Register    Login    Forum    Search    FAQ     Formula Hybrid Home Page

Board index » Technical - Design and Construction » Electrical




Post new topic Reply to topic  [ 28 posts ]  Go to page 1, 2  Next
Author Message
 Post subject: Motor Controllers
 Post Posted: Sun Jul 05, 2009 11:07 pm 
Offline

Joined: Mon Jun 01, 2009 3:30 pm
Posts: 17
Location: Element One: Lawrence Tech University
Does anyone have any experience with Kelly motor controllers. I have read a lot of bad press about them, especially their models rated for 144V+ and when used for regenerative braking.

ANY COMMENTS?


Top 
 Profile  
 
 Post subject:
 Post Posted: Tue Jul 07, 2009 1:34 pm 
Offline

Joined: Tue Oct 30, 2007 6:50 pm
Posts: 78
Location: Illinois Tech
Last year we used one on our car. We were running a 72V bus so I cannot comment on their 144V+ models but even the model we had was not all that good. We managed to burn it while doing regen and got another one which functioned fine for a while until the ends started popping off. In general I think their enclosures are poorly build. The ends are made out of plastic and are glued to the aluminum housing, or just pressed in. (One end had glue marks on it the other did not.) As far as the hardware inside the enclosure, well I can't really comment on the quality of the components since they are all covered with some sort of black paint, (to deter someone from copying/fixing the circuit?) but the setup used is not exactly professional. They use multiple mosfets in parallel in order to get the amp rating they want and multiple electrolytic capacitors to get the capacitance they need to run properly. The problem with that setup is that not all mosfets are created equal, manufacturing variations will cause different resistances, turn on and off times which can cause some of them to get hotter than the others and if one of them fails the rest will follow soon after.

However, in May I saw that Dartmouth was using them and they told me they did not have any problems with them. They even pointed out that the Kelly controllers will do torque control, which is a nice feature and one that most other controllers (if not all of them) do not have. Also to their defense I have not opened up the controllers that we used this past May to see what kind of setup they use, but then again I didn't have to because they did not burn.


Top 
 Profile  
 
 Post subject:
 Post Posted: Tue Jul 07, 2009 7:10 pm 
Offline

Joined: Mon Jun 01, 2009 3:30 pm
Posts: 17
Location: Element One: Lawrence Tech University
Couple Questions:

1. Did you have a heatsink mounted to the bottom of the controller.
2. You managed to burn one during regen. Did you try it again with the new one after?
3. We have always run CAN BUS on our vehicles, did you try using the CAN BUS option on the controller (it says it has it).
4. What is the functionality of the controller for changing parameters? In other words, what sort of parameters can I change?
5. When you burnt the controller, did Steven replace it for free or charge?

If you plan on a 72V BUS again, I would suggest Curtis as they are quite functional. However, I must admit I have only used them on AC induction motors and have no expericence with their DC controllers.

Thanks for the responses


Top 
 Profile  
 
 Post subject:
 Post Posted: Wed Jul 08, 2009 12:23 am 
Offline

Joined: Tue Oct 30, 2007 6:50 pm
Posts: 78
Location: Illinois Tech
1. We did have a heatsink mounted on the bottom but it hardly ever had the chance to get hot.
2. Yes we did. In fact it was running constantly in regen during the 2008 endurance. We still don't know what exactly went wrong with the first one because even though we opened up the controller the black tar-like substance that covers everything prevented us from seeing what exactly burnt and why. But my team tends to loose faith in anything that breaks without me touching it. :twisted:
3. If I remember correctly we needed to have a CAN to RS-232 converter in order to actually program it. But we never used it to send/receive data that was used for the control of the motor.
4. I was not directly involved with the fine tuning so I don't know exactly what kind of parameters you can adjust but I have looked at their website and found this
http://www.kellycontroller.com/mot/PMhelp.html
5. I have no idea. I will ask around and let you know.

Unfortunately, Curtis does not have a 72V PM/Regen enable controller. We ended up using a Sigma drive from PG Drives. Other than the lack of torque control they are very good drives. We bought five and they all work like a charm.


Top 
 Profile  
 
 Post subject:
 Post Posted: Wed Jul 08, 2009 8:04 am 
Offline

Joined: Mon Jun 01, 2009 3:30 pm
Posts: 17
Location: Element One: Lawrence Tech University
hmmm I wonder if you had a precharge resistor across the contactor? Also, perhaps the controllers have too small of capacitors inside. Voltage spikes from driving the motor and then suddenly releasing the power can cause the voltage to spike and possibly burn the MOSFET


Top 
 Profile  
 
 Post subject:
 Post Posted: Wed Jul 08, 2009 2:16 pm 
Offline

Joined: Tue Oct 30, 2007 6:50 pm
Posts: 78
Location: Illinois Tech
No we did not, but the precharge resistor is only supposed to be there to avoid welded contacts on the contactor and arcing. The inrush current should not harm the controller, after all it is the capacitors in the controller that draw the current not the fets. Well if the DC link capacitors are not rated properly then that would still be a design flaw. I will try and get some pictures of the internals of our burned controller and post it here.


Top 
 Profile  
 
 Post subject: Inrush
 Post Posted: Thu Jul 09, 2009 6:07 pm 
Offline

Joined: Mon Jun 05, 2006 4:05 pm
Posts: 112
Location: Rules Committee
It's possible for inrush current charging a cap to cause ringing between the cap and the stray inductance of the wiring leading to the cap, resulting in a peak voltage up to double the battery voltage. That peak voltage would be more likely to kill a MOSFET than damage a cap.


Top 
 Profile  
 
 Post subject:
 Post Posted: Tue Jul 14, 2009 8:49 pm 
Offline

Joined: Tue Oct 30, 2007 6:50 pm
Posts: 78
Location: Illinois Tech
Here are some pictures of the internals of our kelly controller.

Image
Image

You can clearly see the rows of FETs in parallel and the little capacitor bank for the DC link. the black paint is also visible. What is not visible is what is burned but we never actually got that far because of the paint. It was easier to buy a new one than spend the time to debug it, plus they were very cheap.

Now as far as the ringing between the stray inductance in the wiring and the capacitor goes I have to agree that it is possible but I would say it is highly unlikely given the small values of the inductance and the wire resistance compared to the capacitance. But even if the value is high I don't think that the FETs should be effected since the other side of the FET is not connected to the ground but rather to another FET and then to the ground. In addition, the capacitors that form the DC link are rated at 100V and none of them seems to have any damage so the voltage must have never increased to higher than 100V for any significant amount of time, if it did at least one of the capacitors should have some damage on it.

Now you might argue that the FETs don't have to be rated at 100V or higher since the controller was only rated at 72V and that it is bad practice on our behalf to forgo the use of a precharge resistor especially since it is included in the wiring diagram of the controller. And you are right, we do have some bad habits that are hard to break, but since we have the option of using a controller that can take some extra abuse why use the one that can't?


Top 
 Profile  
 
 Post subject:
 Post Posted: Wed Jul 15, 2009 10:26 am 
Offline

Joined: Mon Jun 05, 2006 4:05 pm
Posts: 112
Location: Rules Committee
Quote:
as far as the ringing between the stray inductance in the wiring and the capacitor goes I have to agree that it is possible but I would say it is highly unlikely


Also, looking back at the story of the failure, it sounds like it failed during regen operation, rather than failing at startup.

Thanks for the photos of the inside. I wouldn't be too critical of the parallel MOSFET design. Because MOSFETs have positive temperature coefficients, steady-state current and temperature imbalances tend to be self correcting. It is true, however, that timing of the gate signals becomes critical.


Top 
 Profile  
 
 Post subject:
 Post Posted: Wed Jul 15, 2009 8:56 pm 
Offline

Joined: Tue Oct 30, 2007 6:50 pm
Posts: 78
Location: Illinois Tech
Well your original comment got me thinking and I realized I don't really understand what kind of voltage difference, if any, the FETs see in the situation you described. In the case you have just two individual devices on different dies like in the kelly they might not be effected by the ringing but what about the case when the devices are on the same die? It is definitely something my team needs to look into since we would like to construct our own controller in the future. Although we should probably have some sort of precharging scheme build in if we do.

But, Charlie you guys have used the Kelly controllers, what is your opinion of them?


Top 
 Profile  
 
 Post subject:
 Post Posted: Mon Jul 20, 2009 4:21 pm 
Offline

Joined: Mon Jun 05, 2006 4:05 pm
Posts: 112
Location: Rules Committee
I'm just an advisor, not a team member, so I don't have much direct experience with the controllers. I think there was some discussion that the bandwidth of the torque control mode wasn't as high as they would like, but it's hard to complain about that too much when the other options don't even have torque control.

Charlie


Top 
 Profile  
 
 Post subject: Regeneration for a PMG132 motor
 Post Posted: Wed Oct 07, 2009 11:20 am 
Offline

Joined: Tue Oct 06, 2009 4:00 pm
Posts: 38
Location: Florida State University
Antoant,

I guess you are using Kelly/sigma motor controller for a PMG132 DC motor, right?
we are also considering using PMG132 motor, the question is: when we buy a motor from
the supplier, do they provide/suggest motor drive/controller at same time? or we have to
buy motor controller separately from other suppliers?

another question is the regeneration, you mentioned kelly and Sigma controller can realize
regeneration function during braking, do they have a active clamp/dynamic brake connected
to DC bus?


Top 
 Profile  
 
 Post subject:
 Post Posted: Wed Oct 07, 2009 3:15 pm 
Offline

Joined: Tue Oct 30, 2007 6:50 pm
Posts: 78
Location: Illinois Tech
It depends on the supplier. Most companies/websites that carry these motors also have controllers in their inventory so you could order one from them. But it is usually up to you to decide if the controllers they offer have all the features that you want. If they will be able to help you pick one depends on how knowledgeable they are. Also keep in mind that the controllers that you can buy for these motors are not limited to the ones that are offered by the companies that you are buying the motor from. For example our sigma controller was bought by the company that makes them and as far as I know that is the only place you can get them from, apart from getting a motor/controller package from the motor manufacturer.

If I understand your question, about the regenerative braking function, correctly then you are looking to find out how each controller implements regen which is something I really don't know. I can tell you how I would do it but I don't know if that would be of any help. Your best bet is to contact the manufacturers of each drive and ask them.


Top 
 Profile  
 
 Post subject: Regen: active clamp or resistor
 Post Posted: Wed Oct 07, 2009 5:09 pm 
Offline

Joined: Mon Jun 05, 2006 4:05 pm
Posts: 112
Location: Rules Committee
I think I understand the question about regen and clamps. I haven't seen a controller in this class that does that. The idea would be to have a way to dump the energy to avoid overcharging the battery if it came to that. I think that what they do is just stop regenerating if the batteries are full. That means you are relying on your mechanical brakes in that case. With strong regen, that could be a bad thing, if the brakes suddenly cut out when the battery got full. So it would be nice to smoothly transition...or to have a resistor or clamp to dump the energy into.

But I agree that it's a better question for the controller manufacturers--I know more about the technology than about what's commercially available, and they may be adding that kind of feature as we speak.


Top 
 Profile  
 
 Post subject:
 Post Posted: Wed Oct 07, 2009 6:01 pm 
Offline

Joined: Tue Oct 30, 2007 6:50 pm
Posts: 78
Location: Illinois Tech
That is very interesting. Another thing added to the list of things to test once our car is done.


Top 
 Profile  
 
Display posts from previous:  Sort by  
 
Post new topic Reply to topic  [ 28 posts ]  Go to page 1, 2  Next

Board index » Technical - Design and Construction » Electrical


Who is online

Users browsing this forum: No registered users and 2 guests

 
 

 
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
cron