Burnt out board No.2

[stevekerikeri 1]

Hi.

I have replaced the main board in one of my machines once already and it seems to have blown again. In both boards there is a distinct small hole in the chip U6. Has anyone had any issues like this. The two machines have had relatively little use in relation to what I would expect to cause a board to burn out in a machine that costs what the Ultimaker 2 did.

I have been looking for schematics to see if I can find out exactly what this chip does. If I can work this out it may point out where the fault actually is and provide some indication of how to fix it. I have been in touch with the supplier to see what they think as well.

Thanks.

[neotko 1,417]

For the 2.1.4 board

https://github.com/Ultimaker/Ultimaker2/tree/master/1546%20ultimainboard%20V2.1.4

Edited June 25, 2017 by Guest

[Dim3nsioneer 558]

Are both boards genuine UM boards? If your machine is still under warranty, please contact your reseller.

[stevekerikeri 1]

Are both boards genuine UM boards? If your machine is still under warranty, please contact your reseller.

The first board burnt out not long after the warranty period ran out. The machine was being used in a school where it was active for part of the year then unused for quite a few months.

We have two machines running in the same room. The same one has blown again on exactly the same chip. But the other is fine. In the same room over the last 8 years or so we have been running a cnc router, a cnc engraving machine, a Replicator 2 (last three years or so) along with two computers. This is the only machine that has had a problem. I have been offered a replacement board by my reseller. Which in its self, is not bad as the machine was technically out of the warranty period. If only by a few months. However, I was hoping to build up to around 6 3D printers running in this room over the next two years. The two Ultimaker 2s, a dual head machine and some smaller bed ones such as the new little Ultimaker. The problem is not the $400 or so for the burnt board. It is the down time and the uncertainty of investing quite a lot of capital in further machines if similar problems are going to turn up. I am trying to get a course of around 20 students using this room next year. This is to be based around 3D printing and electronics. The idea of having machines dropping out of use once a year with 6 machines is a cause for concern. In both boards it has been the U6 chip that has gone. I have a co-worker at the school who used to build missile systems for the US in a previous life who is currently checking out the power supply. He seems to think that a power surge would have wiped more than just the U6 chip as other parts of the board do not draw power from this chip. So they too should be damaged. He did get a reading of 54v from the pack but wants to check this with a second multi meter to ensure that it was not just his meter that was faulty. I have some understanding of electronics but nothing like his. Anyway he seems to think that this would destroy the U6 chip (I can understand this) because it is dealing with around 5v (this is all from memory of my last conversation with him yesterday). Anyway it is strange that the chip did not blow instantly as far as I am concerned if it were the power pack producing a voltage of this level.

Anyway I can not see the point in putting a new board into this machine until I have discovered exactly what the problem is as it will more than likely blow again. I have been looking into the cost of UPS and other protected power supply systems to power the room even though I am quite sure it has nothing to do with a brown out or a spike as none of the other equipment (which was not fitted with spike protection where as the 3D printer was) has been damaged. This would cost around $2,000 for a system that would protect possibly all the machines in the room. Which is not that bad as I may be able to get some help with this from the school. I would of course rather spend this on more printers. Anyway at the moment I have one working Ultimaker 2, one working Replicator 2 and a very expensive paper weight. Any suggestions on how to find the fault or how to prevent it from happening again would be useful. I have been using CNC equipment in schools for around 20 years and in the past have always had the luxury of being able to take the machine back to a dealer who was able to examine, identify and repair faulty equipment. Where this was not possible, in the case of large machines such as the big Boxford CNC router we used in the UK an engineer was able to come out and identify problems. I suppose this is the problem with living out in the beautiful winterless north of New Zealand.

[IRobertI 521]

I'm looking at the slightly older 2.1.1 schematics (I don't have Altium to be able to look at the latest ones) but I think the differences are very small with the newer 2.1.4. Anyways, U6 looks to be a voltage regulator with a max input voltage of 46V. So if your PSU is actually putting out 54V then the issue is obvious and would explain things nicely.

I've worked as support for a few years now for the Nordic distributors and while yes, boards do fail, I wouldn't class it as common. So I wouldn't worry that you'll be replacing mainboards every year.

Edited July 6, 2017 by Guest

[stevekerikeri 1]

I'm looking at the slightly older 2.1.1 schematics (I don't have Altium to be able to look at the latest ones) but I think the differences are very small with the newer 2.1.4. Anyways, U6 looks to be a voltage regulator with a max input voltage of 46V. So if your PSU is actually putting out 54V then the issue is obvious and would explain things nicely.

I've worked as support for a few years now for the Nordic distributors and while yes, boards do fail, I wouldn't class it as common. So I wouldn't worry that you'll be replacing mainboards every year.

[stevekerikeri 1]

I'm looking at the slightly older 2.1.1 schematics (I don't have Altium to be able to look at the latest ones) but I think the differences are very small with the newer 2.1.4. Anyways, U6 looks to be a voltage regulator with a max input voltage of 46V. So if your PSU is actually putting out 54V then the issue is obvious and would explain things nicely.

I've worked as support for a few years now for the Nordic distributors and while yes, boards do fail, I wouldn't class it as common. So I wouldn't worry that you'll be replacing mainboards every year.

Yes it looks like that would add up. They were both authentic boards. Do you think that due to the fact that the elevated voltage is not that much higher than the max rating it would have basically burnt out over time. This would explain why it worked fine out of the box and for a period of months after the new board was installed. So basically the power pack could have had a fault from the start or it could have developed a fault after the warranty ran out. I think that the second board did not last as long as the first so I suppose it could have gone later on in the life of the printer rather than straight away. Looks like its time to cut off the plug and throw it in the bin.

[IRobertI 521]

Hard to say. How chips like these will handle going over the max rating is a bit "random" as it depends on so many different factors. Was it hotter/cooler than the test conditions for the chip? How well does the board sink heat away from the chip compared to other configs? How close to the absolute maximum insta-fry voltage did they spec the chip? Does the faulty PSU always stay at 54V or does it fluctuate? etc etc

But verify that it's the PSU first. You have another UM2, measure that PSU as well with the same meter and see that you get a different result (should be 24V).

[geert_2 558]

If you measure on a modern power supply ("switched mode"), then make sure it sees some load: use a lamp or a suitable power resistor (heater) as load.

Most modern power supplies do not work or do not work correctly without load: they may output nothing at all; or a far too high voltage. For a correct measurement, it would be best to use a load similar to the real load it will see when in use in the printer. Connecting a few car lamps might do. And then at least you have some visual feedback too.

[stevekerikeri 1]

If you measure on a modern power supply ("switched mode"), then make sure it sees some load: use a lamp or a suitable power resistor (heater) as load.

Most modern power supplies do not work or do not work correctly without load: they may output nothing at all; or a far too high voltage. For a correct measurement, it would be best to use a load similar to the real load it will see when in use in the printer. Connecting a few car lamps might do. And then at least you have some visual feedback too.  

Will do.

[stevekerikeri 1]

Just to kill this one. I replaced the board after buying a new one. This burnt out again after about 6 months. I was given a new one free of charge but bought a new power supply. So far so good. Been in about 6 months so if it is going to go again it will go soon. I am not expecting it to burn out again.