Electric Mountainboard Project - 72V 45A controller with 7S

[Pedrohlt]

Hi,

I'm working on a electric mountainboard. Here is the specifications:

Board - MBS Atom 95X
Motor - Turnigy 63-74 200kV Outrunner
Batteries - 6 or 7 ThunderSky 40Ah LiFePO4 cells
Controller - LVC Disabled E-Crazyman 72V 45A 1500W (I blew up my Suppo 100A ESC)
Throtle - 10k slide pot

Here is some pics:




Hall Sensors Mount

SS411A



The problem is that the controller don't respond to any command, the only thing I noticed is that when I plug the conectors it sparks(maybe cause the capacitors are charging). I thing the problem is that I'm using only 6S LiFePO4 (I have one spare cell to make 7S) and the controller don't work with low voltage. When I bought the controller from ecrazyman I asked if he could disable the LVC because I'm gonna use 6S, and he said its ok, he can disable the LVC. So, I though that with the LVC disabled the controller should work with 6S, but it's not. And I don't know if the LVC is really disable.

So I'm asking you guys how can I modify the controller to work with low voltages(6 or 7S) ?

Pedro.

 

[enoob]

dude.

sell those thunderskys while there still worth something and before they crush your skull against a rock and use the money to get some rc 22v cells.http://www.hobbycity.com/hobbyking/store/uh_viewItem.asp?idProduct=9176 for the same size and weight of one of the TS cells youll have enough power to kill yourself . plus some extra cash id bet

 

[kfong]

From the last picture, you need 36volts to even run the controller. It has nothing to do with the LVC. You have way too much capacity and not enough voltage. Get some RC packs to reduce the size like enoob says or your bound to break an ankle with all those packs in the way.

 

[Jeremy Harris]

I've modded a 6 FET controller to work on a 4S pack, and it can be done OK, but as it stands I don't think that controller will run on less than maybe 30V or so.

If you want to run on around 20 to 24V, then you need to remove (or short out) all of the dropper resistors for the 12V regulator (the big resistors next to the capacitor at the bottom of your photo). This will let the 12V regulator work OK at the lower voltage, which will then allow the 5V regulator to also work. If the LVC is totally disabled, then there's nothing else you need to do, as the unit should work OK on any voltage from around 18V minimum up to around 36V maximum.

If you want to run at voltages below 18V, then you also need to remove the 12V regulator and short its input and output pads. This gives you a controller that will work from around 12V up to about 16 or 17V, but anything over about 18V will possibly blow the FET gates.

The advice to swap the big Thundersky cells for some nice, light, LiPo packs is sound, in my opinion. Not only will they be a lot lighter, but they will also be much more compact, so will take up much less room on the deck. You may even find that you could fit slim LiPo packs (in a suitable box) under the deck, as they would most probably not be much deeper than the trucks/axles.

Finally, have you tested the Hall sensors and motor? Do you know that the Halls are in exactly the right place and correctly wired for each phase wire? Are you getting good pulses from the Halls?

Jeremy

 

[Pedrohlt]

Jeremy, I was thinking the same thing that you said, to mod a 6 fet controller. Did you mod a 24 or 36v controller? I was thinking to mod an ecrazyman 24v 250W controller (http://tinyurl.com/yfl5q9x) with 6 IRFB4110 fets. Do you think it will handle a Turnigy 63-74? When you say to test the hall sensors is with an ociloscope or with a multimeter?

I know that LiPO is lighter than LiFePO4, but I choose the ThunderSky cell because they are safer, cheaper and have longer life cycle than LiPO. And if I say to my parents that I will have to buy more batteries, they probably will kill me :lol: . Maybe in the future I change to A123 or if it takes to long to Li-Titanate .

Thanks all for helping.
Pedro.

 

[kfong]

You might be able to take apart the packs and flatten them, place some of them on the bottom of the board. You can then put a sheet of wood to protect them. This will allow more room for the top.

 

[Knuckles]

Pedrohlt,

Before you mod anything you can easily test the system no-load.
Simply use some 9V batteries for the ignition circuit. 4x9=36 volts may do it.

The ignition circuit and the main bat circuit to the mosfets are completely independent.
Chances are the LVC is not really disabled but rather just set to 20v in the software code.
If you want I'll show you how to at least no-load test your rig before you spend any more money.

-K

 

[Jeremy Harris]

If you're going to fit new FETs, then it doesn't matter which model of 6 FET controller you get. Keywin will sell controllers with no FETs fitted, but normally needs an order of a few (5 minimum, I think). They are all programmable with a fairly simple interface (see the various threads detailing the method), so the LVC and current limits can be fairly easily adjusted.

If you're running at low voltage, then there's no point in fitting IRFB4110 FETs, as they are far from the best at anything under about 60V. Their advantage is the 100V Vds, coupled with a fairly low value of Rdson, but there are lower Rdson FETs available if you're operating at low voltage. I've been using IRFB3077 FETs,as they have a much lower Rdson than the IRFB4110s and are still OK at the sort of voltages you're running at. There are other FETs around with low Rdson and gate charge values that you could look at if you only want to work below 40V.

Knuckles may be right, Keywin could have just set the LVC to it's lowest value of around 21V. His test method will show if that's the case. You can fool the LVC easily enough, by reducing the value of the resistor that connects to the supply to sense supply voltage. I did this with my 12V controller and have found that it works well.

Your safety arguments for using LiFePO4 are sound, it's only the bulk of the cells you have that looks to be a potential problem.

Jeremy

 

[Pedrohlt]

Jeremy,

I tested the setup adding a RC 2S LiPO and it worked ok with 30-32v. So I'll need to change the power resistor to a lower resistance ones, but I don't know the value of the resistors for the controller work with 7S LiFe.
I measured the resistors:
R1a - 322 ohms (330ohms)
R1b - 322 ohms (330ohms)
R6 - 652 ohms (680ohms)
There is a jumper shorting R01b and the ignition wire.

With the jumper connected (36~60v) - 974ohms (1010ohms) and with the jumper disconnected (72~90v) - 1296ohms (1340ohms). So what value would be the new resistors for the controller work with 7S?

Pedro Henrique

 

[Jeremy Harris]

Pedro,

If you're only working on around 22V or so, then you can get rid of all the power resistors and just short the pad where the power lead is connected to the input of the LM317 regulator.

The only other problem you may have is with the programming of the controller. The lowest cut-off voltage that can be programmed is 21V, so even if yours is set for this value, you won't have a lot of leeway between the normal operating voltage (7 x 3.2 = 22.4V) and the low voltage cut-off.

There is a way to modify the low voltage cut-off, by shorting one of the two series resistors that form the top part of the voltage divider that feeds the sense circuit. There are some photos showing how I did this on this thread: http://endless-sphere.com/forums/viewtopic.php?f=2&t=16651

Jeremy

 

[Pedrohlt]

Jeremy,

On my controller the R01a and the R01b are connected to the Vin of the LM317T and the R01a, R01b and R6 are connected to the Vout. So I have to short the ignition wire to the Vin and let the resistors connected to the Vout, like this ?

Pedro Henrique

 

[Jeremy Harris]

Yes, that looks OK. This will allow the voltage regulator (the LM317) to work OK with a 22V input, but will put an upper voltage limit on the board of about 36V, as the voltage regulator can only take about 40V maximum.

Jeremy

 

[Pedrohlt]

Bad news .
Yesterday something strange happend. I was adjusting the sensors position, the controller was warm and then the motor stops. I thoght the problem was the motor, but when I opened the controller, I saw that one fet exploded . I was ready to run and have fun, and then it happends :x . I think that the fet burnt others components. So I think the best option is buy/make a 6 fet mod controller and try to fix my 12 fet cntrl to my possible future ebike project.

I asked Keywin if he could make a 6 IRLB3034 FETs controller and he said he doesn't has those fets. So, I have two options: ask Keywin to make me a 6 genuine IRFB4110 FETs controller or buy a 6 fet controller without fets and try to buy 3034s from somewhere (don't have on my country), which is not easy (digikey's shipping is too expansive and others place don't have on stock ), but I don't know if I can do all the job of install the fets on the controller.

I will use this controller with 7S LiFE and a 63-74 outrunner motor. Do you guys think a 6 FET mod cntrl can handle 40-50A continuous? Current limit of 100A? I think the IRLB3034 can handle it and stays cool (Rdson almost 3 times lower than 4110's), but it has slower rates (I don't if it matters). Can 6 IRFB4110 do it?

Thanks all for replying.