96v controler?

[RatoN]

Thinking about going with 2 x 48v in serie = 92v (124v full) on a 5304 hub.

Can someone explain to me how to modify either a 36v or a 72v controler to 124v?

To make it even more funner :? and more intersting for you all , I know nothing about electric or electronics! wee!

Any help will be apreciated.

 

[xyster]

Ooh yah, I'd like to know the answer too....

The x5's are supposedly good for 5,000 watts continuous, and 120volts or so max. At that power in a bicycle frame, torque arms would be wise even on rear wheel models.

At full charge, 96v of nicad or nimh batteries would be 112 volts, and 96v (26s) of lithium-ion or lithium polymer would be 109.2 volts. and 96v of lead-acid batteries would be too damn heavy!

The stock 72v crystalyte controllers are reported good for no more than 100 volts.

 

[xyster]

To make it even more funner and more intersting for you all , I know nothing about electric or electronics! wee!

Yah, wee all over the floor if you zap yourself! The current at that voltage and with those batteries is potentially lethal. Might want to study up, or consider paying one of the experienced members who has offered, like Bob McCree or Fechter, to do the mod if they would.

 

[fechter]

That's a crapload of batteries. Don't touch the wires either. It will kill you.

The Crystallyte 72v controllers have 100v rated FETs and caps. The next step is like 150v, but you'd might have to sacrafice some current rating.
The low voltage regulator transistor is rated for 400v, so not much other than the FETs and main caps would need to be changed.

Changing the parts is really more of a mechanical job than electronic. It would require good soldering skills though.

I can double check the circuit and see if anything else would need to be upgraded.

 

[RatoN]

Thanks Fechter ! What kind of FETs and caps should i buy? Could you post a pic pointing out the components I should replace. Btw, my controller is a 36v, but I could buy a 72v if it makes things easier.

And yes Xyster , I might just pay somebody to do it. I wish that I could just buy a prefab one instead of messing with a perfectly good controller.

Now, I really don’t think that it’s THAT much batteries or power guys, really. Wait t'ill i make a e-mini van My only worry is the spokes on the back wheel of the bike. Then again, all depends on the rider.

 

[xyster]

Now, I really don’t think that it’s THAT much batteries or power guys, really. Wait t'ill i make a e-mini van

Because we care

http://www.allaboutcircuits.com/vol_1/chpt_3/4.html
In industry, 30 volts is generally considered to be a conservative threshold value for dangerous voltage. The cautious person should regard any voltage above 30 volts as threatening, not relying on normal body resistance for protection against shock. That being said, it is still an excellent idea to keep one's hands clean and dry, and remove all metal jewelry when working around electricity. Even around lower voltages, metal jewelry can present a hazard by conducting enough current to burn the skin if brought into contact between two points in a circuit. Metal rings, especially, have been the cause of more than a few burnt fingers by bridging between points in a low-voltage, high-current circuit.
Also, voltages lower than 30 can be dangerous if they are enough to induce an unpleasant sensation, which may cause you to jerk and accidently come into contact across a higher voltage or some other hazard. I recall once working on a automobile on a hot summer day. I was wearing shorts, my bare leg contacting the chrome bumper of the vehicle as I tightened battery connections. When I touched my metal wrench to the positive (ungrounded) side of the 12 volt battery, I could feel a tingling sensation at the point where my leg was touching the bumper. The combination of firm contact with metal and my sweaty skin made it possible to feel a shock with only 12 volts of electrical potential.

• REVIEW:
• Harm to the body is a function of the amount of shock current. Higher voltage allows for the production of higher, more dangerous currents. Resistance opposes current, making high resistance a good protective measure against shock.
• Any voltage above 30 is generally considered to be capable of delivering dangerous shock currents.
• Metal jewelry is definitely bad to wear when working around electric circuits. Rings, watchbands, necklaces, bracelets, and other such adornments provide excellent electrical contact with your body, and can conduct current themselves enough to produce skin burns, even with low voltages.
• Low voltages can still be dangerous even if they're too low to directly cause shock injury. They may be enough to startle the victim, causing them to jerk back and contact something more dangerous in the near vicinity.
• When necessary to work on a "live" circuit, it is best to perform the work with one hand so as to prevent a deadly hand-to-hand (through the chest) shock current path.

 

[knoxie]

Hello

Yes you must be careful if you run 100V or more, at 72V I am very careful as the controller when switched off holds a good charge for quite a while in those caps.

When working on the controller the first thing I do at any time is short the battery wires together else you can get one hell of a belt off it.

at 100V DC you will feel it and it could kill you so please be very careful.

Cheers

Knoxie

 

[Mathurin]

I say go for it, dude.

I mean, your bike's allready illegal anyways, so...

Are you thinking of adding a 48/96v switch?

 

[fechter]

You could use power from the battery as part of a theft deterrent system.
Bzzzzzzap!
Still checking on the parts.

 

[fechter]

On DigiKey, there's one called a FDP2532 that's rated for 150v, 79A, .014 ohms.
$3.83 ea.

The controller takes 12 of them.
The current and resistance rating mean you would have to stick to the stock 35 amp limit (ok, maybe 40 amps). You could always make the current limit lower.


You would also need new main capacitors.

It looks like the rest of the parts are made to take at least 150v. I could not positively read the number off the bootstrap diodes without unsoldering one. It starts with 1N4 and ends with 7. I'm guessing it's a 1N4007, which is rated for 1000v.

There's a small resistor that determines the low voltage cutout. That would need to be changed. You could also disable the low voltage cutout.

When I get time, I'll post pics of what you would need to change.

Know anybody with a fried controller? Needs new FETs anyway.

Note: this is only for the Crystallyte "black box" controller type. Other controllers will be similar, but I don't have detailed information.
As far as I know, the only difference between a 36v, 48v, or 72v version is the resistor that sets the low voltage cutout. This is easy to change.

 

[Mathurin]

Or, change the FET's & drive side caps, then rig the low voltage part to run off one of the 48v packs, and let the buisness end run off full juice. Wouldn't need to alter the rest of the controller that way.

That's the way my 36v non-clyte controller seems to be designed, there's two positive cables going into the controller's PCB. One with a fat wire to the FET side and the other with a thin wire, apparently to the low voltage part. Case says something about 36/48v, and low voltage regulator has a 42v max...

I know Velectris is using a similar strategy for the new controller they're developping since they're quite open about it on their forums, right down to their FET's part numbers. No schematics or gerbers though. Three power leads, normal version has normal voltage on both positive, so the 10% of their clients who want unreasonably high voltage can buy an external regulator for the low power side.

 

[Ypedal]

How fast do you really need to go ?.. wowzers..

a 5304 in a 26" wheel at 90v operating voltage should be going something insane like 60 + mph !!!.....

If you really want to break the sound barrier.. just get a different motor.. a 5302 at 72v in a 24" wheel.. 40 + amps would be quite the getup...

You are gonna need ALOT of batteries ! lol... think mucho AH..

 

[Mathurin]

Dude, on the surface of the Earth, the escape velocity is only about 11.2 kilometres per second. 8)

 

[xyster]

How fast do you really need to go ?.. wowzers

At 100v and 40amps, and a realistic 80% efficiency, wind resistance will limit top speed to about 48mph:
http://www.kreuzotter.de/english/espeed.htm

So it's more about efficiency, acceleration, and hill-climbing power in an ungeared motor.

Of the two ways to increase those, higher voltage has a couple distinct advantages we've discussed quite a bit already in other threads.

I definitely would not go with a lower number than the 5304 in a 26" wheel at these voltages, else you'll lose quite a bit of torque (and thrust) for no usable gain in top speed:
http://www.ebikes.ca/simulator/

I run my 5304 in a 24" wheel at 80 volts and 35 amps, at which power it can climb very steep 15% hills still at only 20mph, and accelerate up to it's 44mph top speed at about the pace of a Yugo. It feels like a light, modestly powered ICE motorbike.
I could definitely go for >100 volts, 40-50 amps, relacing the motor into a 20" wheel, and beefing-up the rear dropouts.

 

[fechter]

That's the way my 36v non-clyte controller seems to be designed, there's two positive cables going into the controller's PCB. One with a fat wire to the FET side and the other with a thin wire, apparently to the low voltage part. Case says something about 36/48v, and low voltage regulator has a 42v max...

Right, that's what you need to do with most controllers. The low voltage regulator transistor on the Crystallyte controller is rated for 400v and bolted to the main heatsink, so I think it would work fine without rewiring.

 

[RatoN]

Ypedal, it’s exactly what Xter said. Torque/Distance/Speed, potential.
Thanks for your concern about safety all.

Math, I have no intentions of going airborne… with this vehicle. In the pursuit of logical transportation though, one would come to the conclusion that the next step is to get rid of the wheels, but that is another story.

Till then,

menfin?

 

[tno]

you can probably run 96v sla or lithiums with the 72v controller.

Todd Allen did this with his 408 and it ran for a month until the motor burned out.

http://tech.groups.yahoo.com/group/power-assist/message/29164

 

[Mathurin]

Faut pas confondre manque de pot et pénurie de récipiant. - Gaston Lagaffe

 

[xyster]

you can probably run 96v sla or lithiums with the 72v controller.

Todd Allen did this with his 408 and it ran for a month until the motor burned out.

From ~90V - 106+V , the crystalyte controllers with some 100V components seem to burn out at a rather wide range of voltages.

 

[Ypedal]

Faut pas confondre manque de pot et pénurie de récipiant. - Gaston Lagaffe

Faut pas perdre les pneus at un moment inconvenient - Gaston Daigle 8)

 

[fechter]

you can probably run 96v sla or lithiums with the 72v controller.

Todd Allen did this with his 408 and it ran for a month until the motor burned out.

http://tech.groups.yahoo.com/group/power-assist/message/29164

If the motor burned out, it would be from too much current. That could be limited by the controller. I think that's the first report I've heard of a Crystallyte motor burning up. Time for forced-air cooling :wink:

 

[Ypedal]

After taking a few apart, the thermal transfer of the crystalyte motors could be improved quite a bit by simply opening them up and cleaning the sand casting on the inside of the motor covers... even better would be aluminum side covers ! ( would also shed a few lbs ) .

Keeping it sealed up is more of a weather proofing thing.. i like this aspect.

 

[RatoN]

Request for a 96v controler here

 

[RatoN]

Warning : Not prutty.

 

[fechter]

Oh, you should have made the "not purdy" warning a bit larger.

The board needs to be parallel to the heatsink. The legs on the FETs were supposed to be bent the same way as the original ones to fit properly.

I found it very helpful to mount all the FETs on the heatsink before soldering to make sure the alignment would be straight.

You fried some traces on a couple of them. This is pretty easy to do, I've done it also. The fix is to use a very small wire to bridge from the FET leg to the next component along the trace (the gate resistor). The spacing is small, and it's easy to accidently bridge between traces.

Before powering up, check for continuity between each FET heatsink tab and a screw going into a heatsink. There should be no connections.

At this point, it's hard to say what the best approach is. You don't want to bend the FET legs any more, since they will break off easily. You need to repair the lifted traces with small wires as mentioned above. If you don't have any fine wire, you can take a piece of stranded wire and pull out a single strand and use that.

You should also unscrew the heatsink and solder the legs on the other side of the board where they go to traces.

Sorry we didn't get a good 'how to' thread posted before you tried this.