No schottky diode needed? 300a paralleled mosfet switch

Thread Starter

bigjoncoop

Joined Feb 1, 2019
185
I'm using 12- 100a mosfets connected in parallel as a ~300a switch for a battery spot welder. I posted a question a couple weeks ago about adding flyback diode protection and the consensus was that it wasn't needed for this application.

I just finish building the mosfet array last night. Going over my schematic and other DIY projects that are similar to this, just to double-check everything is correct. But all the other DIY mosfet switches like this are either using TVS diodes and or large schottky diodes and I just want to make sure I should be fine without adding them

The project that I followed in order to make mine has a large 300amp schottky diode between the drain and the source.

Just want to clarify if it's necessary... Thx


Mosfet welder zoom schematic.JPGMosfet welder schematic.JPG
 

bertus

Joined Apr 5, 2008
22,266
Hello,

Your source connection is incorrect.
It is now connected to the gate.
Move the connection to the other side of the 1K resistor.

Bertus
 

MisterBill2

Joined Jan 23, 2018
18,061
In order for a FET such as these to switch on, the gate must receive a drive signal. For the circuit shown only the first device gat will receive the signal. That is a problem.
 

Sensacell

Joined Jun 19, 2012
3,420
Think about what the negative lead (drain terminal) looks like from the outside world.

It's just a bunch of floating MOSFET drains, with zero protection from positive overvoltages. (think ESD for example)

You should place a fast clamping diode from the Drain to the positive supply, to clamp the max voltage seen by those MOSFETS, or for sure the first time you drag the cable over the carpet the MOSFETs will be destroyed.

If there is any significant inductance in the welding lead circuit, this could also generate large voltage spikes at the drain.

Consider a clamp and a TVS.
 

MisterBill2

Joined Jan 23, 2018
18,061
Think about what the negative lead (drain terminal) looks like from the outside world.

It's just a bunch of floating MOSFET drains, with zero protection from positive overvoltages. (think ESD for example)

You should place a fast clamping diode from the Drain to the positive supply, to clamp the max voltage seen by those MOSFETS, or for sure the first time you drag the cable over the carpet the MOSFETs will be destroyed.

If there is any significant inductance in the welding lead circuit, this could also generate large voltage spikes at the drain.

Consider a clamp and a TVS.
A simple solution will be a resistor, about 1000 ohms or maybe 2200 ohms, directly between the battery positive and the drains. that will allow anu stray voltage an easy path away from the MOSFETs , but not affect the current used to produce the weld. And still the circuit appears to be incorrect, because the control is done with the gate to source voltage. And I see no connection for any of the main current to the source of any of the devices. Thus the circuit as drawn can not do anything useful.
 

MisterBill2

Joined Jan 23, 2018
18,061
Now the circuit shown in post #8 will be able to function. So still there is a question of putting a diode some place to protect the MOSFETs .Each one of those MOSFETS has an intrinsic reverse diode, so a reverse voltage will not rise very high at all. With he low value of the gate to source resistor the transistors will switch off fairly rapidly, and so some voltage created by the collapsing magnetic field in the series inductance could be expected, but do not see a series inductor in this circuit. When the weld is completed it is probable that the two electrodes will still be in contact withthe material being welded, and so it is not likely that much voltage will be developed between them.
A diode connected across the electrodes will not cause any problems if the polarity is correct, and so it could be put in. But a 12 volt light bulb can also be used, which will provide the benefit of being able to verify the functioning of the circuit without needing try to do a weld . So that may be a useful choice.
 

Thread Starter

bigjoncoop

Joined Feb 1, 2019
185
Thank you all for all of your input / help.

My actual mosfet switch is built correct as shown in the schematic on Post #8. I just didn't draw the schematic correctly. Thx you for pointing that out.

I've included in updated schematic and a picture of the project I am using for reference.

I would have thought that IGBT's would be the preferred choice in that application?
Max.
I was thinking myself that IGBT's would be have been a good choice but it seems that most of DIY mosfet welders are using mosfets. When making mine I followed another persons build video/guide to make sure it would work. Which I've included and below. I've had tried to make a couple different DIY battery spot welders recently and none of the gave me the results I'm looking for....

A simple solution will be a resistor, about 1000 ohms or maybe 2200 ohms, directly between the battery positive and the drains. that will allow anu stray voltage an easy path away from the MOSFETs , but not affect the current used to produce the weld. And still the circuit appears to be incorrect, because the control is done with the gate to source voltage. And I see no connection for any of the main current to the source of any of the devices. Thus the circuit as drawn can not do anything useful.
If I could just use your idea with using resistors as protection that would be great.
And I'm Not exactly sure what you mean by no connection for the main current to source.

Images:

Weld schematic 2.JPG
This is the project I'm using as a reference:
Weld example-1.JPG

Weld schematic 3.JPG
 

MaxHeadRoom

Joined Jul 18, 2013
28,576
Is the reason for going with a DC battery version due to portability feature?
Otherwise the mains powered AC transformer method is far easier, with primary triac switching.
Max.
 

MisterBill2

Joined Jan 23, 2018
18,061
My reference to a lack of any connection was based on the initial drawing that was missing a few lines. The corrected version looks much more correct.
 

Thread Starter

bigjoncoop

Joined Feb 1, 2019
185
@MisterBill2

you mentioned putting a 1k or 2.2k resistor directly between the battery positive and the drains...

but since the battery positive doesn't connect to the circuit I'm assuming I should run a wire from the battery positive terminal to the resistor connected to the drain.

Also since all the drains are tied in parallel can I just use one resistor or should I add a resistor to each of the mosfets drain going to battery positive?


Weld schematic 4.JPG

OR

Weld schematic 5.JPG

THANKS @MisterBill2
 
Last edited:

MisterBill2

Joined Jan 23, 2018
18,061
One resistor only, and the one purpose of that resistor is to avoid an open circuit when the welder is not in place to do a weld. That will greatly reduce any static build up or other random voltage developing during the setup. It would only need one single resistor. Another option would be a 12 volt light bulb, which would also be a way to do a quick check to verify that the system was working before setting up for a weld. It could also provide a warning that the electrodes were energized.
 

MaxHeadRoom

Joined Jul 18, 2013
28,576
Every spot welding machine I have worked with had a electrode pressure detection of kind before power is allowed, it can also be dangerous if used without and no eye protection.
Max.
 
Top