Sanguinololu ~2amp mosfet limitation?

No matter what amperage power supply I use (I'm using ATX, PC power supplies), I can't seem to get the sang board to draw more than 2.5 amps when heating both the bed and the tip.

The heated bed takes _forever_ (like 20-30 minutes) to get to 120C (and hold it) for ABS. I have no problem adding a relay and pulling directly from the 12v on the power supply for the heated bed - but I'm kinda stumped as I thought the Sang-lo board should be able to do it all...

Edited 1 time(s). Last edit at 09/19/2011 12:35AM by TheCase.

The Mosfets are rated to 55A and I have no issue getting my hotend up to temp. I would guess its a software issue.

Sanguinololu 1.1 w/ Teacup P.I.D. control.

---

FFF Settings Calculator	Gcode post processors	Geometric Object Deposition Tool Blog
Tantillus.org	Mini Printable Lathe	How NOT to install a Pololu driver

I'm talking about a heated bed. Not the hot end. The hot end heats up fine and quick (but I'd imagine that's because its about 300 times less as large of a surface area)

I'll look into software (using Sprinter), but I have my doubts. I've got a rather large heatsink on the Bed side mosfet and that thing gets crazy burn-your-finger-if-you-touch-it hot.

I should add that I've been using a 12amp power supply.

Edited 1 time(s). Last edit at 09/19/2011 01:12AM by TheCase.

I doubt 2.5 amps for both heaters will get you above 40°C on the heated bed. Typically you need some 10 amps for the bed alone.

With 10 amps, 100°C on the bed is about the maximum you can get. To heat up faster, put some isolating stuff onto it until it's ready to print.

BTW., not my own experience, but a collection of what others said.

---

Generation 7 Electronics	Teacup Firmware	RepRap DIY

Can you describe your heat bed? Do you have all the resistors in series or parallel? How are you connecting them to your build plate? Do you have thermal compound where you mount, and insulation on the bottom side? How thick of wire are you using from PS to bed? What firmware are you using? Is this a 1.3A board or older? What kind of power supply are you using? What are the specs on it?

The heat bed is the prusa PCB (red), so no resistors. 1.3a board.

The power supply is a conventional PC power supply - 12amps on the +12v side.

I put a Kill-a-Watt between the power supply and the wall socket - that's how I can see its only drawing about 2.7amps.

I'm using the same gauge wire as the 12v leads that power a PC hard drive. I suppose I could go thicker on those - now that I think about it, that could be the weak point in the system. The do get a bit warm when trying to juice the HB to 120C.

TheCase Wrote:
-------------------------------------------------------
> The heat bed is the prusa PCB (red), so no
> resistors. 1.3a board.
That is 1.3 ohms?

> The power supply is a conventional PC power supply
> - 12amps on the +12v side.
>
> I put a Kill-a-Watt between the power supply and
> the wall socket - that's how I can see its only
> drawing about 2.7amps.

If that measured the mains amps, then 2.7a at 220v makes out 594W, then assume perfect conversion to 12v would turn into something like 49.5Amps ??? Something must be wrong. So can only be 110v mains that would give a 24.75 amps, lets say a 80% efficiency then its maybe ~19.8 amps, and its still way too high than it should be. To get to 12amps it should have a 50% efficiency but thats more of a linear supply not a smps. Not really sure but i think the bed you mentioned should be at about 10amps when cold, probably less when heated. And rest of electronics should be like max 2A including 1.76A for extruder 6R8 resistor. Because with that 80% efficiency scenario you have at least 7 amps @ 12v that are not clear, fact which might suggest at least one short (or super-driven motors on a 150% output psu?). Those estimated 7 amps are also above the limit of specs, as you said the 12v line can output max 12 amps. So this math points out that the psu you are trying to use is probably under alot of stress at the edge of its limits (or beyond). Since it draws that much and the electronics havent set on fire yet, lets hope its the psu that is broken and not the more expensive other stuff.

All considering that initial 2.7amps reading on some 110v mains is correct. Because to be honest i wouldnt put much faith in that thing either, maybe all is ok and only that kill-a-watt thingy is just way off about its own readings. Or maybe its just a metaphorical name, since it basically ... makes amps dissapear ... makes psu give more amps than ever ... or maybe just showed you more power so you would be tempted to lower the consumption ... either way isnt that what a box named "kill-a-what" should do?

> I'm using the same gauge wire as the 12v leads
> that power a PC hard drive. I suppose I could go
> thicker on those - now that I think about it, that
> could be the weak point in the system. The do get
> a bit warm when trying to juice the HB to 120C.

If those get warm, its bad, and they should be thicker so untill it doesnt get warm.

Edited 2 time(s). Last edit at 09/20/2011 12:43AM by NoobMan.

On a serious note, if i would had that myself i would certainly investigate it further.

Trying to do the math again in watts, the extruder should draw 12*1.76=watts, the uP and rest of electronics some insignificant amount of power, maybe 12*0.1A, e.g. 1-2 watts, and the motors depends on what you have and how you set them, the math cant be sure but maybe somewhere between 30w (pololus maybe) and 60 watts (bigger drivers maybe). Untill now within 50-80W. And heated bed like 12*10a=120W max, so grand total somewhere 170-200W including bed.

While 2.7a for 110v gives 297va and to make it 200W on the 12v line the overall eficiency would have to be ~67% which might be considered realistic but at a stretch. For 170W its worse than it should be for a psu that is not even variable, should have a better efficiency than 57% since it is switching mode (smps). These values however would be greatly affected if that power meter is even like 0.2-0.5 amps higher than reality.

But still to output 200/170W on the 12v line the psu has to give out ~17/14amps, but you said it is rated at max 12 amps. This might be an important issue if its like that. Psu would need to run at ~150/120% of it ratings. Try a psu with a 12V rating at least like 20A or more, so it wont run at the edge of its specs, and certainly not beyond. Also as a question, i am a little intrigued to hear a atx psu with 12a on 12v line, i think most i saw are at least 20 amps or more on 12v line. Are you sure of the 12A figure?

I would hope either that the power meter shows little higher readings, or that efficiency of psu drops dead just because it is forced to output more than it was designed for, or even hope that the atx psu is more or less crippled, as it should be changed anyway. The perspective where might be something wrong with the rest of electronics is much worse than any of these.

Cheers and good luck

P.S. I went way too far with many assumptions above, just starting from the one single 2.7A figure you mentioned, so what i said might be pretty well stretched on the overall picture. But i think on the 12V line you need more than 12A for certain.

Edited 2 time(s). Last edit at 09/20/2011 10:00AM by NoobMan.

How have you insulated your heatbed against thermal losses?
Try putting a layer of 1 cm (surgical) cotton wool beneath it.