36v, 300A motor controller,

[cobra1]

I want to build my own motor controller, to control a SepEx motor (seperate field coil)

The motor is a 36v 7.5kw motor (10kw max) so maximum current draw is 278amps

So i have decided on the following specifications,

36v, 300amp capability
Reverse Function
Regenerative braking function

I have looked on google for suitable schematics etc, but there all for low power motors 5-12v

Does anyone have any schematics for something of this size, or any pointers on where to find one??

If not an idea of what is required to start building this would be great,

Cheers in advance

 

[tcmtech]

It sounds like a standard issue golf cart or EV motor controller.
If your going the DIY approach I would go way past the 300 amp capacity and push it to around at least 600 or higher. High current motor control life expectancy greatly improves if you double or triple (or higher still) the switching device amp capacities.

How complicated and how expensive are you thinking?

 

[cobra1]

I would like to keep it fairly straight forward, cost wise probably max of £100 - £150.

i could no doubt buy one of these off the shelf but i would like to learn how they work and if it fails during trials i can fix it easily

 

[tcmtech]

I dont know what components you have available for cheap parts but that doesn't sound like much to work with money wise.
What you will need for the ability to go forward and backward is either a solid state H bridge design or a large electrical contactor system that reverses the polarity of either the rotor power or the field power.

The actual speed control is basically the same as the small DC motor driver designs. Its just the power handling circuits and components that are bigger and the current feedback system that gets changed.

 

[crutschow]

I doubt that you can switch 36V @ 300V for the budget you have.

 

[Boncuk]

I would like to keep it fairly straight forward, cost wise probably max of £100 - £150.

Did you omit one ore two zeros before the decimal point?

My guess is 1,000 to 1,500

 

[tcmtech]

A new factory built golf cart controller with regenerative braking and higher power capacity seems to be running around $350 - $700 on line depending on source and options.

 

[cobra1]

those prices arent too bad, i have found a partial circuit of what im looking for, i will work out a price for it to be built and see which turns out the best option

 

[Ubergeek63]

I want to build my own motor controller, to control a SepEx motor (seperate field coil)

The motor is a 36v 7.5kw motor (10kw max) so maximum current draw is 278amps

So i have decided on the following specifications,

36v, 300amp capability
Reverse Function
Regenerative braking function

I have looked on google for suitable schematics etc, but there all for low power motors 5-12v

Does anyone have any schematics for something of this size, or any pointers on where to find one??

If not an idea of what is required to start building this would be great,

Cheers in advance

well you start with 5 FETS like these:

Digi-Key - IRLB3036PBF-ND (Manufacturer - IRLB3036PBF) 26 USD/10

and five gate drivers:

Digi-Key - 296-1987-1-ND (Manufacturer - TPS2829DBVR) 2 USD ea

and a basic PWM chip like this:

Digi-Key - 497-3079-5-ND (Manufacturer - UC3842BN) 1 USD ea

add a heat sink (dissipation around 50W) and a handful of parts and you have it.

 

[tcmtech]

I am rather curious as to how they get 190 amps through the pins of a TO-220 device and how it can manage 380 Watts dissipation as well on that package size?
Do they use solid gold circuit traces and solder while mounted to cryogenic cooled heat sinks?

I would be concerned about using a 60 volt device on a 36 volt high current H-bridge motor driver system. Thats just not much headroom for any brush noise or inductive spikes. It can be done but a considerable amount of additional snubbing and bypass diode capacity is needed plus substantial line side capacitor capacity for that snubber and bypass energy to go into.

I have a basic E-Z-GO 1206 (36 volt 300 amp) three wire speed controller that uses twelve 60 amp 150 watt rated Mosfets as the switching system and is just a single direction system. The input line has fourteen 1000 uf 50 volt capacitors. Plus the Mosfet banks have six high current bypass diodes and a number of smaller bypass capacitors as well. There also is a built in high current shunt for the current sensing and limiting systems.

An H-bridge version would need four times this amount of switching device capacity, bypass diodes and snubber capacitors. Plus a more complex logic control system on top of all of that.
It certainly could be home built but the realistic price may get way high real fast!

I picked mine up as a new old stock item for around $100 last year. If I added a set of 5 lead contactors (about $40) I could easily make a bidirectional drive unit out of it while still retaining all the control functionality. A few high current diodes (about $20) in the right places and it could have some degree of basic regenerative braking ability as well.

 

[Ubergeek63]

I am rather curious as to how they get 190 amps through the pins of a TO-220 device and how it can manage 380 Watts dissipation as well on that package size?
Do they use solid gold circuit traces and solder while mounted to cryogenic cooled heat sinks?

they use multiple FETs in parallel with inverter rated substrate diodes and synchronous rectification ( shorting out the substrate diodes with the FET as required ).

this keeps the total dissipation under 100W for a full bridge.

They might brag about 380W but if you look close they say that is at 25C case and 125C or 150C junction temp ... an infinite heat sink. a TO220 can take only 2W in open air... it all depends on how much heat sinking and reliability you want. it is best to keep the junction below 100C though 75C would be better for high rel

Dan

 

[Nigel Goodwin]

I suggest you have a look at this site:

Battery Motors and Controllers - FAQs

Only some information is free to non-members, but it should give you some idea of what you're attempting, and may encourage you to join up.

 

[tcmtech]

They might brag about 380W but if you look close they say that is at 25C case and 125C or 150C junction temp ... an infinite heat sink. a TO220 can take only 2W in open air... it all depends on how much heat sinking and reliability you want. it is best to keep the junction below 100C though 75C would be better for high rel

Thats one of the parts about specs and data sheets. The numbers are often in regards to unrealistic or ideal conditions that the device is not likely ever going to be used in.

I personalt try not to use the TO-220 case devices in any application that will likely see much over a 50 watt dissipation number for more than a few seconds.
Most aluminum heat sinks just dont carry heat away fast enough to keep the junction temps down at higher power level in that package size unless they are way over sized for the application.