Golden Motor Build 48v and Beyond!

[icecube57]

disconnecting the hall sensor ground while the motor and controller are running may cause problems for the controller.

you maybe could accomplish the same thing if you ran the brake cuttoff line from the controller to the ground in the motor through your thermal switch.

I thought of the same thing it may have been first on my post it was last on my list tho...

 

[icecube57]

Apparently my idea was flawed it cause fear on ES facebook if it was hooked up to the speed switch that if its in a tripped state and it untripped re-enabling Speed 3 ... thus removing the limp home governed mode that the sudden rush of power could be dangerous.... and the fact that it could trip and untrip at some unlucky momment which is dangerous also...I suppose the best way to prevent motor overheating is to not let it get to the over heated state but be proactive to prevent it from getting there. since im getting the speedict I think Im just gonna install the speedict temp sensor in the motor and run the wires out to the speedict.

 

[icecube57]

SANTA!!!!!

 

[Hyena]

http://www.ebay.com/itm/45-Celsius-KSD9700-NC-Temperature-Control-Switch-Thermal-Protector-140-Celsius-/330739494289?pt=LH_DefaultDomain_0&hash=item4d019c2d91

Im going to put this inside the motor and use it 1 of 3 ways. Use it on the ground of the hall wired.

You been spying on my ebay purchase history ?
I bought some of these same things a month or 2 ago and have them wired in series with the hall -ve. They're NC so for the most part you don't know they're there until they get too hot and go open, shutting off the halls. People will cry that it's dangerous if it cuts out etc etc but if you melt a winding it's going to cut out just the same, only you'll be pushing it all the way home - and against a fair bit of resistance most likely.
If you're going to run wires externally you may as well just hook it up to a buzzer. The whole point (for me) of using these internally on the halls is to save having to run extra wires externally. And on motors like the crystalytes where the wiring is easily damaged by removing the side cover it's good to be able to pop off the opposite side cover, yank the hall wires up, snip one and whack this in between.

Never mind the critics, they're often the same lot who are scared of lipo burning their house down

 

[icecube57]

http://www.ebay.com/itm/45-Celsius-KSD9700-NC-Temperature-Control-Switch-Thermal-Protector-140-Celsius-/330739494289?pt=LH_DefaultDomain_0&hash=item4d019c2d91

Im going to put this inside the motor and use it 1 of 3 ways. Use it on the ground of the hall wired.

You been spying on my ebay purchase history ?
I bought some of these same things a month or 2 ago and have them wired in series with the hall -ve. They're NC so for the most part you don't know they're there until they get too hot and go open, shutting off the halls. People will cry that it's dangerous if it cuts out etc etc but if you melt a winding it's going to cut out just the same, only you'll be pushing it all the way home - and against a fair bit of resistance most likely.
If you're going to run wires externally you may as well just hook it up to a buzzer. The whole point (for me) of using these internally on the halls is to save having to run extra wires externally. And on motors like the crystalytes where the wiring is easily damaged by removing the side cover it's good to be able to pop off the opposite side cover, yank the hall wires up, snip one and whack this in between.

Never mind the critics, they're often the same lot who are scared of lipo burning their house down

In a way i was still gonna do it... lol i think and wire it to the halls but im going to use my spare axle space for that speedict temp sensor

 

[Hyena]

If you're running wires for the speedict temp sensor then that's all you need

 

[icecube57]

If you're running wires for the speedict temp sensor then that's all you need

This time im going to over do things in all aspects... You get better the 4th time around... Over build and run it conservative but break the balls on it sparingly... i think this new motor is gonna be the shit tho.... cant wait to get it and crack it Imma strip it molest it upgrade phase wires and have my way with it... I wanna do a true upgrade of the phase wire but if im cramming extra wires down the axle imight not be able to.. so i may take the phase wires out the axle... take the low temp POS sheath off. Shrink wrap the phase wires like I did with my other motor this should protect them and allow me to run more current throught them. I was hitting and 4-6kw peak. With no issues with the phase wires doing that on my last build. The insulation wont get tacky and fuse together and short and allow enough extra space for the temp sensor. ofcourse the poormans phase wire upgrade hacked off a few inches outside the axle. I think imma use xt150s on the phase wires. 8G to the controller. Its nice being able to do quality wire jobs now. I think its gonna be a game changer for this build. Less complex battery wiring/rats nest. Im excited. Its almost been a year since my accident...

 

[John in CR]

Hyena,
I like the idea of easy access for the mod and no new wires to run out. A few things to consider are:
- Some controllers trigger regen braking with a cut in the halls power line
- 140°C is awfully hot, so placement at the windings is critical
- Shutting down the motor is the 2nd worse thing you can do when a hubbie is that hot. Continuing under heavy load is worse.

 

[Hyena]

Hyena,
I like the idea of easy access for the mod and no new wires to run out. A few things to consider are:
- Some controllers trigger regen braking with a cut in the halls power line
- 140°C is awfully hot, so placement at the windings is critical
- Shutting down the motor is the 2nd worse thing you can do when a hubbie is that hot. Continuing under heavy load is worse.

I'm using standard infineons for now, haven't heard of such a thing but that's interesting. These just stutter when the halls are cut. I've had loose hall plugs before while riding and it's not like the motor and controller are going to instantly self destruct. You feel the jerk/stutter and get off the throttle. Easy.
I agree 140°C is cutting it close, I'm actually using 100°C on mine and being on the conservative side. 120 would probably be a better middle ground but at the time 100's were available. Mine is on the windings but it's also close to the air cooling holes (which are over the windings) so that's another reason why I figure it's best to be on the safe side temp wise, as the air holes may cause the temp sensor to read cooler that what is actually indicative of the overall motors heat. On the last air cooled motor I did have a temp sensor in (and monitored it closely) it rarely got over 90C after feeding it with 4000-5000w. The hottest as you say is when you stop, which is a slight limitation of this method if you're planning to be lazy and get off the bike when it cuts out rather than continuing to pedal it. If I was stuck out in the bush (which is where I ride hard) I'd be tempted to add a jet of biological "water" cooling :lol:
I'd originally planned to use a NO thermostat hooked to an internal piezo buzzer powered from the hall voltage but coudln't find any rated to 100C +

But yeah, I see these as more of a last resort fail safe and really I only have one in my own motor to test so I can put them in motors for noobs. I've never smoked a motor, not even a 9C with 8kw bursts but give one to a noob and he'll find a way to kill it with only 3kw (I had a bloke recently who managed to burn a HT35 into oblivion with a current limit of only 45a)

 

[icecube57]

I can cross the spedict off my list and cross the connectors and wire for my harness off my list. All i have to do now is get some torque arms... controller and ebrake levers a multi function switch...which i should be able to get from lyen.. rear panniers... off ebay

 

[icecube57]

All voltages checked out good. Im top balancing them all and then im gonna do a full discharge to find any runts and find the lowest cell type up the info and put a small sticker with some clear tape with that info on it. But they are balancing quite welll. I charger makes quick work of topping of these cells... you know im raping them right charging at the full charging c rate...

 

[icecube57]

Pack cycling begins... first discharge cycle down on 1st pack 5A or 1C. Cell#2 is the runt...delivering 5Ah at 3.39v and 3.0v delivering 5.115AH. Im not going to post every pack but this is exciting being that i never owned an icharger or plotted anything on graph. At 25A 5C the pack went from empty to full on normal balance mode in 24 minutes Balancing took half of that time.. I took 5.11AH out and put in 5.144Ah

Im putting a tag on the packs with clear tape that has the runt cell # and discharge capacity information.

Also i found out at the max 30A rating my 12v Dell power supply can satisfy the Icharger charging at full rate on 4s battery packs so I dont need to use my louder... second power supply in series for 24v

 

[icecube57]

Ive found myself restarting the charge cycle a few times on my charge to get the most AH and closest balance possible. Ive cycled 4 packs so far. They all have given 5AH atleast. with 5.15 to 5.2 AH on their first cycle. There was my 4th pack that gave me only 5AH at 3.28v which is low compared to its previous tested pack and 5.065AH at 3v. So I charged this guy back up and kept restarting the balance mode charging every so many minutes to give that cell a chance to fall in line with the others. I stop doing it when the charger fails to start charging for more than 2-3 seconds. Im doing a second full discharge on this first ...somewhat runty pack of the group. I expect better results on this one. Atleast I should be able to get in the 5100+.

Im trying to figure out the best way to mate these packs... since teh will have known weak cells how do you pair them together. by capacity or dispears the weak cells evenly an let their fellow brother help prop them u. I am creating these into 4 pack cubes where the leads and balance taps will be solder togetther.

 

[icecube57]

My bike torch/ Flashlight came in today. It has an operating range of 3.2 to 8.4v. I had a Delta Electronics cell phone charger. http://www.ebay.com/itm/Delta-Electronics-ADP-5FH-B-AC-Wall-Power-Adapter-Charger-Supply-/400302477340?pt=PDA_Accessories&hash=item5d33e3101c . This smps actual turns on in the high 20s. I took the case off and I was left with a bare board. I ditched the wimpy wires and soldered some silicone 14 gauge wire. I put one 4s pack across the input no go. I put 8s across the input and it turned on... success. I put two layers of shrink wrap over the dc to dc put andersons on each end. So i got that done... now to the flash light. It was straigh forward. Positive was on the spring of the led driver and the negative was on the spring of the screw on cap which serves as a switch to cycle between the modes. I drilled a hole in the case of the flashlight big enough to feed the two 14g wires through. I beveled the edges of the hole with a dremel to reduce any edge that could cut the wires. I fed the wires through fluxed and soldered the wires and closed her up. I put hot melt in the hole to act as strain relieve and to also prevent the wire from getting cut. Here are some pictures of the finished product.

 

[icecube57]

Got my first true weakling..... 4.848 AH at 3.0v a cell @ 5A 1C discharge. Recharged... squeezed a bit more AH back in ... not much tho... Im doing a second discharge at 2.5A or 0.5C... Hope I can recover some capacity. I never use 100% of my pack but id like to know if its there. If im hypermiling im normally doing less that 1-2A at 20mph which means it will be .25 to .5A from each pack in a 4 P configuration. If i can get more capacity running at a lower C rate that would be nice to know its there...Im going to have to put it in with one of my other packs that has a high capacity like the ones that had 5200 to offset the missing .2ah in this pack. Im paying attention to the low cells to make sure a pack doesnt have two weak cells in the same position.

 

[Hyena]

Im trying to figure out the best way to mate these packs... since teh will have known weak cells how do you pair them together. by capacity or dispears the weak cells evenly an let their fellow brother help prop them u

If you group the strongest cells and weakest cells together you'll end up with one strong parallel group and one weak group but it'll still be the weak one that lets go first when you discharge in series and limit you overall. If you want to get the most capacity from the pack as a whole you'd parallel the strongest cells with the weakest cells so their average was raised.
So with a 5.0 ah and 5.2ah cell in parallel I'd think you'd get a discharge of 10.1ah.
That is assuming you're paralleling those cells at the cell level. Unless you plan to reshuffle your packs or they happen to be same cell number in each brick (eg cell 1 in one pack A highest and cell 1 in pack B is lowest)
14ga for your light is massive overkill, they only draw like 3a @4v max. You could even run it off the CA Aux out if you wanted to save running wires back to your battery.

Anyway enough of that, where's this motor ?

 

[icecube57]

USPS is predicting the 30th... i got my batteries a day early from the West Coast in Washington im hoping for tommorrow but ill take a Saturday delivery.