Maximum allowed output voltage on Pic 18X

R

Rbeckett

Member
I have re-read the data sheet on the CHI035 board numerous times and have been unable to find a max allowable voltage for the output side of the FET board. The data sheet suggests 12 volts in the example at 1 amp, but I would like to use 24 volts at 1 amp for a project I am working on. Otherwise would it be better to link the standard power project board to a stepper driver board and call it good. I really like the abiliy to cut the resistor off and seperate the input voltage from the output voltage all on one board since additional boards and additional wiring connections tend to create issues with durability and noise in the system. TIA for your thoughts on this issue.
Bob
 
R

Rbeckett

Member
Haku said:
In the CHI035 PDF it says the high power board uses IRF502 FETs, which are capable of handling up to 100V and up to 9.6A.

Specs wise it says you're ok, but heatsinks may be needed.

I did once read of someone testing copper pennies as heatsinks...
Click to expand...
Thanks HAKU,
I really appreciate the reply. Where did you find that info on the sheet? I have looked for it several times and must be overlooking it somewhere. I plan to use a large heat sink in addition to a fan to cool the unit to prevent overheating and to protect it from Iron Oxide dust that it will be in close proximity to. This board will also be in close proximity to High Frequencies used to initiate an electric arc across plasma, so I am trying to reduce wiring connections to a minimum and enclose the system in a Faraday enclosure and sheilding all cables to prevent noise from creating errors in the input and output of the system. Any thoughts on those issues? Thanks for the help!!
Bob
 
H

Haku

Senior Member
In the parts list in the PDF there's Q1-4 which is the FET.

I've learnt to look for parts, identifiers on components and schematic diagrams from being given a blank board (or few) + parts list + schematic and being told to "build it" too many times over the years.

You'll have to wait for some replies from the seasoned circuit designers on shielding issues, I'm more a tinkerer/builder :)
 
D

Dippy

Moderator
Best thing for Bob to do is post his exact design and board connections.

No point everyone pondering for hours over noise suppression ideas if Bob's idea is top notch already...

For something extreme I would always do my own PCB so that I could optimise track specs and layout.
 
B

BeanieBots

Moderator
Agree with Dippy.
Also, don't forget that the FETs used may well be rated at 9.6A but they are NOT logic level type and they ARE driven with logic levels. The datasheet will also show that under these conditions they can only sink ~3A on a good day and will get damn hot doing so.

24V @1A will be fine.
 
D

Dippy

Moderator
That's a good point.
I never understood why the non logic-level ones were chosen?
Seems a bit tight.

I haven't got one of these PCBs; what are the track widths like?
 
B

BeanieBots

Moderator
I'd not want to put 9A+ down those tracks myself. (don't know the weight)
Might be a "safety feature" for use in schools. By having the FET limit the current, it means no instant flames when switching 12v from a car battery down to 0v with no load:( (relatively safer)
Also, easier to replace a FET than the board.
 
Last edited:
R

Rbeckett

Member
Hence my initial question on the max voltage. I think that 24 V would be Ok at 1 amp, but 100V at 9.2A is extreme. The suggestion on the example on the CHI035 Data sheet was for 12 volts and I wanted to be sure it could handle the increased voltage before I let the magic smoke escape on the first try. The board will be in close proximity to a plasma torch operating in the 7000 degree Farenheit range and will see high freq, as well as Iron Oxide dust in signifcant quantities. I am using a breadboard to get the coding and input/output schemes worked out, then I will mill a torch holder and install the components, so I dont have a final idea yet. It is really just a beginning work in progress at the moment. When I get it a little further along I will put up a ppost with all of the parameters and code so that anybody interested in duplicating it can. Thanks for all of the help so far, I really do appreciate it especially since I didnt even know what a resistor was 6 months ago.
Bob
 
D

Dippy

Moderator
Sounds like a fair point. Serendipity is wonderful.

Though a sentence or two concerning max amps and volts in the Data Sheet would clarify the point and could save some accidents. "By implication" isn't really good enough to warn little Johnny about sticking an unhealthy volt/current load onto it.:(
 
J

John West

Senior Member
I'd guess a comparison of prices of power FET's with otherwise similar specs, but TTL driven vs higher voltage and current driven, would yield why the FET's selected were used instead of the more technically appropriate TTL drive level FETs.
Good TTL drive power MOS FET's don't come cheap.
 
G

Goeytex

Senior Member
Looking at the schematic I see that the FETs are being directly driven by the
microcontroller.

The 5V ( 4.5V) output of the micro coupled with it low current sinking ability
is inadequate for properly driving the IRF520 MOsfets at anything other than
a snails pace. One must wonder why this FET is was selected for this
otherwise nice project board.

This FET needs an 8 to 10 volt switch signal with more available current to
rapidly move out of its linear area.

There will be unnecessary heat dissipation issues, that will get worse as
switching speeds increase.

If it were my project I would remove the IRF520's and select a good logic
level FET. Or better yet build your own board using mosfet drivers and
modern, efficient power mosfets.
 
D

Dippy

Moderator
It's an old board so maybe the target audience didn't need anything more complicated or costly?
After all, we can design extra bits easily and all of a sudden the board is 1 foot square and costs a bomb.
The Data Sheet could do with freshening up.
Maybe the original PCB designer hadn't even heard of MOSFET drivers... after all, they've only been highlighted on THIS Forum in the last couple of years.


The choice of non-logic level MOSFETs is strange - maybe that part was designed by an accountant :)
I bet the cost difference between (for example) the IRF520 and IRL520 is in the couple-of-pennies range.

From what Bob has said about his application I would 100% design my own board, but at least this board will allow things to be tried out.

As to whether Bob actually needs a MOSFET driver is unknown (to me anyway??). If it's slow switching then a Logic Level MOSFET is fine. If it's for PWMing something then driver is needed.

I'm sure if Bob wants design help then he'll post his proposed schematic and full requirements list clearly .... ;)
 
You must log in or register to reply here.
Top