Xilinx S3 Starterkit, how hot it is supposed to be?

Hi

I just burnt my fingers trying to lift off the Xilinx Spartan 3A Starterkit the all corner with power supplies is extremly hot - I wonder if that is normal or not.

I already own 3 FPGA boards designed by Digilent with burned in Power supplies so I am little worried that I may get a 4th one into my collection of "dead digilent garbage" :(

The board was powered maybe 3 minutes and FPGA was configured with original bitstream for the dataflash programming, and yes incoming power is 5V not 9

Any ideas? Should I be worry? The power supply IC are so tiny that it is defenetly not possible to mount any heat sink on them :(

Antti

Reply to
Antti
Loading thread data ...

What sort of regulator did they use? Current drawn?

Reply to
Tim (one of many)

Antti,

Which regulator gets hot? IC12, IC3 or IC5? All these ICs have thermal protection, so you shouldn't have to worry about this causing it to fail. They'll turn off if they get too hot.

However, that's no help to your burnt fingers, is it? I hope you learnt a lesson here. I notice that a lot of your work seems to be on the software side of the FPGA world. Complex and challenging it may be, but perhaps you should leave the real engineering (such as handling the boards) to us hardware guys whose fingers are so much more used to abuse than yours? ;-)

Cheers, Syms.

p.s. Maybe this would help?

formatting link
I went into an Aveda store somewhere in silicon valley once and demanded this product. I half expected to be arrested, but apparently the UK meaning hasn't travelled the Atlantic yet. My mates back in the UK were delighted to have gifts of 'Hand Relief' for xmas!

Reply to
Symon

FWIW, my Spartan-3E 1600 kit from Digilent had a burning hot power regulator (just short of smoking). Digilent swapped the board for one that didn't do that, but was otherwise identical. I suspect there might be something wrong with yours.

Tommy

Reply to
Tommy Thorn

Maybe is not so normal...

I've a spartan-3 starter kit and a spartan-3e 1600 board both from digilent and quite all IC on them cannot burn Your finger...

Just on the spartan-3e 1600 board the ethernet phy IC is a little bit hotter than other IC but absolutly touchable with fingers

Sandro

Reply to
Sandro

Dear Syms

where do you get those weird ideas that I am on the soft side? It doesnt sound like that you wanted to insult me, but I still wonder... as I am doing almost onle the "hard" stuff (that is hardware) since

1979.

if you dont belive my mileage, try find some guy who knows what is ERA6000 - I have one of them in junk box.

...

actually my fingers are ok, and nose also, I would normally not touch a board that would burn me, as that would already smell or smoke.

It must have been something weird as the same board even with same bitstream is now behaving much better, but at one time it looked like really too hot. And yes, I can say when some IC is really overcooking, it wasnt the case yet, but it was also more then you normally would touch without feeling pain.

Antti (withasmallsmile..)

Reply to
Antti

Antti,

"Hot" behavior is usually associated with something that pulled too much current out of, or into, the substrate/vcc's.

It is not SCR Latch-up, but instead the memory cells themselves are de-programmed (loose their brains) because the substrate perhaps went too far negative or positive, and as a consquence, the device looks like it needs far more current than usual.

In effect, you have forced a random bitstream into the device.

This requires more than 200 mA sink or source.

Of course, sinking or sourcing 200 mA to supplies above any chip vcc, or below ground (0 volts), is not something we recommend (or allow). But, it can happen (power sequence with another chip, connecting a cable to another pcb, ...).

I suggest that if you ever see this again, you attempt to do a readback/verify (if it isn't too hot!).

Austin

Reply to
austin

Of course I'm joking. Although I'm a little worried that you seem to know an awful lot about EDK! :-)

OK, glad all is well. Like I said, the regulator chips have thermal protection, so they should survive getting hot! Cheers, Syms. (withabigsmile!)

Reply to
Symon

Bye hell you are going way back. That was a GPS equivalent of an FPGA. My boss worked on that. That was the good old days.

Andy

Reply to
Andy Botterill

this is what I wanted to hear. yes looks like many different digilent board have power supply issues.

Antti

PS to Symon - I am doing "hard" since 1979, and doing not muchelse than FPGA hardware lately, so I did reall get insulted by your suggestion.

Reply to
Antti

Antti, Wow, you're easily insulted. I apologise. The joke was meant to be that everyone on this board knows you're an excellent engineer, but even the best engineers burn their fingers now and again. However, I'll not try and joke with you again. Syms.

Reply to
Symon

Austin,

did you read my original posting?

the FPGA was programmed with biststream downloaded from Xilinx website the "Picoblaze dataflash programmer" - and the design was working properly.

If the FPGA did take too much current (while being properly confoigured with known good bitstream), then it maybe is reason to worry for someone not me.

but maybe it was random, misbehav, its rather hot in Munich daytime may reach 32C in shade

Antti

Reply to
Antti

Antti,

I was just letting you know the most likely cause of a "hot" chip.

Sorry it is so hot in 'Munchen'.

Aust> >> Antti,

Reply to
austin

No Problems - ;)

eh my joke-tolerance is little low lately, and it hasnt been very high ever I suppose

Antti

Reply to
Antti

I'm glad your story had a happy ending.

--
David M. Palmer  dmpalmer@email.com (formerly @clark.net, @ematic.com)
Reply to
David M. Palmer

Thank you, Mr 'Palmer'! :-)

Reply to
Symon

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.