Topic: Caution! Pandaminers catching fire! (Read 4673 times)

and how can you solve a problem that resets 2-3 times a day?

hard to read but

gpu 0 50c
gpu 1 54c
gpu 2 48c
gpu 3 55c
gpu 4 48c
gpu 5 50c
gpu 6 50c
gpu 7 49c


are good watch them to see if they go past 60c after a day .

these are the data shows.The temperature is normal?

Good day friends!Got my first pendaliner B3+.If I came across earlier on your theme probably would not have ordered this miner.Actually put it on the balcony of the apartment,noise is certainly creepy.There is one problem : periodically popping up a program and there's a label on the screen.The block is also not clear what capacity.Have any questions or opinions?

I'm guessing you haven't heard of KNC, who is now bankrupt, but made alot of BTC miners in 2013

Well, it seems that EMI seems to be the main possibility. After reading many of the posts, their are several complaints that point to EMI. One guy complained that he blew a capacitor, and the pictures show burning on the ground wires, not on the +12V wires and now we have somebody complaining that he's getting a lot of 'noise' from couple of his units. And if you go into the theory far enough, some experts claim that these frequencies are destructive in many ways and in many cases.

How to fix it or help it might be difficult. Some ways that might help would be to make the wires twisted pairs, which is hard to do unless you can make some fancy braiding by turning the plug back through the wire on itself a couple of times. Another method that should help is to wrap the cables with copper or aluminum foil for shielding. Steel foil will not be as good as aluminum or copper, with frequencies this high.

This frequency leak is what likely destroyed the capacitor too. The problem is magnified when you have 2 frequencies that in harmony with each other and the make it a multiplier. Another reason I think this might be the problem is because this box has the ability to do many different hashing speeds. If their is a frequency leak from two circuits and are in some kind of harmony, this will create another frequency. One or more of these frequencies seems to be destructive.

I'm guessing you don't have any bugs hanging around and no dogs. The frequency might drive them nuts. haha.

P.S. Did panda warrant the burned boards?

Do a post-mortem on the cables to see if they are really 18AWG.
http://nerdralph.blogspot.ca/2016/06/when-does-18-26-when-buying-cheap-cables.html

I got my start in mining BTC with S7s, and used server PSUs with genuine 16AWG cables.  For my GPUs I use 18AWG splitter cables connected to the 16AWG cables from the server PSUs.  Even thought the splitter cables are using true 18AWG wire, I have to be careful how much power I push through them because the connectors will heat up more than the cables.

This really has to be bad design on the part of the manufacturer. Never have I seen something so trivial overlooked to miners a company produces. This could hold them liable over huge damages, but they're a Chinese company anyway...

they use 5 sets of cables and split them into 7.

so it is 5 sets  of 18 gauge cables

 an 18 gauge  pcie cable  max is 150 watts.

so 750 watts at the cables would be safe in theory   since these are around 83%  efficient

750/.83 = 903 watts at a k-watt meter/pdu meter  in theory is the  safe max

well my b1 pulled 1015 watts on zec at the pdu meter = not safe
and my b1 pulled  1100 watts on eth at the pdu meter = worse then zec

So  they are selling a psu  that is not safe to run all the gpus inside of it.

This was indeed on a B1.
I don't know how it goes on a B3 model. Mine hasn't arrived yet (still after months of waiting). But I think if you would use the same PSU, it will give the same problem as these PSU's are just crappy!

Well, the mining area where these miners are located is very ventilated. And for the electricity, I'm an industrial electrician, so I've used the right safety requirements. I've seen enough electricals go wrong back when I worked as a technician in a factory, so I've learned my lessons well years ago  This is obviously Pandaminer going cheap on the PSU's. Today my brother discovered 2 more PSU's starting to melt. So this time we were in time before it screwed up my miners... I'm ordering a bunch of new PSU's as we speak. So this problem should be solved in a few days for me though.

Did this happen on the B1?

Do you think it also happens on the B3? Just buy new wires and connectors and swap them out and you should be fine.

Here is why I don't buy the pandaminer.

- 1 month warranty
- Laptop gpu's
- Closed case
- China quality

looks like an over-current (short-circuit) in pandaminer for me - no need to shill

moftkhor, can you post a picture of the PSU label?  Is it made by Great Wall, their current PSU supplier?  Great Wall is a big supplier of power supplies.

A few days ago after I first read this post I went into the DC and felt all of the Panda miner power connections. Most of them felt DC room temperature and only one of them was even mildly warm to the touch. While checking them again today we found one with very hot connections and on closer inspection the black wires were turning gray at the PCIe connector at the Panda miner, and several of the PCIe connector black wires were starting to melt for a few inches with some copper wire showing. We cut the power right away and let it cool just a minute and then disconnected the PCIe connectors before they fully cooled and plastic welded together. I few PCIe pins left the plastic in the connector, but I believe that it can be cleaned out and the miner will still function. I think we caught it in time to prevent major board damage.

The interesting thing is that the 2 most burn connectors were single connector cables, not double connectors.
The connectors with wires that were melting were closest to the back or intake side of the miner.   

I thought I would pass this on for others to keep a eye on there miners, and have them check even if they are hosted at a DC.

If the PSU is the issue I would strongly recommend replacing it with a Platinum or Titanium PSU.
   

Thats dangerous and guess what I don't trust mining even more now. Lol. What if my house catches fire due to those little panda's. Dam that would be bad idea and bad investment ever.

Is your house too hot? Or the mining area looks congested something like that? It may be happening that overheating and bad electrical wrong is causing all these stuff. PSU's can be low quality so you might wanna double check them.

Oh now I know why my garage door opener does not quite work as well as it did.

I will build a faraday cage for it,  thanks for that info.

So far all the problem reports are from the B1 model which uses RX480 cards, also they may have changed PSU suppliers since 6 months ago.  Has anyone have any insight into the recent production to say things have changed for the better?

Anything by OklahomaWolf: www.jonnyGURU.com

I don't have the gear to test EMI, but the scope will give me 20-30mV of background noise at the load tester without the unit being tested even being fired up whenever those big Enhance OEM units are online.

The big ones I use that don't throw out EMI: all EVGA/Super Flower 1600W units, Corsair AX1500i, Coolermaster MIJ 1200W, Seasonic X-1050. Also running an Andyson N700 and Seasonic Prime Titanium 750W in the same room as the load testing gear - those don't do EMI either.