1 (edited by ysb 2013-04-14 21:44:44)

Topic: magic smoke on ramps...

hi all.. it's my turn to ask for help...

this week-en, i decided to changed my sanguinololu for a arduino+ramps1.4...

i triple check the wiring but it seems that i do something wrong . when i tested the hotbed, some magic smoke appear on my arduino.. i switch all in emergency and check the board...

the arduino seem ok, and the ramps board also... i tested all the functions all seems to work apart the hotbed that stay on whatetever i do..

i look more closely and see that one of the 3 mosfets (the one that control the hotbed.. it's logic that this one burn if i do a shot-circuit somewhere) is a bit burned..

i want to change it as it seems to be the only problem on the board: it's a IRF512 N-Channel Power MOSFETs/ 5.5 A/ 60-100V.

as i'm not a specialist, someone can help me to find an equivalent mosfet as the ifr512 seem to be a bit old ?

ty smile

2

Re: magic smoke on ramps...

What heated bed are you using?

3

Re: magic smoke on ramps...

elmoret wrote:

What heated bed are you using?

Rds on that MOSFET is. 0.5 ohms. P=I^2*R, so if its a 50watt bed, that's about 8w dissipated in the MOSFET. That's a tremendous amount for a TO-220 package without additional cooling.

Read about Rds and theta_ja, and choose a MOSFET according to your needs.

4 (edited by ysb 2013-04-14 22:15:00)

Re: magic smoke on ramps...

hem.. i use a qu-bd silicon pad 6x6...  i thought that the ramps can drive this hotbed without problem no?

http://store.qu-bd.com/product.php?id_product=36

they said on the site that it work out of the box with ramps

5

Re: magic smoke on ramps...

ysb wrote:

hem.. i use a qu-bd silicon pad 6x6...  i thought that the ramps can drive this hotbed without problem no?

http://store.qu-bd.com/product.php?id_product=36

they said on the site that it work out of the box with ramps

RAMPS 1.4 calls for a STP55NF06L, which has a much lower RDS. Sounds like you bought a Chinese knockoff.

6 (edited by ysb 2013-04-14 23:17:20)

Re: magic smoke on ramps...

arf....

the STP55NF06L  is a N-CHANNEL 60V - 0.014 OHM - 55A

and mine is a 100v - 5.5a - 0.8 ohm ...

damn...

edit : do you think i can change the 3 mosfets IRF512 by 3 IRF 540 N-Channel Mosfet, 100V, 33A  with RDS of 0.055 (i can source some on a store near me)

they are as not good as the STP55NF06L but i think that they can do the job, any advice ?

7

Re: magic smoke on ramps...

ysb wrote:

but i think that they can do the job, any advice ?

Again, read up on theta_ja and Rds.

The qu-bd bed pulls 11.5A. P=I^2*R. If Rds is 0.055, then that's 7.27w. With a theta_ja of 62.5C/W, that means the junction temperature will be at 62.5*7.27 = 454C above ambient. No, the MOSFET will not last long at that temperature. The maximum junction temperature is 175C.

You either need a heatsink (and calculate it's C/W, add it to the theta_jc value) or a better MOSFET. The STP55NF06L is 0.014ohms, so 1.85W, or 115C over ambient, which is still under the 175C spec - though a heatsink/fan would still be a good idea here.

8

Re: magic smoke on ramps...

thanks for you explication.

9 (edited by ysb 2013-04-17 18:04:28)

Re: magic smoke on ramps...

for one with interrest in.. i replace the faulty mosfet with a mosfet IRLB8743 .. seem to be a good choice (even for our base sanguinololu)

30V (i work with a 12v power supply, so it's good)
RDS : 0.0032

so.. (lets do math) 11.5A^2*0.0032 .... so 0.43W ....Tj=25 ... 11 degC above ambient... no bad smile ... the heatskink is not necessary here.. but i will put a little fan (to cool all the board) for my peace of mind..