the penultimate firmware (as far as I remember 1.01.01.22).

This always happened when as Last e.g. the temperature measurement activated and the 3055 has turned on again in this state again.
In this situation, this firmware has actually rattled a relay (acoustically roughly with mains frequency)
towards the end of the boot process for one to two seconds.

PS.
From the 1.01.01.22R1 (and the new ... 25) this error was fixed.