To further clarify, the symbol for an inverting driver should look like this:

​

Those tiny circles at the front of the triangles representing the gates mean that the output is inverted with respect to the input. And the lines above letters A and B at the output labels mean the same thing.

 

That you've been using the wrong dirver is a real posssibility. But this diagram that you've shown is of a normal, non-inverting driver:

Those tiny circles at the front of the triangles representing the gates mean that the output is inverted with respect to the input. And the lines above letters A and B at the output labels mean the same thing.

I was basing my answer on the data sheet given. I hate it when the data sheets don't have a "truth table" included when they are logic based chips.

 

Thanks to Shortbus's observation, I made a little progress today.

Having the outputs high and dropping them on the push of a button worked. The lamp pulsed 4 times. I expected it to stay on continuously through the 4 pulses, but it didn't. That's interesting to me. There is no delay between pin activation. The lamp gives a good illustration about how long it takes the current to get flowing... I guess. Is that how long it will take for a magnetic fields to build?

After testing with the lamp as circuit protection, I swapped the lamp for a shunt and pressed the button again. The 10A fuse blew immediately. Not surprising. I then swapped out the blown 10A fuse with a 20A. The first MOSFET blew AS I CONNECTED the battery. Current should NOT have been flowing until I pressed the button.


I guess the MOSFET was damaged during the first attempt that blew the fuse? But why? Also, why can't the MOSFET handle 10 Amps? I have to dig up the data sheet again, but I'm pretty sure it was rated at least 20 amps.

 

Those tiny circles at the front of the triangles representing the gates mean that the output is inverted with respect to the input. And the lines above letters A and B at the output labels mean the same thing.

Thank you. Some questions I have in LTSpice just vanished.

 

@Ben Varvil and cmartinez

Did you install pull-down resistors to ground at their inputs?

C, did you mean that? Or did you mean 10k between gate and source on the mosfets? Without the 10k between G and S if the drive is shut down with a gate still charged won't it still be 'on' if the power is turned back on, unless some time has elapsed?

 

@Ben Varvil and cmartinez



C, did you mean that? Or did you mean 10k between gate and source on the mosfets? Without the 10k between G and S if the drive is shut down with a gate still charged won't it still be 'on' if the power is turned back on, unless some time has elapsed?

I meant pull-downs at the driver's inputs. I didn't read the driver's datasheet, and some of them have open collector inputs, so a pull down resistor is needed. Ben hasn't posted exactly which driver he's using, and I'm unfamiliar with his controller's i/o characteristics, so I suggested those resistors just to be on the safe side.

 

Ben hasn't posted exactly which driver he's using,

Interesting, the numbers on the chips are no longer there. Maybe they burned off?
Their brothers still in the case read: MC34151P then AKAA then 1806G

I looked at the IRL540 MOSFET data sheet again: https://www.vishay.com/docs/91300/91300.pdf
It looks like 28A is the "avalanche current" absolute maximum... which feels like it applies to my application.

As was mentioned earlier, I'm likely pulling much more current than I think I am. So, it feels like time to redesign the coils. I'm balancing the numbers: Wire gauge, number of coils and overall resistance... It feels like the system I have will be able to handle 3A, motivating me to shoot for 4ohms. Does this sound like a good plan? 4ohm, coils?

THanks for any thoughts.

 

I looked at the IRL540 MOSFET data sheet again: https://www.vishay.com/docs/91300/91300.pdf
It looks like 28A is the "avalanche current" absolute maximum... which feels like it applies to my application.

As was mentioned earlier, I'm likely pulling much more current than I think I am. So, it feels like time to redesign the coils. I'm balancing the numbers: Wire gauge, number of coils and overall resistance... It feels like the system I have will be able to handle 3A, motivating me to shoot for 4ohms. Does this sound like a good plan? 4ohm, coils?

THanks for any thoughts.

I think it's worth a try.

 

You might look into the specs on these 2 diesel fuel solenoids. They both go BANG when you energize them.

This one for Westerbeke generators: https://www.westerbeke.com/service bulletin/sb_175.pdf
is on a 10 amp fuse and is fast and strong.

This one is an emergency stop solenoid for old GM diesels.
https://www.amazon.com/Trombetta-Heavy-Tubular-Solenoid-P515-A57V12/dp/B007W2NFJ2

It has to pull a release lever on 2 spring loaded air intake doors.

"I believe this solenoid has about 1.5'' or so travel and right around 70lbs hold in force."

 

The coil in the video I posted pulled 5 amps It got warm really fast too.

You show the driver chip with no numbers that can't be good.
If you want this to work you need to use smaller wire and make the coils right cause what your making will not do what you say you want.

Post what code your using on the arduino

The driver chip you say your using needs pull up on the input You need 4 diodes on the gate to ground.
This is what your doing your killing your gate driver the dang thing is not cutting off cause the input is not turning off your gate putlling to much power
thats 10 ohm there you have on the gate you do no that means 1.2 amps could flow out to the gate.

That driver has a

Whats that say the last line 400 mA the max clamped for the whole chip is 1 amp you done cooked the name off the chip.
this is how you should look

 

Post what code your using on the arduino

Thanks for your post. Still processing. In the meantime, here's the code.

 

You show the driver chip with no numbers that can't be good.
If you want this to work you need to use smaller wire and make the coils right cause what your making will not do what you say you want.

I'm on board to wind new coils with thinner wire. I'm trying to make better decisions about gauge, resistance and amperage... This is proving to be tougher than I anticipated. The Trombetta people (thanks DNArobotics) tell me the P515-A57V12 pulls 13amps with a coil resistance of .92ohms.
"What gauge wire?"
"it's pretty thin."

I'm doing some back of the envelope calcs and getting pushed towards using thicker wire...
For example:
If I selected a thin, 30 gauge (0.01" diameter) wire with 103ohms per 1000' I'd only need 38 feet to get my 4ohms. 4 ohms gives 3 Amps, but 30Ga wire gets to 1600F at 3 amps. So no go.

25 Gauge wire (.0179" dia) gives 4 ohms at 123 feet. 3 Amps gets 25 gauge to 800F. 123 feet is better, but that is only 20 wraps around my 2" diameter tube.

16 gauge (0.0508" dia) gives 4 ohms at 1000'. 6.4 amps gets it to 400F, which makes me think 3 amps would produce manageable heat. 1000' also gives me 160 wraps. BUT, 1000' of 16 gauge wire is 7.81lbs. 4 coils at 8 lbs each is pretty ridiculous and super expensive.

Could you please explain what you mean about using thinner wire? Is there an assumption that I have to use much less current? 0.5A?

I've been studying your center tap design (what I'm assuming you mean by "make your coils right") with alternating fields. Here is what I see when I compare alternating vs non-alternating:

I recognize this is over simplified, but it seems that alternating the fields would push the plunger back, not advance it forward. What am I missing?

The driver chip you say your using needs pull up on the input You need 4 diodes on the gate to ground.

I did reading on pull-up resistors. Thank you. I feel like I get it. I still don't understand what the diode from the gate to ground does, but I'll trust you and hopefully figure it out later.

thats 10 ohm there you have on the gate you do no that means 1.2 amps could flow out to the gate.

No. I thought the 10 ohm resistor on the gate was to prevent ringing. I see now that V=IR says I'm pumping 1.2Amps to the MOSFET. I see you selected 30ohms so that the gate sees 400mA.

Whats that say the last line 400 mA the max clamped for the whole chip is 1 amp you done cooked the name off the chip.

I agree, I definitely cooked the driver chip. Why does the specification list chip performance for 10mA, 50mA and 400mA. Is that just to give designers a spectrum? and designers are to assume the highest value is the max?

I'l wire up new driver chips, swap out the MOSFETS, add pull up resistors, gate/ground diodes and swap out the 10ohm for a 30ohm. From there, I'll fuse the system at 15 amps and see if I can get motion out of the magnets.

If you were to further clarify what you mean by "make your coils right" that would be much appreciated.

Thanks again for your post. Got me thinking.

 

The diode on the gate keeps the gate from going negative. That keeps your driver outputs from burning out.

The coil with center taps make's this so you can step both ways the same as a unipolar stepper motor.
your trying for a Bipolar which needs a

Bipolar drivers use H-bridge circuitry to actually reverse the current flow through the phases. By energizing the phases with alternating the polarity, all the coils can be put to work turning the motor.

Unipolar vs. Bipolar
Unipolar drivers, always energize the phases in the same way. One lead, the "common" lead, will always be negative. The other lead will always be positive. Unipolar drivers can be implemented with simple transistor circuitry. The disadvantage is that there is less available torque because only half of the coils can be energized at a time.

 

You might get other opinions, but I don't think your coils will get very hot if they are only energized for a fraction of a second.

 

His code has one coil on for 1.4 seconds and I'm sure it getting everything hot LOL

 

This table might help you. Determine the resistance you need, then find wire that will carry that current, then determine how much for each coil.
https://www.pupman.com/listarchives/1998/April/msg00222.html

 

His code has one coil on for 1.4 seconds and I'm sure it getting everything hot LOL

two things: First, I looked at my code again and realize that there IS a delay between coil activations (200ms), which explains why the lamp pulsed rather than being on continuously. Second, output 13 is just the on-board LED to tell me when something is happening. It is not connected to a coil.

 

I don't think you understand how a coil works
you fire 1 2 3 4 your loop is 1.4 seconds that coil is on for 1.4 anyway you look at it.
Like I said it's made wrong the coil needs to be at most 24 or 22 gauge wire.
and 4 ohm and I would center tap coils would work best they can pull or push the shaft the easy way
The one I made is 1 ohm 24 gauge and I get's hot at 5 amps and it was on maybe half a second.

 

I don't think you understand how a coil works
you fire 1 2 3 4 your loop is 1.4 seconds that coil is on for 1.4 anyway you look at it.
Like I said it's made wrong the coil needs to be at most 24 or 22 gauge wire.
and 4 ohm and I would center tap coils would work best they can pull or push the shaft the easy way
The one I made is 1 ohm 24 gauge and I get's hot at 5 amps and it was on maybe half a second.

I realize my windings look like one coil, 1" wide. But there are four separate, 1/4" coils, each with independent access to power. The code is push button activated... so each coil is activated for 200ms then everything goes off, waiting for the next button press. I don't see where "that coil is on for 1.4 anyway you look at it" is coming from.

22 Gauge wire gives 4 ohms at 250'. I'm seeing charts saying current rating is 3A. I'll go this route.

 

If button = high
{
Do this
means it dose
everything that's here
you have a 1.4 second delay
guess what it does that
your coil is on for 1.4 seconds
}
Now read up on how a coil works you'll see that it is trying to move power even after you turn it off so your coil is on for the whole loop
you told it
(If button = high) do this what happens is if this was 1 the (do the loop) starts happening it doesn't check nothing till it's done.
So you can hold the button down or press once all the do stuff will happen till it get's done.