Electric Motor Secrets

[Jetijs]

Thanks guys
Here's an update:





I put the circuit in an acrylic box and all the wires go to plugs on the aluminum plate, so that it will be like plug'n'play It looks messy with all the wires, but that's ok if everything works well. There are two plugs for each motor. One for timing and other for power. There are four switches so that I can switch each pair of coils on and off. This will allow me to adjust each phase of each motor separately to get the best timing and performance. There are also four red LEDs above the switches, each one indicates which pair of coils is currently running. Those red LEDs are great because with a 3k resistor at 12v they consume only 1mA each and still glow rather bright. The box has an acrylic cover with a hole cut in the middle right above the MOSFET's, this is intended for a cooling fan.
So with a bit of luck, tomorrow both motors will run
If everything goes well, then all I will have left to do is find a good coupler to connect both axis together.
Thanks,
Jetijs

[StevanC]

prosperous

StevanC,
I took your advice and used those "clamps" along with terminal bloks, that you can solder in the circuit board. I found only one way how to arrange the MOSFET layout so that I can get to all of them with a screwdriver. Also my previous board had the MOSFETs too close together so that I could not attach heatsinks to them. Now the new board is ready and everything looks good:



Note that heatsinks are still not attached. Every MOSFET driver has its own socket so that I can easily change them if the need arises. Also each MOSFET can now be easilu changed. No wires are soldered on the board, everything will be attached via those blue terminal blocks. Now I need to test the circuit, put it in an acrylic box and attach all this to the motors. It's a shame that I have so little time to work on that
Thanks,
Jetijs

"You are on the narrow path of becoming a mighty jedi over time, young padawan"



That's the way I like it (the board) - may the mighty Murphy be with You ;-)


all You have left to do is to find time for time...

[Mark]

Well you better make the time! We can't sit around here all day waiting for you to get this motor finished, get to it! If you dont get that thing running in the next 2 days I'll have to come and confiscate it

OK so I'm a little anxious to see it finished, what can I say other than great work and I'm still jealous.

[Jetijs]

StevanC,
I took your advice and used those "clamps" along with terminal bloks, that you can solder in the circuit board. I found only one way how to arrange the MOSFET layout so that I can get to all of them with a screwdriver. Also my previous board had the MOSFETs too close together so that I could not attach heatsinks to them. Now the new board is ready and everything looks good:



Note that heatsinks are still not attached. Every MOSFET driver has its own socket so that I can easily change them if the need arises. Also each MOSFET can now be easilu changed. No wires are soldered on the board, everything will be attached via those blue terminal blocks. Now I need to test the circuit, put it in an acrylic box and attach all this to the motors. It's a shame that I have so little time to work on that
Thanks,
Jetijs

[StevanC]

Oh, you mean those. I use them where I can, but this time I need something I can solder in a bread board. Just like those chip sockets. Also it is very hard to use those clamps if you have many transistors/MOSFETs close to each other, because you just can't reach the screws with the screwdriver.

"Use the 'force' and clever layout, young padawan."


I take care all my power elements are easy to reach/replace. And if I succeed, murphy's law protects me:

"A part that is easy to replace, newer fails: the part most hard to replace will fail instead."
COROLLARY:
"easy to replace designs are most rough"


so use low RDSon (5-7mOhm now(!)) and put them inline over a trivial heat conduct in a favorable step (pin wise) to just screw them on thew rail - from the other end of the rail make 'spaghetti' to the bread board?

But what do I know?

[hh1341]

Oh, you mean those. I use them where I can, but this time I need something I can solder in a bread board. Just like those chip sockets. Also it is very hard to use those clamps if you have many transistors/MOSFETs close to each other, because you just can't reach the screws with the screwdriver.

Jetijs,

How about sticking a small breadboard onto your circuit board?

You can also hack up those sockets with a sharp knife, to hold any number of leads(3 in your case)The sockets I am referring to are the DIN chip sockets.
Carl

[Jetijs]

Oh, you mean those. I use them where I can, but this time I need something I can solder in a bread board. Just like those chip sockets. Also it is very hard to use those clamps if you have many transistors/MOSFETs close to each other, because you just can't reach the screws with the screwdriver.

[StevanC]

Do you perhaps have a picture? I could not find anything on google

phew!

RadioShack.com - Cables, Parts & Connectors: Connectors & connectivity: Terminal strips & barrier blocks

:-)
helps?

[Jetijs]

Do you perhaps have a picture? I could not find anything on google

[StevanC]

Thanks StevanC
That is a great idea! I wonder if there is something like that also for the MOSFETs? That would make changing fried MOSFET's very easy
Anyway, I already disconnected the inverted driver chip from the rest of the circuit and soldered in a new driver on the free space of the board.
Thanks,
Jetijs

Actually it is (!)

it's called 'clamp' and has two holes and a screw for each:

|oooooo|
|oooooo|

^-- looks like this...
;-) cut a three rows or sixer out and just use.... (did You see my Youtubed SSG?)


I can fit an MJL to it and have my SSG run, then i just swap various BJTs and see how they compare:
I ran my from the MPSA06 and he is the winner among the small (to93 sized) ones in front of the venerable but fast 2N2222...
the draw was 320mA to 270mA

[Jetijs]

Thanks StevanC
That is a great idea! I wonder if there is something like that also for the MOSFETs? That would make changing fried MOSFET's very easy
Anyway, I already disconnected the inverted driver chip from the rest of the circuit and soldered in a new driver on the free space of the board.
Thanks,
Jetijs

[StevanC]

Jetijs,
I can only advice allways use socket for 4+pin packs.
I intend to make my next 555 swither on a socket only (with two optos inluded) - socket rules in case like Yours.

Don't have vacuum solder sucker?

[Jetijs]

Started testing the circuit. Found two problems. One set of two MOSFETS has the UCC37322 MOSFET driver and not UCC37321. This is an inverted output driver, so two sets of MOSFETs are working in inverted mode I don't know how I could make such a mistake. Also one of the rest six MOSFETs has and transil diode across the drain and source that is rated at 100V and not 400V as the rest. This makes the MOSFET oscillate in low frequency rather than turn it steady ON when the optoswitch triggers. I can easily chanfe the transil diode, but resoldering the inverted MOSFET driver will be hard if possible with the equipment I have. I guess that I will rather just disconnect the inverted driver from the rest of the circuit and solder another non-inverted driver somewhere else on the circuit board.


Edit:
No, the transil diode on that one MOSFET was not the cause of the oscillations. I guess that the MOSFET is just dead.
Edit: Yes, te MOSFET was dead.

[Jetijs]

Sorry Mike,
I missed your post. Yes, the inductive spikes will be collected and sent into the primary capacitor so that much less energy will be needed from the power supply. In my previous schematics you see just the driving circuit with no recovery part. This is the exact reason why I am using bifilar coils, because one coil will be used as primary power coil and other coil will be the recovery coil. You have to use bifilar coils in order to get the output right beck to the input. You can also use just single strand coils, but then you will have to send all the inductive spikes in a second battery or somewhere else. Here is the circuit with the recovery part (this is for the V2.0 motor):

[mlurye]

Jetijs,
Are you planning to utilize BEMF in your system? For now (based on your schema) it will affect performence of your motor.
Just as an idea. I'm not sure if this schema will work.