DIY Electric Car Forums banner

1 - 11 of 11 Posts

·
Registered
Joined
·
163 Posts
Discussion Starter #1
hello, i have been contemplating making my own controller for a while now but i dont have any experience in complicated cuircits...ect, only basic wiring stuff like cars 12v system. i have some knowledge of electrical devices, like i know what a resistor does, and a diode, and capacitor but every time i read a post about building a controller i find it hard to keep up with them. i was wondering if someone would be kind enough to start me off maybe by explaining the basics of how a dc controller works, like what each( general) part does in the controller. if you guys think this would be very hard to build for me at my electronic knowledge level, just let me know. but i am very interested in learning how to do this.

i need(want) a 144+ 500amp controller for my current ev that has a 72v 500A controller, i am willing to build a smaller 72v 80A controller to start off( if you guys think it would be worth it) and maybe, just maybe if i get good enough (LOL) someday build a high power controller, but thats later....

thanks
 

·
Registered
Joined
·
163 Posts
Discussion Starter #3
yea, i looked them up a few times but they seem to just be selling a kit for 600$ that has directions to put together. if i could buy their items list, and make my own that would be better
 

·
Registered
Joined
·
224 Posts
I have to believe that Pauls DIY controller kit would be your best bet, also. You would get everything needed, AND, have the tech support just a phone call away.

Hard to beat that. Just read the Logisystem threads, to get an eye opener. Just my 2 cents worth.
 

·
Admin: 'one of many'
Joined
·
4,838 Posts
My plan is to buy the P&S circuit boards and the programmed chip and then source the rest of the stuff myself. That saves a lot of shipping of parts to the UK that I can get locally.

You should also have a look at jack bauer's EV BMW thread as he also got the P&S boards and then modified the design to suit what he needed. I am hoping to get jack's help on this when the time comes.

I believe Jimdear2 is also planning to use the P&S controller for his pulling tractor, The Big 13", and hopefully will be posting a 'blow by blow' account (or ouch by ouch account if Jim is doing the soldering!:D) of building it if all goes well.


My feeling on building a controller is this:
It is easy to understand how the controller works in principle.
Use a circuit that produces a pulse width modulation signal, a nice variable square wave. Use that to control a big power circuit that switches on a high voltage and high current DC input to produce a 'chopped' DC output. Vary the width of the controlling pulses and the high power output will make a motor spin at different speeds.

However, if your understanding of electronics means that you don't understand how that works, or how to build any part of that system, or how to test it, or what can go wrong with it, and you get lost when reading about the complexities of how it all works then perhaps it isn't going to be as easy to design and make one from scratch.

That is how I decided not to try myself. I have built power amplifiers in my youthful past, which are sort of the same principle, but I built from designed kits and either they worked or they didn't. I can assemble to a set of good instructions but no more then that.
I know my limitations.:)

If you want to give it a go and can put in the time to learn what you need then by all means go for it, I believe that is how P&S got started.
Here's a link to their thread on EcoModdr.
 

·
Registered
Joined
·
661 Posts
hello, i have been contemplating making my own controller for a while now but i dont have any experience in complicated cuircits...ect, only basic wiring stuff like cars 12v system. i have some knowledge of electrical devices, like i know what a resistor does, and a diode, and capacitor but every time i read a post about building a controller i find it hard to keep up with them. i was wondering if someone would be kind enough to start me off maybe by explaining the basics of how a dc controller works, like what each( general) part does in the controller. if you guys think this would be very hard to build for me at my electronic knowledge level, just let me know. but i am very interested in learning how to do this.

i need(want) a 144+ 500amp controller for my current ev that has a 72v 500A controller, i am willing to build a smaller 72v 80A controller to start off( if you guys think it would be worth it) and maybe, just maybe if i get good enough (LOL) someday build a high power controller, but thats later....

thanks
he basics are simple enough. The devil is in the details.

Woodsmith gave a basic overview. There are a handful of basic components
between the battery and the motor (both which are part of the system) that
make up the controller. The motor is a key component because it serves as
the inductor for the system.

In its most basic form, a motor controller is a buck switching voltage
regulator that takes a high voltage source (the battery) and provides a
lower voltage to the output (the motor). Motor speeds a governed by
voltage, the higher the voltage, the faster it spins. So to get the basic
idea, check out tutorials for buck voltage regulators.

The buck regulator has three basic components, a switch, a diode, and an
inductor. They are organized in a T organization like so:

B-C-S-+-I->output
. | .
. D .
. | .
.....GND......

B: Battery
C: Capacitor Bank
S: Switch
D: Diode
I: Inductor

Note that one end of the battery, the cap bank, and the other end of the
output are also grounded. Finally the diode is reversed biased, so that
when the switch is on, current does not pass through the diode.

When the switch is off, the battery is disconnected from the rest of the
circuit. However, when the switch is on the inductor gets charged.
Inductors have the property that current through them cannot change
instantaneously. Like capacitors and voltage, it takes time for them to
charge and discharge. Think of it like a bucket that doesn't magically fill
up just because you turn the water spout on. It takes time to fill up, and
time to empty.

As the inductor is charged, the voltage to the output rises. When that
voltage reaches the target point, the switch is cut off. Now once again,
the inductor cannot simply get rid of its charge all at once. But if we
leave it disconnected, the voltage will continue to rise as that charge
tries to find an outlet. This is where the diode comes into play. It serves
as a "relief valve" for the charged inductor while the switch is off
completing the circuit between the diode, inductor, and output. So the
output is still supplied with power at the regulated voltage even though
the switch the battery is off. After some time, the charge in the inductor
will start to drain and the voltage will start to droop.

So to keep things regulated the switch is pulsed on and off. When on power
is supplied to the output directly from the battery, which charges the
inductor. When the off, power is supplied from the inductor through the
diode. The ratio of on time to off time determines the voltage for the
output. The longer the switch is on relative to off, the higher the voltage
to the output. So tiny pulses will provide a low voltage, while large
pulses will raise the output voltage significantly.

Finally there is the capacitor bank next to the battery. When the switch
turns on, the batteries are asked to provide a lot of power right now. They
get "slammed" and it's really hard on them. The purpose of the cap bank is
to help with that pounding. Like inductors, the cap bank is charged by the
batteries. But this happens when the switch is off. Unlike inductors
capacitors can be discharged very quickly when you "slam" them. And the
type of capacitors used in these types of devices, called low ESR
(effective series resistance) caps, are not bothered by the pounding. They
will heat up a bit, but not fry. So they provide the buffer for the
batteries so that the batteries don't get punched in the mouth each time
the switch turns on.

Now the details, and there are a lot of them. First that in a motor
controller, the motor serves as both the inductor and the output. They are
one and the same.

Second is that like the caps, the switch needs to be very low resistance
(so it doesn't heat up) and very high power (to carry the charge from the
batteries to the inductor). MOSFETs have both of these properties.

Third is that motors can draw extreme amounts of current when the car is at
a standstill. This is a huge problem because batteries will happily try to
deliver all the current that is available to the target. The problem is
that the switch and wiring (and motor too) will smoke if too much current
passes through them. So motor controllers monitor the amount of current the
motor is asking for, and will cut off the switch prematurely if the current
exceeds a certain level.

Fourth is that MOSFETs need to be driven hard in both directions (turning
on and turning off) because they heat up significantly during the
transistion. But the counter is that it takes a lot of power to get them to
switch fast. So there is a lot of detail that needs to be managed in terms
of the switch driver.

Finally the missing piece is the manager for all of this. In all modern
controllers, a microcontroller pulses the switch, monitors the current, and
gets input from the user (via the gas pedal) to see how fast the motor
should be spinning. So at some level microcontroller programming and
connectivity is on the radar.

I'm sure that I'm missing something, but those are the basic concepts.

It's always been my though that a simple motor controller could be put
together with large simple blocks. Most folks use lots of MOSFETS and
diodes and caps. They build their own boards and program their own
controllers. Any comments on the following substitutions:

Switch: IGBT modules
Diodes: IGBT modules
Controller: Arduino (not my speed, but simple for a beginner)

Nothing can be done for the cap bank. Lots of caps are required.

Thoughts?

ga2500ev
 

·
Registered
Joined
·
163 Posts
Discussion Starter #7
i think the best idea so far for me is to eventually buy the paul/sabrina kit once i save up the money( im 17). i am about 95% done converting my 1999 hyundai accent hatch, all i have to do is finish the rear battery box and underpan, but i think the 72v controller will have a relatively low top speed.

soo, anyways when i build the controller, i will hopefully learn, and study how each part acts, and what it does. then later on try to copy, or build a very similar controller to paul's since i know it is a reliable layout/system.

does that sound like a decent way to see if controller-building is for me?:D

thanks for the suggestions so far guys, this forum has saved my butt a few times already :p
 

·
Admin: 'one of many'
Joined
·
4,838 Posts
Don't forget that the P&S controller is open source so the schematics should be available if you wanted to have a look at it.

However, there is no beating the reserch and development that has gone into it and all the hard work they have done to iron out any bugs over time.
 

·
Administrator
Joined
·
6,060 Posts
Hi OMT,

I have just received my paul/sabrina kit - a real jigsaw puzzle!

All of the BOM's (parts lists) are on the website - including where to get them

If you are strapped for cash start by buying the control board this is all of the microprocessor complicated bits - most of the cost is in the big hairy bits on the power board

Paul and Sabrina are real nice people and I'm sure they will sell you the bits you need
 

·
Registered
Joined
·
230 Posts
There is not a lot of opportunity to spend less than you would on the P&S "Cougar" controller. He did an excellent job of selecting parts that are reasonably priced and readily available.

I could build a controller for less by using parts that I can get for practically free, but you probably wouldn't be able to get the same parts. For instance, scrap PC power supplies usually have diodes, heat sink hardware, good 200V capacitors, and an optocoupler or two. But you won't be able to get the same set of parts from the ones you take apart.

One of the big expenses is the power devices. For 144V you will need 200V or 250V devices. Yes, 35% more than the traction voltage. Especially if you skimp a bit on the capacitors, the voltage spike from turn-off will be pretty high. You can save a little money by using cheaper FETs, but then you'll need more. With more mounting hardware, connections and a bigger box. Or save more money by using older IGBTs. But then you'll be designing your own driver circuit. And then really be spending money, when you end up with a nice collection of burnt-out parts as you learn what can go wrong.
 
1 - 11 of 11 Posts
Top