Author Topic: solved | lease help: MOSFET seems to allow current through despite being told no  (Read 1452 times)

0 Members and 1 Guest are viewing this topic.

Offline pepelevampTopic starter

  • Regular Contributor
  • *
  • Posts: 101
  • Country: nz
Hi fellas!
I'm trying to throw together an arduino logic controlled circuit that charges a capacitor bank, then dumps the power into a solenoid to hit things I don't like. 
I know it seems straight forward but I'm running into a bugger of a problem: The first mosfet on the left won't turn off! :(

We're battery powered today - with arduino suckling off a lithium ion & 5v regulated boost.
And the capacitors get their daily juice from a lithium ion, 5v regulated then multiplied up to 40v dc/dc boost.

I have two mosfets today:
* one for joining the DC/DC boosts to ground, allowing them to charge the capacitor bank.
* one for joining the solenoid to ground, allowing it to suck on those juicy capacitors & get motivation to hit stuff.

Each mosfet is secretly two in parallel (see the 2x) to ensure maximum current, I treat them as a single mosfet logically though.

There's an interesting caveat to this circuit though - I have basically two separate circuits- the power circuitry and the arduino logic circuitry. This is to avoid tanking the arduino's voltage when the caps charge. No bypass cap can handle it, so it has its own battery.

In order for the mosfets to see 'my gate voltage is 5v above my source voltage' I have connected the two circuits by a 150 ohm resistor at the anodes.

Tried so far:
The mosfet at the left there just does not shut off. It is continually charging the caps which I don't want.
I have checked both mosfets in the pair there with my multimeter and they check out. My transistor tester blew up because I forgot to discharge the 40v from the caps before testing their ESR.
I have also joined the circuits at their cathodes (common ground). I still get the same crap behavior.

Please enjoy the absolute crudity of the diagram. I do know how to draw schematics but I'm on the bread board at this point & I'm just in the sandpit.

Help me please. For some reason I can never get a mosfet circuit to work.


The MOSFET in question is a monster:
IRLZ34PBFm
60 volts, 30 amps, with 5v gate drive voltage.
https://www.digikey.co.nz/product-detail/en/vishay-siliconix/IRLZ34PBF/IRLZ34PBF-ND/811911

datasheet:
https://media.digikey.com/pdf/Data%20Sheets/Vishay%20Siliconix%20PDFs/IRLZ34,SiHLZ34.pdf



Here is a song and music video about my issues with mosfets. It should work, it just never does:

« Last Edit: January 03, 2020, 02:37:46 am by pepelevamp »
 

Offline Mr Evil

  • Regular Contributor
  • *
  • Posts: 76
  • Country: gb
    • mrevil.asvachin.com
Re: please help: MOSFET seems to allow current through despite being told no
« Reply #1 on: January 03, 2020, 12:45:33 am »
The MOSFET on the left doesn't have a gate stopper resistor. The note next to it indicates that you know why they are needed, but I bet you don't know that the ringing can be bad enough to stop a MOSFET from turning off.

Offline pepelevampTopic starter

  • Regular Contributor
  • *
  • Posts: 101
  • Country: nz
Re: please help: MOSFET seems to allow current through despite being told no
« Reply #2 on: January 03, 2020, 01:23:21 am »
oh NO! hmmmmm

hey I think i figured out the culprit. what wasn't mentioned in the above post/schematic is that I'm using an arduino shield for a display. The bloody thing hogs all my io pins and also messes with my ADC pins. fucking garbage.

the only pin I had left available to trigger the charging mosfet was A1 - A5, and they are all floating for some bloody reason thanks to the library used by the shield. I'm currently rejiggering it now with a different display etc. will post & mark as solved if this actually works now
 

Offline pepelevampTopic starter

  • Regular Contributor
  • *
  • Posts: 101
  • Country: nz
Re: please help: MOSFET seems to allow current through despite being told no
« Reply #3 on: January 03, 2020, 02:37:23 am »
yep it was the arduino shield / driver. working fine now. well kinda :P but my issues are for another story.

cheers ears
 

Offline mcovington

  • Regular Contributor
  • *
  • Posts: 191
  • Country: us
Um... The Arduino cannot put a voltage across the gate and source of a MOSFET unless the Arduino is connected both to the gate (logic out) and source (ground) of the MOSFET.

If your circuit works without such a connection, it's by luck, or leakage somewhere.
 

Offline pepelevampTopic starter

  • Regular Contributor
  • *
  • Posts: 101
  • Country: nz
dont need to connect the grounds. connecting at anode is enough with gpio on the gate. it can do some funky stuff though with the voltage drop of the arduino eating into the logic high voltage reported to the mosfet and stuff though which was starting to eat at me.

but its storted now. i've changed it to common cathode, connecting grounds via jumper & not a resistor due to the arduino's ADC needing to sniff the capacitors. a lot simpler to understand all the voltage references this way.

i thought i'd confused myself somehow with the ass backwards way I was joining these circuits or input ringing but my problem was totally unrelated - it was the arduino shield being a complete pig & mangling my GPIO pins with its library.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf