Giant DH Team, Cromotor, 18fet Lyen, 111v Lipo

[markobetti]

I like the speed that the camera directly translate. this give real impression of your true speed!

Hahaha, thanks! The lines start to blur really fast!

at 2:21 i heard your motor temp alarm that chime right?
Hey hey!.. overtemp a little bit?

I currently don't have a probe in my motor, I have the probe connected to my controller. My motor wasn't even warm after this.

HAHAH DOC, YOU STILL WANT TO PROVE CRYSTALYTE IS BETTER... SORRY ,..NO CAN DO...AND YOU SHOULD BE HAPPY CAUSE YOU KNOW WHAT YOU AND ZOMBIE ARE GETTING, EVEN WE IN CROATIA WILL NOT HAVE ANY FOR OURSELVES

 

[zombiess]

My controller apparently supports up to 150amps+. I'm afraid I'm going to damage my wires or have a wheel go out on me at speed. I'm using the Turnigy 10g silicone wires for my phase and main leads with 6mm bullets... nothing has gotten warm yet, but I'm just taking my time. I think i'm going to order some 8g wire and some even bigger connectors! I'm just going to slowly increase the AMPs now until I'm comfortable everything is stable.

An 18 FET IRFB4115 controller will not support 150A of battery current, you will blow the FETs. I have mine heavily modded to increase it's power handling through better current sharing and heat sinking and I keep it limited to under 110A continuous if I'm playing around in the danger 35mph power wheelie zone. I can probably burst 125A for a pull or two, but it's not something I plan to attempt as replacing FETs is kind of a pain in the ass.

If you haven't modded your 18 FET controller similar to the way mine is done, keep the amps to the low side if you want it to live. I always try to keep my FET temps under 65C with a max of 80C. I've only ever exceeded 80C once and that was because I was climbing 15-30% grades at 40% throttle, something you just shouldn't do. What I should have done was reconfigure the pack for 12S or 18S and then run 100% throttle, would have stayed a lot cooler.

My new 36 FET IRFB4115 controller should be interesting if it works. 20" longer swing arm vs my current 8" swing arm and lower to the ground, probably around 21" tire height because it's going into a 16" moped wheel. Top speed should be 75+ but I'm trying to shoot for pure acceleration.

Careful with riding and making sure you don't over do it with the controller. If you want 100A you should be running a 24 FET IRFB4115 controller unless you mod yours similar to what I have done. My 36 FET should handle about 220A battery current at 125V, but I plan to run it at a lower setting, like 150A-175A if the front end will stay down.

Just the other day I was showing off how I could do a burn out. Tried again, tire was sticky and it hooked. I rode the battery bag (just like riding the tank on a motorcycle) into a 2ft off the ground power wheelie for about 10 ft. Didn't scare me, kinda thought it was funny more than anything, just wish I had all my accidental wheelies on video. Some great comedy and warnings for those playing with this motor on high power. Even limited to 70A I'll have the front end off the ground a few inches in a 5 mile ride at least 5 times purely by accident and rarely from a dead stop. This motor is torque heaven/hell all in one.

 

[binlagin]

Thanks for the advice Zombiess, I've considered upgrading to the 24fet controller... but I think i'm going to put on some more KM's before I pull the trigger on that one. Unless someone wants to buy a discounted 18-FET controller

In the video, I was pulling 125A... my controller wasn't warmer then 55c. Is it still possible to damage the FET's with out overheating it first? I think I'll bring it back down to 95Amps

What would the max discharge for my 20c batteries be? I'm afraid they might be the next weakest link in my build. 30S 2P....

I dreamed of ebikes last night... I can't wait to commute on this bad boy.

 

[gensem]

Raise that current!!!
My cromotor is laced to a 17" moped wheel 18fet 4110 and im using it with 24s lipo 65a main 160a phase... and its not that wheelie prone at 6-6.5kw peaks.
Have a look at my last videos.

 

[zombiess]

Thanks for the advice Zombiess, I've considered upgrading to the 24fet controller... but I think i'm going to put on some more KM's before I pull the trigger on that one. Unless someone wants to buy a discounted 18-FET controller

In the video, I was pulling 125A... my controller wasn't warmer then 55c. Is it still possible to damage the FET's with out overheating it first? I think I'll bring it back down to 95Amps

What would the max discharge for my 20c batteries be? I'm afraid they might be the next weakest link in my build. 30S 2P....

I dreamed of ebikes last night... I can't wait to commute on this bad boy.

Was that 125A your constant battery current or just a peak? What are your phase amps set for? I can tell you from experience that if your case was 55C the internal temps were most likely over 100C and the FETs themselves were probably over 120C! That experience is the fact that I have temperature probes on my FETs and also monitoring the internal case temp.

Just remember that my controller is modded almost to the max with better electrical insulators that have better thermal transfer and also I feed and pull power from the tabs of the FETs using copper buss bars which also double as heat sinks. The only mod I have left to do to it is add some copper to the ground traces to see if it helps cool things down and maybe some cooling holes and a small fan. If you want to see what I did to my controller I have lots of pictures located here:
EB318 controller mods
http://dynamic.opticalanarchy.com:8080/zombie/bike/eb318mod/

And my current project, an EB236 36 FET IRFB4115 controller
http://dynamic.opticalanarchy.com:8080/zombie/bike/eb236mod/

Here is a brief rundown of what I've seen temp wise:
At 105A battery 125A phase I the FET's would stay in the 55-65C range and the internal case temp would be about 65C. The case itself was around 33C. This is all with the ambient temp around 15C. As soon as I stopped applying power the temps would drop quickly, FET temps would drop from 60 to 45C in about 1 min and the case temp from 60C to 50C in about 3 mins. My controller is exposed to the open air but has no ventilation. What you really need to worry about is how hot the internal temp of the case is getting to since there is no ventilation, that means the capacitors will be stressed from heat and their life will shorten. Also, running FETs over 80C will shorten their life span, continuous use over 100C the life span drops even faster.

I must admit, I'm surprised that it lived, maybe I'm under estimating what these are IRFB4115 18 FET controllers are truly capable of, but I can tell you, I hate blowing up good stuff because I got greedy. I've done it many times from ebike controllers (luckily just 1) to boosted car engines (a few). Once it gets warmer I doubt the controller will survive very long.

If I were you, I'd consider ventilating the controller at minimum, but I guarantee you are in the danger zone. If you climb a hill at part throttle on those settings you'll almost certainly kill it. To give you an idea how little the case temp tells you watch this video of my climbing some stupid steep hill. At the part where I stopped my FET temps were 88C and that was about 100ft before I stopped so I probably hit 90C!, the internal case temp was in the 60C range and out outside case temp was barely warm, probably 25-30C at most. That 88C temp on the FETs is what made me decide to stop. 40% throttle at 125V climbing a super steep hill creates a ton of heat fast.
[youtube]MXejC1bt6EU[/youtube]

If you want to continue and push it, please post regular updates about the controllers survival. I'm very curious to how far one of these can go but have been un willing to test one to destruction myself, then again at 125A phase, 105A battery my bike can power wheelie at 30mph if you aren't fully leaned over the bars so it's pretty sketchy to ride. I strongly suggest turning the block time down to 0.1S (you'll need hex edited software) and keeping the phase amps on the low side.

Your setup sure does look like it kicks serious ass, I just want to see it keep kicking ass.

 

[binlagin]

So your saying... the day I heated up my controller to the point I couldn't touch it... I'm EXTREMELY lucky that it's still working? It was that hot, for at least 5 minutes of heavy accelerating(WOT) in very heavy snow(low RPM).

After that incident, I installed the temperature probe on-top of the heatsync where the FET's attach to inside the case with super glue. How ever, after that mod... I'm basically maxed out on my electronics skill after that. I don't think I'll be doing any hard mod's to it... other then maybe some ventilation as you suggested. I will defiantly keep in mind that the probe isn't reading the temperature of the FET itself... just it's heatsync.

I'm kind of a newbie when it comes to the electrical side of things.

I asked earlier but I still don't really understand the difference of a 65Amp battery main lead vs the higher Amp Phase connections. I think it has something to do with the power "bursting" through each of the 3 phases? So when I say I'm running 125Amps... I'm only speaking of the phase current.... and only when i'm 100% throttle? This is where i start getting lost! lol

Thanks for the detailed response Zombiess... I think it's going to take some time to digest all that.

 

[binlagin]

I commuted to work for the first time on my bike It was preety wet... but that didn't stop the grin on my face!

OMG is all I can say. I was pulling 5,000watt... I think I'm going to turn down the phase current to a max of 75amps for commuting to keep me in line and maximize efficiency. The speed difference between 3500watts and 5000watts is very small.

I looked down at my controller and saw it was reading 90degrees, I swore it was reading 45degrees moments before... foruntatly I must have pushed a button and changed the setting to Ferinheight

My cells where charged up to 4.05v'ish before leaving... I arrived at work with them all reading 3.78-3.8... I'm just going to take it easy on the way home. I hope I can squeeze out another 10km

 

[zombiess]

I commuted to work for the first time on my bike It was preety wet... but that didn't stop the grin on my face!

OMG is all I can say. I was pulling 5,000watt... I think I'm going to turn down the phase current to a max of 75amps for commuting to keep me in line and maximize efficiency. The speed difference between 3500watts and 5000watts is very small.

I looked down at my controller and saw it was reading 90degrees, I swore it was reading 45degrees moments before... foruntatly I must have pushed a button and changed the setting to Ferinheight

My cells where charged up to 4.05v'ish before leaving... I arrived at work with them all reading 3.78-3.8... I'm just going to take it easy on the way home. I hope I can squeeze out another 10km

You might want to consider charging to 4.15V per cell, good amount of storage to be had between 4.05 and 4.15.

I'm hoping I can commute to work on my bike this week but I've been needing my car. Weather also hasn't been looking that great lately from what I heard, crazy winds are coming back, 40-50mph gusts. My commute is just under 10 miles/16km round trip and takes 12-15 minutes one way depending on stop lights and if I'm in a hurry, but all the bike paths I use are now under construction again (not sure how much more than can improve our already nice bicycle paths unless they force walkers to stay to one side and all dogs to be leashed) like they are every year when it's nice to ride. With my 12 AH pack I manage to do a round trip using about 5.5-6AH of the pack which leaves plenty left. My average speed is between 19 MPH and 22 MPH.

Every time I ride to work I arrive more awake and in a better mood, same for the ride home. It's just so much more fun than driving my car.

 

[itchynackers]

For reference, I've done some testing on my lipo cells. My experience is that there is about 10% of rated capacity charging from 4.10 to 4.20 volts. So yes you are missing a bit of capacity with minimal cycle life loss.

 

[zombiess]

For reference, I've done some testing on my lipo cells. My experience is that there is about 10% of rated capacity charging from 4.10 to 4.20 volts. So yes you are missing a bit of capacity with minimal cycle life loss.

I personally chose 4.166 as my preferred charging voltage, after the batteries come off the charger and sit for a day I usually drop from 125.0V down to 124.6V or 4.153 per cell. Since I almost never discharge down to 20% left I think I'm going to get many years of service from this 12AH pack which will soon be 18AH.

 

[binlagin]

Well... just made a huge newbie mistake everyone's been warned about 100 times here on ES.

I was monitoring my voltage across the whole pack. I went from 111v down to 75v inside of 60seconds. I have methods LVC cutoff... but I need to rewire my battery pack for it to work, and I'm out of connectors. Stupid excuse I know.

All 10 of my batteries(60 cells) where below 3.2v and slightly squishy. I am trickle charging all the packs up to nominal voltages and hoping I haven't done catastrophic damage.

I guess this serves me right for trying to commute at 5kw.

 

[zombiess]

Well... just made a huge newbie mistake everyone's been warned about 100 times here on ES.

I was monitoring my voltage across the whole pack. I went from 111v down to 75v inside of 60seconds. I have methods LVC cutoff... but I need to rewire my battery pack for it to work, and I'm out of connectors. Stupid excuse I know.

All 10 of my batteries(60 cells) where below 3.2v and slightly squishy. I am trickle charging all the packs up to nominal voltages and hoping I haven't done catastrophic damage.

I guess this serves me right for trying to commute at 5kw.

Whoopsie, Been there done that got a few slightly puffy LiPos that are still fine after careful monitoring for several cycles. The LiPo cliff is a steep and unforgiving one if you take the plunge and you can hit it so fast on a high powered setup. This is exactly the reason I'm very careful how far I discharge my batteries. Do you know how far below 3.2V you went, was the 75V under load or resting voltage? 75V is 2.5V per cell and if that was your resting voltage your might have done permanent damage to the pack if the voltage didn't come back up on it's own after the packs cooled. If you only went down to 2.8ish resting your packs will most likely be slightly puffed but still possibly OK (but maybe some permanent puffing). Now it's time to start monitoring them carefully for damage so you don't have one go up in flames on you taking anything along with it.

If you run high power on 30S LiPo and you see your resting voltage is 113V or lower, you MUST be very careful. My suggestion is pedal and just use the motor to assist to make the bike feel weightless (aka go slow, like 10-15mph slow) if you have any distance to travel. If you drop below 110V resting voltage, turn it off and pedal or call a friend, no power allowed period! My general rule of thumb is to stop powering around when I see my resting voltage at 115V and to take it easy and keep the speeds below 25 if I have any distance to go.

Please monitor those packs VERY carefully, charge in a safe location at low amps with a balance charger, but don't charge them where it's cold to over 4.0V because when they warm up the voltage will climb.

Do you mind posting the stats of each cell in your pack so we can assess the damage done? How puffy did they get, I'm sure they were quite warm. The safest way to ride at higher power is to have 33% more battery than you think you'll need. That's why I'm going from 12AH to 18AH when I get some more $$$. I don't want to have to think about my battery pack as long as it's in balance which I check at least weekly.

 

[binlagin]

Again Zombiess... thanks for the wicked advice. I really owe you a drink or two I really like your rule of thumb... I will defiantly be applying that to my future rides.

Because I charge in parellel, I'm constantly balancing all my cells. I have not yet moved to a bulk charging solution.

The batteries where indeed warm but not to hot. Most cells where hovering around 3.0-3.2v with no load. There was a few at 2.7-2.8, one cell at 2.5 and one cell at 1.9v. The puffyness is very minor. The packs are basically the same size... just have a "squishisher" feel then my other packs that did not get abused.

I am currently charging in fireproof bags with my window open ready for quick ejection. I'm parallel charging all the packs at 0.5A with balancers connected to each pack.

All the cells are now within 0.1v balance of eachother and approaching 3.7v on each cell. Should I stop charging at this point and let them chill out a bit? or should I just charge them up to 4.0?

 

[binlagin]

Update:
I've stopped charging the batteries... all the cells are around 3.75v and have sat for 15min or so. They seem to have lost their "squishyness" i've been describing.

 

[zombiess]

Sounds like you might be OK, but don't be surprised if that cell that dropped to 1.9V is now weaker and doesn't want to balance and discharges faster than the rest. You might have a pack good for replacement cells if that's the case.

Just monitor the heck out of everything the next few 5-10 cycles before you trust it. Always measure each cells balance with it disconnected from any harness.

 

[binlagin]

Looks like my batteries recovered... but I beleive I blew a MOSFET on the way into work today.

The temperature reading was 40c... but when I hammered it at a green light from a dead stop, my motor started to chug and die.

I believe the day I over heated the controller in deep snow to the point I couldn't touch it, weakened the MOSFETS. Anyways... I've msg'd Lyen to see if he repairs or if I can buy some new MOSFETS.

I can't say you I wasn't warned!

 

[zombiess]

Looks like my batteries recovered... but I beleive I blew a MOSFET on the way into work today.

The temperature reading was 40c... but when I hammered it at a green light from a dead stop, my motor started to chug and die.

I believe the day I over heated the controller in deep snow to the point I couldn't touch it, weakened the MOSFETS. Anyways... I've msg'd Lyen to see if he repairs or if I can buy some new MOSFETS.

I can't say you I wasn't warned!

Easy way to tell if the FETs are popped, flip the bike upside down and rotate the wheel by hand, if it feels notchy you have a phase short most likely caused by blown FETs. Disconnect, one phase lead at a time until the notchyness goes away, that's your bank with the bad FET. Just follow that color phase wire to the board then figure out if it's a high side or low side (doesn't really matter in your case, but there are 2 banks per phase) using a multimeter checking for continuity between the pins. Pin 1-2, 2-3 and 1-3. If you get continuity (multimeter beeps), you found your dead bank. Good practice is to swap out all the FETs in the same bank even if only one is dead and keep the others as spares.

If you disconnect all phase wires and the motor is still notchy then you have a short between the phases in your wiring or in the motor. Unlikely anything in the motor would short, but check the phase wires going into it if this is the case.

If everything checks out OK after all this, then you have a hall sensor issue. These halls are rated to 150C so it's unlikely for one to fail unless you shorted a phase wire to one.

Hope this helps. Swapping out FETs isn't very difficult (with the right tools), but you usually need a really hot soldering iron and a desoldering iron / sucker to get at the leads good. When you pull your controller check for any capacitors that look like the tops have vented. If you get the controller super hot on the inside like you have been doing you can cause the capacitors to fail really easy from heat and high ripple currents. You might want to consider just upgrading to a 24 FET controller since you probably won't turn down the power. I just got done building a 36FET controller myself (a lot of work!, but the next one will be much easier now that I know some tricks)

You can't get too greedy with the power or you will continue to blow up controllers. Even if you go to a 24 FET controller, don't set it any higher than you currently are at. If you do fix this one or get a new one, turn down the power. You really shouldn't be running more than about 60A through an 18 FET IRFB4115 controller unless you have modded it like me and monitor the temps of the FETs closely until you learn what gets them too hot. Ideally they should also be all be matched for what's known as Miller Plateau resistance at a given gate voltage so they all turn on at almost the same time which makes them less likely to fail. When I mod my next 18 FET I'm going to match the FETs and do the same mods I've been doing to make them live at high power.

The fact I only run a 20" tire also helps my stuff live because I don't stress the controller anywhere near as hard as larger diameters do.

 

[binlagin]

Replacing the MOSFETS... what a pain!

I just replaced 3 MOSETS on the yellow phase.

1) I used a multi-meter to confirm that these where the bunk MOSFETS. As Zombiess stated
2) Removed the heatsync
3) With lots of patients and solder sucking... I was able to get the damaged MOSFETS to drop out.
4) I then cleaned out the mounting holes and droped in the new MOSFETS.
5) I had to redo the solders a little bit because I had created shorts while removing the chips. But after a little more time and solder sucking everything was short free.

I reconnected everything and I got my wheel to turn in the stand! I'm going to finish mounting the controller to my bike and take it for a spin! I'll set the phase wires to 65A to minimize the chance of pop'n another FET.

Thx Zombiess!!!!

 

[Doctorbass]

Now try repairing a 24 fets !!! :lol:

Chxs asked me to repair his controller and he was suspecting that the rest of the non blown mosfet might be weak...

So i replaced all those!

And.. what the real pain is... is when you must remove all beefed traces, clean and remove the old flux, then test all brand new mosfet to test OK.. and than to solder all those and to rebuild the beefy traces...!

NOW THAT'S what i call a pain !... 24 fets to replace! :wink:

Doc

 

[binlagin]

Just got back from a run at 65A phase... everything working good! I'm going to leave it at this setting for a few days It's plenty fast for commuting

 

[hillzofvalp]

I would think an easier way of removing bad fets would be to cut the legs and then remove each leg remnant individually.. right? or are you saving them?

 

[binlagin]

I would think an easier way of removing bad fets would be to cut the legs and then remove each leg remnant individually.. right? or are you saving them?

This is basicly what I figured out by removal of the 3rd MOSFET. I defiantly learned alot during this process.

I was able to make it to work today on my fixed controller. Aprox 10km ride... top speed 60km/h... 2000watts. Average speed, 40km/h... 1200watts.

 

[gensem]

Im using a 17" moped wheel with 24s (88v nominal) lipo and im doing around 75kmh... did you limit the top speed?

 

[binlagin]

The only limit I applied was 65A at the phase. I was running 30s. Acutally when I'm fully charged at 30s... my controller won't turn my wheel. I have to move to a 24s setup, use up some volts then my controller will work with a 30s configuration.... no idea why this is.

I really need to get a 3way switch for speed settings and a momentary switch for regen braking.

Any have some suggestions for either?

 

[gensem]

The only limit I applied was 65A at the phase. I was running 30s. Acutally when I'm fully charged at 30s... my controller won't turn my wheel. I have to move to a 24s setup, use up some volts then my controller will work with a 30s configuration.... no idea why this is.

I really need to get a 3way switch for speed settings and a momentary switch for regen braking.

Any have some suggestions for either?

LoL... 65phase is really low... whats your main amps? 65a main also?
Please always include your main amps and phase amps
Im running 60a main 150phase on 24s lipo.