HDC2450S FET Bank Temperature Differential

Hello,
Upon reviewing some logs from temperatures 1 and 2 from our HDC2450S, we're seeing a big temperature differential between two the two different banks. The data shown in the graph below are from ?T queries. They are t1 and t2, as noted in the manual. (This is the file named 'bad_graph.png').





The second graph is for comparison -- it is the same product under the same load. ('good_graph.png').






Is it possible we have a fault internal to the controller? This temperature differential has caused the 'bad' controller to trip the 'overtemperature' flag on the controller during runs.

The pictures should be uploaded in a zipped folder; the picture links didn't show up.

The T1 and T2 temps will vary. One will always be higher than the other by a few degrees. The controller goes into thermal limiting whenever the temperature on either T1 or T2 goes above 70 degrees C. This is when the overtemperature led fault comes on.

From testing conditions, it takes a rather long time and very high current to get to this point in a given amount of time.

How much current are you currently drawing from the motors, and for how long? How long does it take to reach this overtemperature condition?

Sorry, graphs should be attached. We're pushing some serious current through these -- well over 200 A for 15 seconds or so.

I'm fine with a couple of degrees C difference. But this is more than that.

That definitely does look bad.

2 things to verify. What is the current firmware? Go to console and in the out data box send this command: ?fid. The date code is what I am looking for. The most recent one is 10/10/13.

The graphed data, is this from the same controller, or is there another controller being referenced that is either the same or a dual channel?

I can tell you it's not going to be 10/10/13, since they've been installed longer than that. They're difficult to get to, so it's difficult to get that information for you.

Each graph is specific to one controller. We're using the HDC2450S, so they return two temperature values-- one for each 'bank', even though they're paralleled.

So, for each graph, what you're seeing is the two temperatures for the same controller.

Ok, this is what the firmware says. Are there known bugs with the firmware? Is it possible that the cooler bridge is somehow broken?

FID=Roboteq v1.2 RCB500 05/25/2013

If the thermal sensing circuit has broken down, more likely the numbers reported back would result in any one of the following conditions.
The temperature for one side would be stuck, and overheat may be on.
The temperature would be stuck at a specific value lower than the thermal cutoff point and overheat will be off.
The temperature would fluctuate in an abnormal way that would lead to overheat and thermal limiting and the other way around.

If you have another controller, you can doing a comparison to see if it has the same behaviour.

Worst case scenario is to contact customer service to send in an RMA for a repair or an analysis. If within the 90 period, you'd be covered under warranty.