JRK Power turning off

After having my JRK on for awhile the power trips off and back on again. I can see the JRK's light turn from yellow to red.

Does anyone have a idea why this is?

I'm guessing it is the temperature of the main processing unit. I have a heatsink attached, but no fans are pointed at them. When I finally pulled the power the heatsink on the JRK unit that was turning on and off was much hotter than the other one.

Is there a way for me to see the temperature of the JRK?

So, I'm guessing that there is a sensor of some sorts on the JRK that shuts the power off after the chip gets to a certain temp? And then once it goes below this temp it trips the power back on?


I'll try to get something set up that I can run for some time, but for now I just want to get it set up. It seems fine for 20-30 mins. So, maybe something I can worry about later if it automatically shuts down the power.

I don't have a small fan that one would use in a computer or something like this. I have medium sized fan I use for my bedroom. You think I can get away with just pointing that thing at the jrk and blasting it?

Yep - the JRK 12v12 uses the ST VNH2SP30 H-bridge which has an internal thermal cutout/shutdown. You want to minimize how often that activates because it is stressing the chip circuits to get that hot.

There's some more info in this thread but not a huge amount. (https://www.xsimulator.net/community/threads/jrk-12v12-overheating.10327/#post-133655)

The JRKs bond the thermal pads on the chip to the bottom layer of copper on the board and the max current that the JRKs can handle is a function of the board copper area and how much heat that can dissipate. You can really increase the current handling if you heat sink the bottom of the board. Most heat sink the top of the chip but that doesn't cool all the heat generating parts and the plastic of the bridge has fair heat conductivity but only about a quarter of what the thermal pads do to the bottom copper. The bridge chip has an internal current limit that activates around 50 amperes or so. The JRK is only rated at 12 A continuous so if you heat sink the board, you can pick up a lot of headroom but I don't know just how high you can go.

If you heat sink the bottom of the board, get some thermal pad material with high thermal conductivity and isn't more than a mm or thereabouts thick. You want to clamp or use mounting screws to get good contact across the bottom of the board and you need to be careful not to over tighten since there are only three screw holes to mount with due to the USB connector. Using thermal pad like that on the bottom of the board also means nothing can stick down and pierce the pad to short out against the heat sink. You probably will need to clip and smooth the feedback and power/motor connector pins to have those connections be flush with the bottom of the board.

This is how I'm mounting mine. Preliminary tests with two fans for a wind generator (about 12A but I didn't measure it to know) showed around 30 F temperature on the top of the bridge chip on an unheatsinked 12v12, and only 2-3 F rise on the top of the chip on a board with a heat sink. I haven't run them under heavy load yet but no reason not to expect an even more dramatic difference.

The last pictures are a slightly different arrangement where I clamp (first photo was a clamping test to make sure the contact points were good) a JRK onto a heat sink on top of a box for a seat belt tensioner. Same thing, though. A fan is pulling air through the JRK heat sink and blowing it down onto the power supply. The photo of the aluminum heat sink and fan assembly is the big boy I'm using for my motion motors and should have much more cooling headroom. The first picture is how you want the JRK board to be on the bottom for this method - flat with nothing sticking out to short to the heat sink. The thermal pad is an electrical insulator but has high thermal conductivity and is also very fragile. Anything sticking down will poke right through and short out.

But, as others here will also note, you can use a heat sink and fan on the top of the chip too. That's worked for lots here for a long time. The way I'm doing it is way overkill for most installations but I like keeping electronic things cool.

Thank you so much for all the information. I probably will be sticking to a fan and heatsink type build for cooling the JRK's since I don't think I will be able to get the bottom part of the JRK smooth. I just don't have a lot of experience working with electronics and I'm more inclined to move in a direction that is an easy fix. I do see the benifits that your setup provides. Cooler is always better, and this seems like the best setup. What sort of material is the red stuff made of? Is that plastic? So you are using a regular CPU heatsink and just drilled some holes to mount some screws into?

That's it exactly. I used a 3D printer to print the boxes out of red PLA and the heat sink fan assemblies are cheap aluminum CPU coolers from Amazon.

One other thing I thought to mention is that you can also look into other reasons for the heating which may not just be straight up current. @Avenga76 has a number of posts here where he discusses back EMF from the motors and what that does and how to block it with chokes in series with both leads to the motors. I haven't tested the high current chokes I've gotten to know if I can recommend them, but check Avenga's posts for more info. I don't have time to find the links but others will know if not Avenga himself.

Good luck!

Interesting thoughts on heat sinking the bottom rather than the top

I found this bit from TI (http://www.ti.com/lit/ug/sluu483/sluu483.pdf) that explains the principle pretty well though it's not exactly the same. The JRKs have a lot of vias under the thermal pads of the bridge and it is soldered down to a smallish area of copper - the board size of the 12v12. Polulu is getting the heat at least to the copper on the bottom of the board but the example in the TI document compares to an unsoldered chip to show how that impedes the heat flow. But the part about trying to get the heat out through the plastic package is the same, the thermal conductivity is lower so it's not as easy to do and the chip circuitry runs hotter. Thermal conductivity through the bottom of the chip is designed to be high and the vias that Polulu uses on their boards do a great job. The heat sinks on the bottom with thermal interface material just work to get that heat off the circuit board as efficiently as possible.