Golden Motor Build 48v and Beyond!

[icecube57]

Im assembling my Bosch packs I got in today. Grinding down the plastic so they stack and soldering on bus bars. Im trying to build a 72v 13.2AH. I only have half of that now. 72v 6.6AH Its a start. I just like the way they are stacking and they hold well together. The bus bars are easy to solder on. Once i get done creating two 36v packs Im going to solder the battery status board to the top of the pack so I can have a quick glance a pack life. Doing it more for vanity than anything.

[icecube57]

Well I just completed my first voyage at 84 volts. Well its really 80v after surface charge is gone and 74 under a 45A load. I have 6 bosch packs running 3p 2s. I plan to go to 6 p 2s .... when money gets better i just dropped a ton on a wedding ring ... ugh... all i get is a cheap band... that blows....I soldered bus bars on them connecting all three packs together. They were luke warm after i got halfway into pounding them. My GM was above warm. I could leave my hand on it without a problem but you can tell it was hot. I have the three speed controller and was wide open at 45A thats like 3300w at 74v under load. I think i will back off but there is a hell of a difference in the highest speed and the lowest speed. I need to know can someone calculate my shunt resistance or do i need to put more batteries in parallel to get the amps higher. Also the speedo on my CA worked fine on my old controller connected to the hall wiresbut on this one its connected internally and I get erratic reading. It sometimes work and sometimes dont. Im tempted to tap the hall wires on this controller also to see if i can get a stable reading. I gotta calibrate my meter. I have three pictures that i took when i was doing the reverse current trick to get the shunt reading. I think the voltage and the amps are to low I may have to try with more batteries in parallel. But the watts up gives the voltage of the battery and the amps being drawn the voltmeter is measuring the volt drop. I believe this is all the information needed to calculate the shunt value.

[dnmun]

i didn't really follow that. if you have 33V at the input to the voltage regulator, that is ok, but you need to keep track of how low the 12V rail goes when the battery voltage drops during discharge.

i estimated there is about 40mA current draw when it is running, and 27mA when idle.

so make sure the input to the 12V regulator doesn't drop below 9-10V or so, but 14V minimum is best. you can measure the output of the 12V regulator, it may be 15V in the golden motor controller, but look for that as the output and if it gets pulled up then you need to increase the input resistors. make sense?

[icecube57]

Thanks to the info but ive put that controller on the back burner. Prob sell it off for cheap. Im going to get a 36v controller I can overvolt to 48v. Atleast I could use this ample supply of 36v packs i have now. Im using my new controller now... the SOB runs warm in standby and warmer while running. i need to see if i can get some type of fan blowing in it.

[pwbset]

do i need to put more batteries in parallel to get the amps higher.

Considering a single fatpack will burst to darn near 45A paralleling more isn't going to help that issue. If you have 3p and are only getting 45A max that's a limit of your controller or motor or both. 3p fatpack will happily give you 90A+ sustained.

[icecube57]

No what im talking about is when Im doing the reverse current trick to get the shunt value of my controller. I was using one AA lithium to feed the negative current through the controller. I may need to put more in parallel to get the voltage and amps high enough to get decent measurements to calculate my shunt value for my CA. I posted pictures of the voltage and the amps that was being drawn from the the AA battery and the voltmeter shows the volt drop in mv i just need someone to try and calculate the shunt value other wise... i can get parallel more batteries to get more current going through the shunt for a more accurate reading.

[pwbset]

Oh right sorry. I just did the same thing the other night with 2 D cells in series (the same cells I'm using to trigger my SCR tab welder gate.. convenient). I was getting about 3.5A steady and then just adjusted my Rshunt CA value until they lined up. Not that great, but better than eyeballing it and guessing. I don't know how to help with the WattsUp sorry. Good luck!

[icecube57]

Well I had the watts up meter powered up through the auxilary connector. I connected a AA battery on the source side and it showed 1.47v i hooked the negative on the load side to the phase wire and the positive wire to the negative connector on the controller. Then it started drawing power. The voltage sagged to .7x volts it was drawing an amp or two and i measured the shunt wires on the CA connector and thats whats displaying on the meter. I just need a coherent formula. When i try to figure it m decimals seem off... I think its 2.5 million something but i need a seasoned professional to help me. All the information should be there on my photos i took on rapidshot on my camera.

[Hyena]

Are you still running the standard phase wires on this motor ?

I'm going to be running my 500w at ~60v/40a when my new lipos arrive in a few days, so it's good to know yours is holding up ok at a bit higher voltage. You're my canary, when your motor falls off its perch I know it's time to keep a close eye on mine

[icecube57]

Motor wires seem cool. Stock wires are coming out of the motor about 1 ft . Then I have Andersons and 10G wire going to the controller. I did that to reduce the resistance a little bit to avoid them getting to warm. The motor itself was hot to the touch but not burning. I could leave my hand on it without having to remove it. This was after 10-15 of hard WOT riding in 90-95 degree weather. i will back off the amps a little bit and go down to one of the lower speeds on my controller. I have the speed selector switch on the handlebar so i can get those extra amps when i need it.

[icecube57]

I cleaned up my wiring a bit and programmed a new CA shunt value i calculated about 2.67mohms averaged by the three readings i posted. Im still off a bit but not by much. I need to go to the store and get some D cells and solder them together so i can get a higher reverse current to get a more accurate reading when i check again. I fixed my CA speed issue. I just cut the speed sensor wire put a crush cap on it and took the other end and tapped the yellow hall wire. It now reads correcly and doesnt cut out. Keywin needs to check into that. I actually use my speed select switch today.Speed 1 sucks balls. Its limited to 10-15A and peaks at 15mph. 20 if you are on an incline. The max amp tapers down so much that once you reach 15mph it doesnt assist you. It like 0-2A at 15mph. You can pedal faster than propels you. It can be good for a soft start maybe. Past 10mph I have to go to speed 2. I think speed 2 is 30A unrestricted. Speed 3 maybe 30A with timing advanced... not sure. All i know is keywin did deliver me a controller but not quite as expected. At this higher voltage it dumps alot of heat into the motor. After a while the motor become saturated with heat there is a noticeable reduction in power. Again the motor is hot that i can leave my hand on it. It does not burn or feel uncomfortable. The batteries are still delivering 30+A at 72-74v but the motor doesnt have that punch like at the start of the ride. Methods if you read this check your pm i need the info and or cables to flash the parameters of my controller. I hope i can. This is a new series board keywin is using Also what should the lvc be on these bosch packs?

[chet]

I picked up a nice cromalloy Trec Mountain Bike frame, USA, with rusty rims for $5 at the police auction.

[icecube57]

Ok i did more testing on my controller and battery. When my voltage was sitting in the 70s the speed selector really didnt make the much of a difference. When it started getting into the 60s then it started to clearly show that speed 1 was 10A speed 2 was 20A and speed 3 was 30A. Also regen started to kick in when I entered the 60s. Its kinda nice I ended up with .75AH in regen. The controller doesnt have an LVC that i was able to hit. I think keywin didnt set one and thats why keywin included a pot to solder on and set the LVC. Also the controller ran at 33v when i tried to put 1 pack on it. It ran and it was rideable and the speed selector worked. Keywin said i couldnt go below 58v but i ran the controller at 33v. What damage can occur by running below 58v. I kno Regen was working at 33v also. Its not a full braking regen. I would say its like a 10% drag. and it kicks in when you get above 10-15mph and stops regen when you drop below it and it free wheels normally. I was getting about 5A of regen most of the time. I peaked at 10A during regen and 11.7% regen recovered. The pack discharged to 6.2AH but its not totally accurate. 5-10% variation until i test my shunt and set the CA again So maybe this is a controller with no lvc as long as the power stays high enough for the logic to stay on.

[methods]

I did not read this thread but in response to your IM>

Sounds like everything is working as it should.

You need to mod your controller to get past the 75V limit.
In a nut shell, you are going to spoof the CPU into thinking your input voltage is lower than it is by some percentage.
This will allow you to use the advanced features (like regen) at higher voltages (like 90V).

That is what you are after right?

SAM-Pilot and Knuckles both pointed out things that you would need to know in the last few pages of my "18 fet Infineon boards are here" thread.
Go check it out and let me know via PM if you have any more troubles.
I am not monitoring this thread.

-methods

[icecube57]

I ordered some torque arms from ComCycle when I recieved them they were for a a X5 rear. I needed torque arms for a Front X5. I sent them back and a few days later I got two front ones. I was going to put one on each side but then i didnt want to hack my motor wire harness. So i put them on the opposite side. I took off my loose X5 J Arm torque arm. It was less than snug fitting on the axel. Probably would have sput with the right conditions. I put my a washer and 2 new torque arms on the inside of the forks on the non wire side of my motor. They fit super snug. I had to gently tap them on with a hammer. There was a problem the way the drop outs were made I couldnt mount the torque arm parallel with the fork. They are set forward . I used the strap that came with the Clyte X5 torque arm to secure it to the fork. Is there any problems with this configuration.

[icecube57]

I went out on my infamous proving ground. The Silver Comet Trail. Its like the dried salt lake in the desert where all conditions are perfect for speed trials. I found me a nice open stretch and let her do her thing. I topped out at 41.8mph @ 72v under load. The thing is it could probably run wide open all day without heat issues. When I started doing jack rabbit starts and playing around wasting juice it really started to heat up.

[icecube57]

Here is some more GM Eye Candy. I got bored and decided to go up to 96v... But true 84v nominal. I have my two Bosch 36v 6.6AH fat packs in series with my 12v 12AH nimh pack. It sat at 97.4v with all 3 packs straight off the charger. It sagged down to 85.2v under load. After the surface charge was gone it sat at 93.5. Its consuming almost exactly 1AH a mile. 60w/hr per mile. Super high I know but its under solid WOT. Free wheeling speed is 70.9mph. I was doing about 3300w continuous. I peak at 45A. I only maxed at 38mph loaded speed but it dont have any open stretches in my neighborhood. I estimate about 46-47mph top speed. With me on it. After a 5 minute ride the motor was barely warm.

[icecube57]

Here is the configuration.

Two 36v 6.6AH Bosch Fat Packs.
One 12v 12AH NIMH Tenergy Sub C Pack w/Watts Up Meter attached for better consumption accuracy.
One Ecrazyman Controller 60-84v. No LVC.
One DP Cycle Analyst.

I went 19.57 Miles.
Consumed 6.322AH on my CA and my Watts Up said I consumed 6.338AH
My starting voltage was 97.4v Resting
My ending voltage was 80.4v Resting
My CA said I peaked at 44.03A. My Watts Up Meter said I peaked at 36.55A.
534.09Wh Consumed.
27.1WH/MI
The ride was 1 Hr 2 Min and 41 Seconds.
Top Speed 42.7mph 18.7Avg

Watts Up 12v 12AH NIMH Tenergy Booster Pack Results (4 Sub C Packs in Parallel)
6.338AH Consumed
Started 13.2v
Ended at 12.70v
Lowest voltage was 11.43v under 36.55A Peak Load
76wh Consumed

The motor was just warm after this run. Could leave hand on it forever without pulling away. Outside Temp was 85 degrees. I did a few high speed burst to try and get a max speed. To crowded on trail to get an open straight area to test top end speed. Speed 1 topped out at 18mph@19A peack speed 2 was topping out at 32mph@35A mph and speed 3 was giving me the rest of the extended speed range where topped out at 42.7mph@35A. Its not limiting amps though on mode 2 and 3 Unless i run at 60v or less. In the higher voltages Speed 1 is always the economy mode but speed 2 and speed 3 only change top speed and not amps. The controller stayed warm to the touch during the whole run. Wasnt alot of loading on the motor except for short speed boosts or crossing intersections it was a solid WOT test. I would say 30% of the ride was WOT. The Surly Fork was pretty solid. It flexes alot while riding over the asphalt surface. But the rake on it is pretty high so I suppose it dampens the ride just a little so that is not a fully rigid fork.

[icecube57]

Today I spent about two hours installing the temp sensor in my motor. I realize the motor isnt solid. The stator isnt. It consists of two plates. Instead of it being solid steel like an X5. Its pretty hollow. I though that the GM ha more mass than the 400 series Crystalyte motor. I thought wrong. There is nothing to absorb heat so it must be immediately reflected into the air.

The sensor was a bitch to install. It required me to cut all the wires. Remove the cable from the axel. Thread the sensor through. Strip back about 4-5 inches. Tape it tight with electrical tape. Thread it back through tha axel hole and solder all connections and heat shrink them. I zip tied all the wires close to the stator. The probe is on the stator but is on the outer ring below the windings.

[icecube57]

I blew up 2 controllers in the past two months. Learning from my mistakes I ordered 1 more and keywin gave me the shaft. He didnt ship me the new 116 MCU controller. He sent me the older 72v 45A model. I was pissed. It wasnt pimped out at all. At the very least I could have got a CA connector. I requested one in the paypal email. He shipped me a Plain Jane Stock!

I took a deep breathe and got out the iron. I looked at the Infineon Dummy threads and other Technical info on this model controller.

I did the R12 mod. The original Tested at 1120 ohms. I soldered some 1/4 resistors together and soldered them to the tap points under the PCB of the controller. I got it down to 805. This gives me a 1.391 factor when I go to program my lvc.

The programming cable Keywin sent me on a previous order was screwed too. Every wire was in the wrong spot. I had to resolder all the wires and match them up with the pins on the controller. I programmed the controller with a 42.5 lvc. Which should give me a newly adjust lvc of about 59.13. With a newly adjusted regen voltage of about 83.46v.

I also set the
Phase Current to 90A
Rated Current to 45A
LVC to 42.5
Tolerance to 1v
I set the speed limit to 60
3 Speed mode to Switch Mode
Speed 1 to 30
Speed 2 to 100
Speed 3 to 120
Indicate Mode to Comm GND
Converse to 25
Block Time to 1
Auto Cruise Time to 15
Slip Current Change to Up 15 Mark
EBS Level to 2

I soldered in connections for speed selection switches.
I have two 3 position switches installed on my handle bar.
I soldered in jumper wires to enable and disable regen.
I soldered in my CA connector.
Soldered a new hall wire connector to match my harness.
Installed a power connector to get away from the spade connectors. (That spark is super scary)
I drilled a hole into the controller case to install a toggle switch for the MCU/CA. (controller power switch)

One switch is for Speed 2(30% throttle) and Reverse

The other switch is for Speed 1(Speed limit 60) Speed 3 (default 100% throttle) Speed 4 (Advanced Timing 120%)

My CA still has a 59v LVC set.

I also setup my Temp Gauge on my handle bars and soldered and shrink wrap the connections. Reads ambient air and motor temp.

[icecube57]

I blew up 2 controllers in the past two months. Learning from my mistakes I ordered 1 more and keywin gave me the shaft. He didnt ship me the new 116 MCU controller. He sent me the older 72v 45A model. I was pissed. It wasnt pimped out at all. At the very least I could have got a CA connector. I requested one in the paypal email. He shipped me a Plain Jane Stock!

I took a deep breathe and got out the iron. I looked at the Infineon Dummy threads and other Technical info on this model controller.

I did the R12 mod. The original Tested at 1120 ohms. I soldered some 1/4 resistors together and soldered them to the tap points under the PCB of the controller. I got it down to 805. This gives me a 1.391 factor when I go to program my lvc.

The programming cable Keywin sent me on a previous order was screwed too. Every wire was in the wrong spot. I had to resolder all the wires and match them up with the pins on the controller. I programmed the controller with a 42.5 lvc. Which should give me a newly adjust lvc of about 59.13. With a newly adjusted regen voltage of about 83.46v.

I also set the
Phase Current to 90A
Rated Current to 45A
LVC to 42.5
Tolerance to 1v
I set the speed limit to 60
3 Speed mode to Switch Mode
Speed 1 to 30
Speed 2 to 100
Speed 3 to 120
Indicate Mode to Comm GND
Converse to 25
Block Time to 1
Auto Cruise Time to 15
Slip Current Change to Up 15 Mark
EBS Level to 2

I soldered in connections for speed selection switches.
I have two 3 position switches installed on my handle bar.
I soldered in jumper wires to enable and disable regen.
I soldered in my CA connector.
Soldered a new hall wire connector to match my harness.
Installed a power connector to get away from the spade connectors. (That spark is super scary)
I drilled a hole into the controller case to install a toggle switch for the MCU/CA. (controller power switch)

One switch is for Speed 2(30% throttle) and Reverse

The other switch is for Speed 1(Speed limit 60) Speed 3 (default 100% throttle) Speed 4 (Advanced Timing 120%)

My CA still has a 59v LVC set.

I also setup my Temp Gauge on my handle bars and soldered and shrink wrap the connections. Reads ambient air and motor temp.

Regen does work at the voltage im running.84 charged. 72v nominal. But after 5-10 minutes riding around my neighborhood. The temp slowly rose until it went past 160F. Thats when the meter shows its HI. I was using the regen function.(Always On. Kicks in when you release the throttle.) I think thats one reason why it rose so quickly. Its dumping all that extra energy into the windings. The regen is alot softer on this controller compared to the one on the newer 116 controller. It slows you down ok on level ground but its not going to aid much on downhill. I wonder does the brake by wire regen work more agressively than the Always On Regen. I will remove the external regen jumper and check the temps tommorrow.

[icecube57]

Today I took the bike for a spin on the trail. Mainly to test the Temp Probe and see how it affects my riding style. I was riding at 96.7v fresh off the charger today. I did 32 miles. Consumed 6.4AH. I took the fatpacks down to about 31.7 under load and 33.25 resting. My 12v 12AH booster pack was at 11.3 under load and was 11.7 resting. I mainly stayed in Speed Limit Mode. It peaks at about 15A and maintains a 21-22 mph @6-7 A perfect for cruising and pedaling.

[biohazardman]

Lots of miles not much juice used so you pedal allot. Looks like you did a good job on the build. Love the temperature probe. ;^) They make some for the BBQ folks that are a wee bit smaller. Not sure one needs the rare, med, or well done though. Wonder if there is a wireless model. Guess I will have to check cuz it does seem a good idea not to roast a spendy motor.

[icecube57]

You also gotta remember at higher voltages you also gain extra wh. Vs at lower voltages where you seem to fly through watt hrs like candy on Halloween to get the same performance. So it seems like I went pretty far on nothing. But when you look at the wh per mile I was averaging 17wh/mi for the whole trip. On the way out I was 13-18wh/mi. I even got it down to 8-9 wh/mi in some areas. On the way back I realized I had over 60% my pack left so i was being lazy and was doing about 25-30wh/mi on the way back. The way my bike is geared I can effortlessly pedal along at 20-22mph. I noticed in some areas i was barely consuming 100w. But for the most part it was a constant 2-7A draw at WOT in the SPEED LIMIT MODE which was real easy on my batteries and easy on my wrists cause I wasnt trying to regulate the throttle to keep a constant speed it. It was like a poormans Cruise Control.

I noticed that while riding @ 85-95 ambient temperature the motor runs in the 135-150 when cruising at a 20mph pace. Under inital acceleration you see a spike and as long as you were steady with the speed the temperature would go back down but if you were playing around with the throuttle it would keep climbing. When in the noramal or timing advance mode where it could get the full 45A it would climb up 5-10 degrees if it was a short burst of acceleration. If it was held constant for 30secs or more... you might as well not look at the Temp Gauge cause it would say HIGH in a minute or so. But if you pulled over for 5 minutes. It would drop back down into the 150s.

The temp reading is coming from inside the motor right under the windings. This would be the highest concentration of heat. The GM does not have a solid stator it just has two thin plates for structure. So there is nothing to absorb and wick the heat away it has to immediately be radiated and cooled throguh the axel and motor side covers or just sit and roast. Even when it got past 160F I could still hold my hand on the motor cover. All I can say is the temp probe is useful especially if this is your main mode of transportation and you are trying to stay in the 20mph constant with you speed without overheating in warm to hot weather.

Looking at the performance aspect of having this... The motor is going to running hot. Probably around 170-180 in agressive commuting. It seems to spike and run cooler/or maintain temp at speed. When you stop It climbs a little bit and plateaus off and starts to drop down. I say this because when I stop the motor temp rises and then drops 5-10F for every 5-10 minutes. So when my meter reads high at 160F The motor cant be that much more warmer for it to cool down and register so quick on my meter.

Torque arms are holding great. Everything is still tight. Im not being cocky (knocks on laminated composite wood nightstand) but I think its going to hold up to the stresses. I would like to use regen. Its just not strong enough at 84v and 96v is just pushing the limits for the controller.

Can anyone tell me if the Always On Regen is weaker than the Ebrake-by wire Regen. If so I will just enable push button regen and get a switch to install on my handle bars. Regen will defintely help with Brake wear but no use of having it enable when it isnt worth a damn. Its just extra heat being generated for nothing.

[Hyena]

Yeah I'd definitely be chasing a smaller neater looking thermometer.
What about something like THIS?
It's much smaller and would neatly fit on the handlebars next to the CA