[xrotaryguy]

Hello all,

I have not posted here in a while, so I figured it was about time!

My first question is this: If I use a Kelly KDHB controller for regen with a series-wound motor, do I need a 2nd controller for propulsion?

More specifics:

I am working on a hybrid boat project. The boat will give river tours. 3 hours down-stream on electric only, 1 hour back up-stream via a 165 hp Honda motor. The battery pack needs to regen during that 1-hour trip.

I have determined that the battery pack charges at 56 V and 108 A max. I have heard that Kelly controllers are inadequately heat sinked, so I want to double those numbers for the Kelly controller rating or come as close to it as possible. I don't want to replace a controller, EVER!

Motor: Transwarp9.
Pack voltage: 48V (running higher voltage on the water requires extra protection).
Pack Construction: cylindrical 18650 cells.
Max pack discharge: 180 A.
Max pack charge: 288 A

 

[xrotaryguy]

Nobody? I know people have dealt with these controllers before. Maybe I need to post in one of their threads.

 

[TheSGC]

The Kelly Controller REGEN was designed for PM motors, not series wound. So far there are NO reliable controllers capable of REGEN on a series wound motor because there are too many variables to deal with on the motor side. Tesseract has a good thread on series REGEN here and why it's just not practical.

 

[major]

My first question is this: If I use a Kelly KDHB controller for regen with a series-wound motor,

Hi xrot,

Is this possible? I do not see regen capable series motor controllers offered on the Kelly web site.

major

 

[xrotaryguy]

It was my understanding that a KDHB controller could do REGEN if paired with a second controller, like the KDHA? We have an Alltrax controller already. Adding the KDHB is supposed to make REGEN possible.

http://www.kellycontroller.com/dual-kdhb-assembly-p-101.html

Or maybe I need two Kelly controllers. According to this thread, that appears to be the case. Specifically, the text to which I refer is:

Kelly has a series wound REGEN setup using two controllers. For example I have a KDH09401 with REGEN. The setup would use a KDH09400 to drive the field coil and a KDH09401 to drive armature REGEN.

Apparently this setup will work with any of the Kelly controllers that have a REGEN and non-REGEN model of the same series, and Kelly has the diagrams to do it.

This is the response I got:

David,

Right now we provide dual controllers for series motor regen, such as 09400 to drive field coil, and 09401 to drive armature regen.

Single controller isn't able to do regen with series motor.

Thank you,
Steven

I also talked to Kelly about this via Alibaba and email. Here is the response I got:

Hi Jackson,

You may consider Dual KDH/B assembly.
This assembly can do regen braking for series wound motor.
http://www.kellycontroller.com/dual-kdhb-assembly-p-101.html
The price of assmbly don't include the controllers and DC/DC converter.
You need a KDHA controller and a KDHB controller.
Could you please tell us the rated power and voltage of your motor?
We will recommend two specific controllers for you.

Thanks,
Selina
Kelly Controls,LLC

Unfortunately, when I mentioned that I was using a 20 kW motor, she told be that Kelly didn't have a controller that would handle that. I am not sure that the motor size is such an issue though. The HV system is as follows:

Motor: Netgain Warp9
Battery Voltage: 50 V
Continuous Discharge: 288 A
Continuous Charge: 108 A

I wonder if noise coming back from the motor could be the issue.

 

[major]

It was my understanding that a KDHB controller could do REGEN if paired with a second controller,

Hi xrot,

I had not seen those links about a dual set-up for regen with series motors. Not much in the way of detail. So my guess is that they are separately exciting the series machine for generator mode. This could eliminate the classic instability associated with the series generator. However, there could be trouble with the excitation of a series field due to the low duty cycle. The current control loops for both the armature and field controllers would be difficult to say the least. And I wonder if Kelly is up the support task to get it tuned for you.

Besides that you state some fairly high continuous current requirements which make me think Kelly would not be a wise choice especially when you say you want a reliable controller. Also, 288A continuous is beyond the capabilities of the 9 inch motor.

Another thing which is strange about your statements is the use of the 18650 cells for a large energy pack.

The use of a 9 inch motor at 48V is going to limit you to about 10kW continuous operation, in either mode, generator or motor. Do you think this will be effective with a 165hp engine running full bore?

edit: I just saw this and see what your objective is.

3 hours down-stream on electric only, 1 hour back up-stream via a 165 hp Honda motor. The battery pack needs to regen during that 1-hour trip.

Might work

Just a few thoughts from me.

major

 

[xrotaryguy]

Update:

I talked to the famous (or infamous, depending on how happy you are with Kelly service) Steve Li in an email today. He says that Kelly is able to do regen on a series motor with a single controller. He recommended either the HPM72801 or the HPM12901 for my application. However, he also said, at least I think he said, that he would need up to one month to complete the firmware. ruh roh! Not sure what to think about that.


Hi Major,

Good point about Kelly potentially having trouble making this work. I think you're probably right about them separately exciting the field and rotor windings with the two-controller setup.

The 288 A is not a motor requirement. That is simply the max continuous rate for the battery pack. Go go Gadget Li ion batteries! The motor will likely be tooling along at a leisurely rate nearly all the time. However, the motor would need to regen at 108 A (in theory) in order to recharge the battery pack during the one-hour, up-stream trip.

Someone else designed the battery pack last year. I'm stuck with it.
Likewise, someone else selected this warp9 motor last year and I'm kinda stuck with it too. I can change the motor to something better suited to the application, but the mechanical portion of this project is already, more or less, complete. I don't want to re-engineer everything if I can get around it.

We shall see how this all shakes out.

 

[CPLTECH]

I recall that the series motor uses an advanced timing on the brushes that causes fire to shoot out when in regen mode. Hence regen has not been recommended in the past. A better authority than I needs to comment.
Can you hang a separate 48V alternator on ICE to charge? (48V x 108A = 5184W)

 

[major]

I recall that the series motor uses an advanced timing on the brushes that causes fire to shoot out when in regen mode. Hence regen has not been recommended in the past. A better authority than I needs to comment.

Yeah, but running at 48V and under 200A he could set it to neutral. It may be predrilled and just need the CEH rotated.

major

 

[xrotaryguy]

I don't know if the transwarp9 motor's brushes can be re-clocked to neutral or not. I know that some Netgain motors have this feature. I can't find any information on this for the transwarp9 though. And yes, I have heard that advanced brush timing can lead to the "fireball effect."

Just talked to Steven Li again. He confirmed that he could sell me the controller and then have me reflash the micro so that it would do regen at a later date (when the firmware has actually been written). Of course, if Kelly's development team discovers that regen isn't going to work so well after all... I don't know what happens at that point...

Brian

 

[major]

IJust talked to Steven Li again. He confirmed that he could sell me the controller and then have me reflash the micro so that it would do regen at a later date (when the firmware has actually been written). Of course, if Kelly's development team discovers that regen isn't going to work so well after all... I don't know what happens at that point...

Hi xrot,

Yeah, I'd be leery of paying for a product up front with the means to make it work coming later, after they figure out how to do it. And I suspect that additional hardware is required as well as the firmware.

You might have noticed, I have a peculiar interest in unusual EVs and/or HEVs. So I have been thinking about your application. And you would be better served using a "true" separately excited machine. You want reliability and several vendors offer SepEx controllers for 48V and 4 to 500A which have had years of successful field experience in golf carts, forklifts, etc. I have used both Sevcon and Curtis SepEx controllers with good luck. These are quality robust controllers.

Coming from a motor background, it probably seems too easy to me. But I see it as no large or overly expensive task to rewind the series 9 inch motor to a 1 or 2 ohm field such that it would be compatible with those controllers. A resourceful builder like Jim Husted may be able to swap out the field coils with an existing suitable set. But I'd first check with the manufacturer. He may already have such a design or if not, may be interested in doing it for future product.

The application you describe seems well suited for a SepEx motor, and the SepEx generator is a much better bet than trying to hack a series machine. Also, the SepEx will give you contactorless reversing, set speeds and maybe some other desirable control features.

Something to consider

major

 

[xrotaryguy]

Interesting idea. I absolutely agree with you about series being a messy way go go. Converting the motor that has already been mechanically designed into the chassis might be the way to go. Besides, I am having a hard time finding sepex motors that are big enough. I want to stay in the 20 kW region. The boat will weigh 6000 lbs loaded, so it needs a pretty big motor. I live near Dennis Berube too, so maybe he can do a rework on this thing.