EB324 Controllers keep blowing up! Problematic controllers

As many know Xie Chang has dropped the EB224 controller and started using a new EB324 controller design. This controller design ends up splitting phase be pretty far apart which is not ideal. I ordered 20 of these controller and sold a few after doing some personal testing on them up to 160A battery and around 200A phase without issues. Before going through and selling more I wanted to make sure these were OK before I sold 20 of these and found out all of them were going to go bad. I'm glad that I'm waited (largely due to my laziness on building controllers).

So I had a customer return 2 controllers that I he had tried out, both failed on him very quickly and at low power, low speed. I took the damaged controllers apart and diagnosed them. To my surprise they all had the exact same failure, phase B (green) had exactly one FET die. Physically the FET looked OK, it was just internally shorted. I repaired one with a matched set of FETs and installed it in my own Greyborg Warp setup for further testing. I rode it for 20 miles and today when I was at my furthest point from home... it died. Let me tell you it is not fun to try and pedal a 100lb bike home 6.5 miles with 52T/16T gearing on 25" diameter tires. Thankfully most of the ride was downhill and I was able to get back home without killing myself too bad.

I just ripped apart the controller for diagnosis and took and educated guess that the exact same FET failed again. Removed it and bingo, rest of them were good. Removed a 2nd one and matched it against my known good spares on the FET matcher and then soldered them both back in. While looking at the controller I thought this is being caused by a design flaw. The FET that has now shorted 4 times on 2 different controllers has the shortest path to the phase wire. All 3 failures were identical in nature, low speed, low load (but high phase current due to being a buck inverter).

I decided to unsolder the phase wire and attach it in the middle of the trace between the high and low FETs. This makes the current path a lot longer from the high side FETs which means the resistive/inductive differences should be closer together percentage wise vs it's stock location. I took some pictures to illustrate what I suspect is a solution to this problem. I need to test this controller for 20-50 miles before I go ahead and start selling the rest of the 24 FET controllers I have. I will be carrying a spare 6 FET controller as get home backup in case I suffer another failure. Pedaling this beast sucks.

At first I suspected I could have been suffering some issues of solder flux conducting enough to turn the gate on of an FET, but I started cleaning it off between all traces and especially around the FET legs. Still had the same FET fail. I leave it alone on the big areas as I am not worried about it doing any harm there.

Pics

High side, FET that keeps failing is the 2nd to last from the right.


Here are the current paths in pink, the shortest one is the one that keeps popping at low throttle
View attachment 1

My attempt to mitigate the differences in current path length

Coolbeans man. So would a 6fet controller really be ok pushing all that weight around?

snellemin said:

Coolbeans man. So would a 6fet controller really be ok pushing all that weight around?

Click to expand...

Yeah, keeping it below 40a batt and 80a phase is fine. My wife's bike runs good like this. Its a limp home backup and I have plenty of space to store it in the bike since it is so tiny.

Yeah I had similar problem on the 18FET version. It kept shorting out at the same FET at low power. Sometime even downhill. So I gave up and went with a Kelly controller. Haven't looked back. Maybe I will explore what you are saying here, but might not be worth time.

Any who good luck on the repairs. Even with the Kelly I always carry a sensor less controller backup. If anything goes wrong, I can get back home with it, albeit at low power and speed.

would you consider a short piece of copper conductor buried in the solder sufficient to get more current to the source leg on the outside mosfet?

i have found that the mosfet closest to the current sensing shunts on the BMS are the ones to go first because of the heat they create, so it may be a function of that adjacent loside mosfet heating up too.

would adding a schottky diode to eat up some of the flyback current help?

I do not believe heat to be involved into the failure, at least not heat that may be removed quickly through the tab of the high side FET (The side that keep failing). During the last failure everything was cool and no prevoius hard riding, cruising around 10mph it shorted.

I don't know an external diode would help or not since I am unable to watch a scope during the failure. The current path issue is an educated guess on my part.

I use to have a picture from a customers 18 FET controller where 3 of the high side FETs had literally melted the plastic hole insulators used on to-220 installs. After talking to the customer I count out there was a throttle issue causing the trike it was installed on to always go about 5 mph. It finally killed the controller when he was stopped for a period of time and the heat kept building in one phase until the weakest FET gave up causing a chain reaction. All three were bad, but besides melted insulators looked OK from the outside vs the typical catastrophic failures most get.

The more I work with these cheap co trollers the more I learn. I am 99% sure my failures were caused by unequal current paths. I can not see any other reason so I am following Ockham's razor.

Keep us posted on the fix. I have a feeling this might not be only defective on the 24FETs design but also the 18FETs version.

mvly said:

Keep us posted on the fix. I have a feeling this might not be only defective on the 24FETs design but also the 18FETs version.

Click to expand...

I have had much better success with the eb318 boards. If I can figure out how to get these 24 FET controllers to work more reliably I will sell them and then I am done building these type of controllers. Too much work for me to build. If they would have kept the eb224 boards I would have been sold out by now and out of the controller building game.

I moved the wire on mine. A 40 watt soldering iron is too wimpy to do this job. Thankfully my neighbor had a 100+watt soldering iron to do this and it worked great. I went for a short test ride to confirm the controller worked and it did. I will post back after I do some more tests when its not raining out side.

Update: Today I just went for a 12 mile ride to my old high school. I upped my battery voltage to 18s, ~70 volts. About 5 of these miles consisted of speeds of 5-15 mph and I had no problems!

well, just got back from killing it again (died at low acceleration from around 10mph while making a corner). Now I'm thinking I might have the phase current set too high. I don't know what the actual phase current is since the software doesn't have this 24 FET version of the controller in ( have to program it as an 18 FET) it and previous versions (eb2xx) made it higher than programmed. I don't know if the EB3xx boards work the same way or not.

Next step is to pulled it apart and see which FET died, it's still on the same high side set, but I'm not sure if it's the same FET or not. I'll fix it and turn down the phase current to see if it stops blowing up.

Phase current is programmed 140A, but I don't know if that means it's 140A or if it's actually more like 186A because I'm programming the controller as an 18 FET.

I might be stuck with a bunch of useless/low power 24 FET controllers... yay. The EB224 design was far superior to these EB324's, I haven't had any issues with any of the ones I have sold or run. I'm kinda glad I have a bunch of EB236 controller boards that I can build up since those have been quite reliable as well. There is certainly a flaw in this design since Phase B keeps puking and none of the others. Of course phase B is the one they split in a funky way which is something that generally should be avoided.

I think I'll give this one more go and if it goes again scrap / sell off the EB324's I have left. Might have to limit these to something like 100A battery, 130A phase (but programmed much lower). The 18 FET controllers I built could run these same currents, but would overheat quickly, these stay cool at this power level.

zomb, on that hi side group, where do they bring the high side drive return to the phase "bus". Hopefully very near the High Side Source star point.

is that serpentine trace draining the charge off the gate from the emitter of the transistor above it? is that serpentine trace present on the other gate drain down transistors? that long?

maybe there is some inductive feedback from the trace or that trace adds enuff extra resistance that the one gate does not drain down enuff. i assume they are all 100R surface mounts so you might see if you can bridge that trace or add another surface mount resistor directly on top of the 100R to get the gate resistor equivalence down by 5% or so. so much for matching the fets though.

Well, 2 more failures later and I'm calling it quits, this controller is ready for the garbage. I'm thinking most of them are since they don't seem to handle the same power level as an 18 FET IRFB4115 can. I'll try one more that I have built but if it too pops the same phase then this new design is junk.

Every time it's failed it's been the exact same FET on the high side of phase B so there is something going on with it, but I have no way of narrowing it down since I'm unable to run a controller under load with a scope. I have a feeling that this new EB324 design just doesn't work as good as the old EB224's which have been 100% reliable for me at much higher current settings.

I'm going to put a 18 FET or 36 FET controller on my Greyborg and hopefully call it "done".

I tried turning the amps down to 80A batt (actual) and set it to 100A phase (somewhere between 100-150A, no idea since it's programmed as an 18 FET).

Does this mean that my controller set to 70A batt (actual) and 88A phase (117A actual) is as high as I should go? I really wish I could go higher on the battery side.
This may be a dumb question but what if the fet that keeps blowing up is just taken out?
Is it possible for you to put one of your extra cromotors (if you have any) on your test stand and test it under load with a disk break to create a load?

I have another 24 FET I want to try on the bike and see if it too pops the same FET or if it behaves differently. Goal is to run 100A battery side, I don't care how many phase amps I run as long as it's > 110A. Bike is still fast and the controller will stay cool for a very long time at this power level.

I have noticed that each time I repaired this one controller I seemed to get less and less run time before it failed again. Crossing my fingers with the other one I'm going to try out.

Scott, I have run as high as 160A battery on one of these same controllers on my high speed wind motor. I only put around 5 miles on it though, while testing it and it was used as an on/off switch mostly, very little part throttle which is where I seem to be getting my failures. I'm not suggesting you do this, I'm trying to narrow down the issue. The older EB224 controllers I built with IRFB4115 FETs had no problems running 150A battery 225A phase, too bad those boards are no longer available

I'm no expert here, just I'm intrigued with the problem. These board have been reliable.

I suspect there are two potential causes for same FET failure: 1) The high side is driver for this particular FET may incorrectly driving too high a voltage, or triggering accidentally via some interference. Would need to look more closely at the circuit and layout. Could reinforce ground paths around the high side drivers, or potentially re-route the drive signal.

or 2) this particular position of FET is too far from a low ESR cap. suggest adding a ~1uf metallized polyester film capacitor nearest the power source and this fet, and replace all the large caps with Rubycons, that have very low ESR.

Mark.

hardym said:

I'm no expert here, just I'm intrigued with the problem. These board have been reliable.

Click to expand...

I am one of the first to use these boards, where have you heard they are relaible? They have only been around a few months and making me miserable.

I meant the Xie Chang boards have been around for a long time. I've repaired a couple, usually broken due to operation/installation errors.
What kind of large Caps are populated on the new EB324 boards?
Mark.

Any progress on a solution? If I keep my amp settings the same would it hurt to up my voltage to 125?

Scott said:

Any progress on a solution? If I keep my amp settings the same would it hurt to up my voltage to 125?

Click to expand...

Small progress, I switched to a different controller vs the one I kept repairing. This new one has gone about 8 miles now (I'm worried to ride too far from home while testing, last time the controller died when I was 7 miles away, thankfully it was mostly down hill).

Unless you are using a small diameter tire there is no reason to run 125V. 125V into a 20.5" tire will get you to 60mph.

zombiess said:

Unless you are using a small diameter tire there is no reason to run 125V. 125V into a 20.5" tire will get you to 60mph.

Click to expand...

My tire diameter is 25.8in and I can go 58 mph (GPS speed) at only 60 amps. I would really like to hit at least 60 mph, because I feel like 60+ (on a friends private street) would sound better than 58 mph on my Tesla application.

I had a similar problem on a 12fet ( ages ago ) that I changed the fets out off ( changed them from stock to 3077's ) the controller worked as expected with the stock fets pushing upto 70A though the controller without issue but after the upgrade ( same current settings ) I started to get the same set of phase fets blowing and only one side. It turned out that the gate connection was dodgy ( intermittent on one of the high side fets due to a damaged track ( I could not physically see any damage) I beefed up the connection between the 2 gate legs of the fets and this solved my problem. Im not saying that its the same problem here , but It could be a possibility.

gwhy! said:

I had a similar problem on a 12fet ( ages ago ) that I changed the fets out off ( changed them from stock to 3077's ) the controller worked as expected with the stock fets pushing upto 70A though the controller without issue but after the upgrade ( same current settings ) I started to get the same set of phase fets blowing and only one side. It turned out that the gate connection was dodgy ( intermittent on one of the high side fets due to a damaged track ( I could not physically see any damage) I beefed up the connection between the 2 gate legs of the fets and this solved my problem. Im not saying that its the same problem here , but It could be a possibility.

Click to expand...

I switched to a new controller, still an EB324 that was blown up by someone else, I fixed it and it is now working at 175a phase 175a battery for over 20 miles, most of them with me trying to blow it up. I'm thinking the issue was the board now.

I have one of these controllers in a customers hands and I just had him up his amps to 120A batt/ 140A phase to see if it lives. I'm replacing it for him for free if it blows, I want someone else testing besides me.

Here is some temp data for my EB324 controller.
My bike set up for this data was 24s lipo which was about 93 volts with sag. My controller is set for 120 amps battery and 140 amps phase. I have a cromotor V2 in a 25.8 in wheel with maxxis hookworm tires.

First controller temp, my controller runs a lot cooler than my cromotor. The cooling is done really well; As soon as I let off throttle the controller INSTANTLY drops about .3 degrees C and continues to drop around .5 C per second. Since the CA v3 for now can only take in one temp input I made a switch to switch between motor temp and controller temp. My temp sensor is mounted to the FET that kept blowing up on one of your controllers. The sensor is on the FET labeled “V2B”.

Today it was around 40 degrees F or about 4.4 degrees C. The controller started out at about 16 C (16 degrees Celsius). On my first hard acceleration to 57 mph the controller got up to 26.2 C and the motor got up to 45 C. I only held full throttle for 45 seconds or .35 miles. 3 seconds after I got to 57 mph the controller started to cool down. 10 seconds later the controller was at 22 C
After 20 seconds the controller already cooled from 26 C to 20 C, after getting to 20 C it took one minute to get to 19 C probably because it was getting close enough to the outside air temp.

For my second run I pinned the throttle for .4 miles (33 sec). Quickly (6 sec) my controller got to 27.2 C getting to 54 mph (I’m not leaning down I’m riding straight up). After I got to 54 mph while pulling 57 amps my controller immediately started to cool down and was 25 C 20 seconds later (while still at full throttle).

To get to my old high school I had to go up a 6% grade hill which is pretty long the first section is only 3% grade and because of traffic I only could go 35mph at about 25-30 amps. The controller stayed around 30 C. Once I got the 6% grade part of the hill I went full throttle and got to 45 mph pulling 70 amps and the controller got up to 33.3 C this part (6% grade part) was .35 miles for 20 seconds. The temp dropped only after I let off the throttle. 20 seconds later while doing some slow 25 mph riding it was down to 28 C. At this point I stopped my GoPro and do not have any more data to provide.

Scott, your letting out your major accomplishment you told me about earlier today. You joined the 70mph club on 125V with the EB324 controller I built

Scott has been helping me test the EB324 controller since I sold him one long ago. It lost an FET on phase C for some reason after less than 10 miles, but has been good since I fixed it. Over 100 miles now and counting and he tried 30S aka 125V today at 120-140A.

Since both he and I have now had success in getting 2 controllers to live I will be finishing up the rest of the controllers I have left and finish up selling them. Don't know how else I'm going to move 500 IRFB4115 FETs I have in stock... already installed on heat sinks ready to build controllers.