Ecrazyman controller with Kollmorgen?

I think so, though I can't remember where. Electrically, it's your run-of-the-mill PMAC motor, so that controller should work fine. You'll have to do a little work matching up the halls, though.
 
Haven't heard of anyone who has tried that. You might be the first. I think it should be no problem up to 48v. Motor flies apart somewhere around 60v.

The color code shown below should get it going one way or the other. To reverse rotation, it will take switching some of the wires. I'm not sure about the wiring, so test carefully and watch the current.
Kollmorgen hall and phase wires.jpg
 
I had one of my Kollmorgen's open last week to reverse the rotation, which worked like a charm btw, just swap those A & C wires pictured for both the hall and phase wires. I already have an ECrazy controller that I only need to modify for an LVC to match my 36V lifepo4 Ping packs. I'll report back with results hopefully this week. My only issue is time, but I should be able to fit it in between epoxy coats on my electric surfboard project that has to be finished and in the water for testing by Thursday. Temps have been so cool that the epoxy is taking forever to cure, so there's no reason the Kollmorgen conversion shouldn't get done. Everything is already in place, motor mount, sprockets, batt rack, throttle.

I wish the controller wiring wasn't trial and error. On the motor it's the 1-2-3 firing order is pretty obvious, but I haven't found the same for the controller (only different color code combinations to match different motors. This seems to be a pervasive problem with 3 phase motors, and I strongly suggest that we here initiate a standard convention. Instead of a standard color coding convention, which isn't even used 100% on 2 wire brushed motor (red and black typical but not 100%), I think hall and phase wires should be labeled 1-2-3. Then we can start a database of known motors and controllers with their 1-2-3 sequences. Face it, electric motors are the future of transportation, and just imagine the confusion that would exist if V8 engines had color codes for their cylinder firing sequence instead of numbers. Let this forum be the start of keeping things simple for 3 phase motor wirings, and eliminate the 6 combination guesswork.

John
 
A little OT, but as long as the pic is up.. could I drill a couple of big holes in that yellow fiber plate there for cooling? Or is that a bad idea? Probably a good idea for 48v, to keep the magnet glue from overheating..

I'm planning on trying this same controller with a Koll at 48.. haven't had a chance to try yet.
 
Hey guys, I'm doing exactly the same thing (36V pack but a 48v ecrazyman with a Kollmorgen). I had it running at 48V on the bench a few weeks ago but I just got it on the bike and set up for 36v last night. Not working so great. I posted this in the E-bike Technical section:

Hi folks, having a bit of a problem. My controller "stutters" at low speed, in fact it's hard to get started at all from a dead stop. I suspect there must be something up with the hall sensor wiring, unless this is normal. The stock controller was super smooth. Jonathan loaned me a 36v Makita pack and I reconfigured my 48v Shenzhen controller for 36v operation. I placed a 3.9k resistor in parallel with R3 and jumpered R1. I then hooked it up to my Kollmorgen as follows:

Motor->Controller
Power
Green -> Blue
Blue -> Green
Hall
Brown->Black
Blue->Green
Yellow->Yellow
Green->Blue

However I believe I rewired the Kollmorgen hall wires in the past so the color coding may not be too useful for others out there with Kollmorgens. When I get this sorted out I'll post a picture of the configuration that "works." FYI swapping green and blue reversed the direction of the motor.

I took the bike for a ride this morning; with the wheel unloaded it spins up nicely, but with any weight on the bike it sounds like a fan with a bad bearing. The bike is also pretty slow, though it does run.

I also have not had any luck finding connectors for this controller, so I used a small screw driver and removed the blades from their housing and used spade connectors and duct tape to hold it together. I could not remove the blades from the hall sensor connector so I just stuck bare wires in and taped the hell out of it. Sounds like I may be better off chopping the connector and soldering the wires together.

Any thoughts?
 
Yes, I think you could drill some holes in the fiberglass for cooling. The plate just lines up the hall sensors, so as long as they stay in the correct position everything should be fine.


John in CR said:
I wish the controller wiring wasn't trial and error. On the motor it's the 1-2-3 firing order is pretty obvious, but I haven't found the same for the controller (only different color code combinations to match different motors. This seems to be a pervasive problem with 3 phase motors, and I strongly suggest that we here initiate a standard convention. Instead of a standard color coding convention, which isn't even used 100% on 2 wire brushed motor (red and black typical but not 100%), I think hall and phase wires should be labeled 1-2-3. Then we can start a database of known motors and controllers with their 1-2-3 sequences.
John

Yes, wouldn't that be nice. Not only do the colors not match, but often the motors and controllers are marked A-B-C for the phases and halls, but this is not consistent between manufacturers. Even for the same manufacturer, it's not consistent between models.

I suggest we adopt the Crystalyte controller / motor color and phase designations as the "standard", since this is one of the most common.

A database of motor and controller wiring would be great too. Something for the Basics section?
 
If that motor works with the controller, without the same problem people are having with their BMC motors, please let me know. I am still not sure what I really want to do now that I need to get either a new motor, or controller for my setup.
 
No luck with this tonight. I spent about 3 hours and can't figure out whats wrong. I tried two Kollmorgens and many wiring configurations and always the same result: the motor seems to work on the bench, but under load it stalls and stutters and has anemic performance.

I wonder whether there is some additional modification that needs to be done to use the 48v controller at 36v? I never tested it at 48v besides on the bench so I don't know whether that works either. But certainly something is wrong right now, as the stock 24v Kollmorgen controller with old worn out SLAs performed better.
 
fitek said:
No luck with this tonight. I spent about 3 hours and can't figure out whats wrong. I tried two Kollmorgens and many wiring configurations and always the same result: the motor seems to work on the bench, but under load it stalls and stutters and has anemic performance.

I wonder whether there is some additional modification that needs to be done to use the 48v controller at 36v? I never tested it at 48v besides on the bench so I don't know whether that works either. But certainly something is wrong right now, as the stock 24v Kollmorgen controller with old worn out SLAs performed better.
Well it could be that with voltage sag, your going under the LVC.

You could try and bypass it. Just read the sticky in the tech section, it has information on doing it.
 
fitek said:
I wonder whether there is some additional modification that needs to be done to use the 48v controller at 36v? I never tested it at 48v besides on the bench so I don't know whether that works either. But certainly something is wrong right now, as the stock 24v Kollmorgen controller with old worn out SLAs performed better.

Did you jumper one of the voltage regulator resistors?
You need to do this for 36v in addition to changing the LVC.
 
Fechter, I jumpered R1, which if I understood right is the big fat resistor nearest the bottom left corner of the PCB (opposite side of the FETs, further from the wires hanging out of the metal box).

R3 has a 3.9k resistor in parallel.

I did not check if there is voltage sag. Should I except this from 36v 6ah Konion cells?

This reminds me of my own controller build where I didn't get the switching voltage high enough for the MOSFETs. The motor shaft would turn fine unloaded, but then using my fingers I could get it to stall. This isn't quite that bad as I can't stop it with my fingers, but even the weight of the unloaded bike (45lbs) makes it crap out.
 
OK, you jumpered the right resistor. It should be OK.

Yes, see if you can check for voltage sag. You could also try putting a lead acid battery in series to try a higher voltage.
 
Ok, so I swapped Kollmorgens, no go. I tried a few more combos, then added an SLA in series, then reverted the controller to stock, and still no go. So I hooked up a 25A Crystalyte controller Jonathan loaned me, and it behaved exactly the same, with the same wiring.

So then I figured my wiring must be wrong.

So I redid it using the image Fechter posted, and it didn't even spin, it just went back and forth. I looked closer and all my Koll's wires are different than the ones in the photo.

Here's how I got the Crystalyte to work:
At this point I got fed up with the color coding of the wires and decided to think in terms of firing orders. I can see the firing order of the Kollmorgen power wires (A->B->C for how I have it mounted). Then I guessed the firing order of the Crystalyte hall wires based on their order in the connector. So I hooked up the hall wires to the Kollmorgen in that order, disregarding the color coding of the Koll. Then I hooked up the power wires also in the same order. It didn't work 100%, but the wheel DID try to spin in the correct direction, just as it had in my tests with the Ecrazyman controller. We're talking about sensor wires and power wires arranged around a circular path, so the only two things you need to do is: 1) attach the power and hall sensors to fire in the same direction and then 2) advance either the power or hall wires until they are aligned with each other. So I took the power wire that was attached to A and connected it to B; the one attached to B I switched to C; and the one attached to C I attached to A. Once again it worked only so-so. So I did it one more time, and voila, it ran smooth.

I rode around the block and it doesn't accelerate as fast at the bottom end as with my massive SLAs but it goes faster and I can lift the bike off the ground.

My buddy Victor dropped by and helped me adjust the chain so that it doesn't fall off while pedaling, so I was pretty jazzed :)

I believe the ecrazyman controller will work just fine then.

I would post the final wiring, but I left it on the kitchen table at home, so no diagram for you (until tonight anyway).
 
Jonathan said:
fitek, so was the root cause of the problem hall sensor wire matching?

Yes, I think it was. I will hook up the ecrazyman controller tonight (if I have the chance, doing some beer brewing) to confirm. If the hall and power wires aren't aligned the motor spins but not well under load.

Next week if I can work out the chain skipping issue I will try commuting to work with this combo.
 
Yes, I think you're on it. The wiring of brushless motors has no consistency whatsoever.

Hook up the ecrazyman the same way you had the Crystalyte and it should run backward (depending on where the reverse switch was on the Xlyte).
 
I think you're right Fechter-- when I initially hooked up the Crystalyte after the Ecrazyman, it stuttered around in the opposite direction. Easy enough to fix... just swap two power and two hall wires, trouble is figuring out which wires "go together!"

One note, the image you pasted has the power wires on the Koll going from left to right C B A and then the halls going from top to bottom A B C + -. When I wrote in my notebook a couple months back I had the hall wires as C B A + -. I think this may be the correct notation, but I'll comment on that when I retrieve my notebook from home.
 
Yes, please post the configuration that works.
It's possible not all Kollmorgens are the same also.
 
Here is the wiring for the Crystalyte controller, using the image you posted as a reference. Too busy packing for a camping trip last night to try the ecrazyman.
[pre]Kollmorgen Crystalyte

Power
A Green
B Yellow
C Blue

Hall
A Yellow
B Green
C Blue[/pre]
 
I'm a bit confused. My plan was to get the 36 volt controller from ecrazyman but if you can use the 48 volt controller at 36v maybe I'll go ahead with that one (never know when you might want to go up a few volts). But... My thought was that the under voltage protection on the 48v controller would stop you from using it at 36 volts.

Also... Does the ecrazyman controller use a pot or hall throttle?
 
EVTodd said:
I'm a bit confused. My plan was to get the 36 volt controller from ecrazyman but if you can use the 48 volt controller at 36v maybe I'll go ahead with that one (never know when you might want to go up a few volts). But... My thought was that the under voltage protection on the 48v controller would stop you from using it at 36 volts.

Also... Does the ecrazyman controller use a pot or hall throttle?
You can easily bypass the low voltage control.

It uses a hall throttle.
 
Well I soldered & cursed away another night on hooking the Koll up to the ecrazyman. I can now make it spin CCW but not CW like I need. My head is spinning in color coding so I just tried combos almost at random in vain. Unfortunately that's probably all my ebike time for this week, so I won't be riding to work for at least another week yet again... grr. At this point I wish I had a DC hub motor lying around or that I had just bought a DC motor a couple weeks ago as the cost in gas from commuting via car rather than bike is quickly approaching parity while I tool around with this stupid thing.

Anyway, griping aside, here is the CCW rotation:

[pre]Power
A Yellow
B Blue
C Green
Hall
A Yellow
B Green
C Blue[/pre]
 
Back
Top