Switching flipper coils with MOSFET

Thread Starter

Raymond175

Joined May 28, 2019
57
Hello,

I'm building a pinball machine and I'm in the process of designing the Arduino driven flipper coils using a N-channel enhancement-type logic MOSFET. I've been watching Youtube videos and reading all kinds of articles describing on how to wire this correctly.

Now I've created a electrical design that I like you to have a look at and tell me everything that is wrong with it, and if anything is, also what is correct.:)

Diagram (v1).png

Some background information:
- The flipper coil has a dual winding with a low and high resistance path with flyback diodes
- The flipper mechanism has an EOS-switch (S1) that switches the low/high resistance paths
- The EOS-switch has a capacitor (C1) to prevent arching/sparking
- I've added an flyback diode (D3) to prevent EMF flyback
- I've added a pull-down resistor (R1) to switch the MOSFET "off" by default

All the feedback on this design is appreciated!

Kind regards,
Raymond
 

Thread Starter

Raymond175

Joined May 28, 2019
57
As mentioned I'm using a N-channel enhancement-type logic MOSFET.
I clarified this a few moments after posting the thread, so maybe you've missed it.
 

Thread Starter

Raymond175

Joined May 28, 2019
57
Unfortunately I do not have the specifics of the MOSFET component yet, but it might be a 60V / 10A one that I still have to sort out.
 

Thread Starter

Raymond175

Joined May 28, 2019
57
It looks fine but the devil is in the details. The components matter.
Oh, I know :D
But I had difficulties creating and understanding the schemas for the P-channel and N-channel MOSFETS, so I really wanted to see if I had the design right. If the design is fine, I'll look into the details of the components.
 

wayneh

Joined Sep 9, 2010
17,496
Oh, I know :D
But I had difficulties creating and understanding the schemas for the P-channel and N-channel MOSFETS, so I really wanted to see if I had the design right. If the design is fine, I'll look into the details of the components.
My advice for the MOSFET is to go large. By that I mean, if you think you need 10A, spend a fraction more and get 20A. And/or spend to get a lower Rds-on. Sometimes these 'upgrades' cost very little. It's certainly worth looking.
 

Thread Starter

Raymond175

Joined May 28, 2019
57
The diagram is a little poor,
I am assuming you are using the dead end for a termination point and a wire crossing for no connection point, method?
Max.
This is very possible, I have no experience with drawing electrical designs. For that reason I've registered for a 4-week course to learn draw/read electrical designs, but it only starts in May...
 

MisterBill2

Joined Jan 23, 2018
18,167
Do you realize that having a diode across the coil will change the response of the solenoid action quite a bit? It allows the current to keep flowing, which reduces the voltage spike, but since the current keeps flowing the magnetic field collapse is much slower.
And what is the purpose of switching coils? There are many circuits for dropping coil current after a solenoid has moved, and none of them use a switch or dual coils. That is the 1950's method.
 

Sensacell

Joined Jun 19, 2012
3,432
The diode D3 is redundant - the other coils are already clamped with diodes?

I agree that there may be response time issues with the design, its worth testing to see if its slow on release.
 

Thread Starter

Raymond175

Joined May 28, 2019
57
Do you realize that having a diode across the coil will change the response of the solenoid action quite a bit? It allows the current to keep flowing, which reduces the voltage spike, but since the current keeps flowing the magnetic field collapse is much slower.
And what is the purpose of switching coils? There are many circuits for dropping coil current after a solenoid has moved, and none of them use a switch or dual coils. That is the 1950's method.
Are you talking about diode D3? I added that because every tutorial tells me that I need a flyback diode across the coil, but you are right that there are already 2 diodes in the coil itself, is that sufficient to stop back EMF?

And there are not multiple coils, it is a single coil (still used in pinball machines build today) that have a dual winding for the strong kick and weak hold of the flipper. As soon as the flipper has moved in the "up"-position, the EOS-switch switches the coil to weak-mode:

1611660493487.png1611660784838.png

Basically the whole right part of the design is the flipper mechanism itself:
1611660904178.png
 
Last edited:

Thread Starter

Raymond175

Joined May 28, 2019
57
The diode D3 is redundant - the other coils are already clamped with diodes?

I agree that there may be response time issues with the design, its worth testing to see if its slow on release.
Is diode D3 still redundant if you look at the dual-winding coil in the picture in the post above?

And do you expect response time issues only because of diode D3 or in the design all together? And are those issues caused by introducing the MOSFET or microcontroller, because the rest is a standard flipper mechanism still used in the new pinball machines.
 
Last edited:

MisterBill2

Joined Jan 23, 2018
18,167
Really, given that the purpose of the diodes is to protect the mosfet, it is D1 and D2 that are redundant.. All of those inductive spikes travel in the same wire, and the only device that they can damage is the switching transistor. So only the one located at the mosfet is required. That makes the wiring a lot simpler and reduces the chance of wrong-polarity connection errors. AND it puts the components back on the PCB where they are simpler to service.
 

Thread Starter

Raymond175

Joined May 28, 2019
57
Really, given that the purpose of the diodes is to protect the mosfet, it is D1 and D2 that are redundant.. All of those inductive spikes travel in the same wire, and the only device that they can damage is the switching transistor. So only the one located at the mosfet is required. That makes the wiring a lot simpler and reduces the chance of wrong-polarity connection errors. AND it puts the components back on the PCB where they are simpler to service.
Diodes D1 and D2 are part of the coil itself, I can't change anything there, nor does it need any special wiring, as it is all internal in/on the coil itself.
 
Top