E-Bike XB-502 conversion project

[ElectricGod]

More on the RV-120-regular motor...

I bought a 3 jaw chuck so I can hold round objects. I could have gone with a smaller chuck, but the theory is that I can hold large stuff now. I was limited to whatever could fit in the jaws of the cross vice or across the face of the rotary table, Now by holding an ID, I can work on stuff a bit larger.

I made a mounting ring for the chuck so I can use the T-slots to mount the chuck, but I need to work out a few details before I can use it. For now, I have mounted it with neo magnets, which means everything is magnetized. I was constantly blasting it with air to get rid of the magnetized metal bits.

This shaft to sprocket adapter originally had a 10mm ID and 3/4" OD. Once I had the chuck centered on the table, I used a 1/4" mill to increase the ID to 15mm. Now it's more of a sleeve with a gap in it than an adapter.

I'll have to mill a keyway in the shaft and make a rectangular key that spans the entire gap.

 

[ElectricGod]

I went for a test ride now that it's all back together. Everything is working well again.

I should try switching the motor to delta and dropping a tooth or 2 on the motor sprocket. This may get me the acceleration I want and still retain the top speed.

I'll make a motor bracket in the next few days for the RV-120, do some shaft modding and I'll be ready to install it on the EV.

 

[ElectricGod]

I spent a good bit of time today refining my turn table and 3 jaw chuck. While still not as good as precision machining can do, this is acceptable for anything I'll need and doable on a drill press!

As a test of how close I am to dead center between the turn table and 3 jaw chuck, I milled down the OD and got this thin wall that's .5mm thick. Running a dial indicator around the OD of the part while turning the table reveals 2-5 hundredths run out. Most of that is in the rotary table. I have tried to remove all the slop and movement in it, but the table top is not perfect. There's a tiny bit of wobble in the rotating portion. A little bit of the run-out is in how well I centered the table and the chuck. A little is in the self centering 3 jaw chuck. Still...not too horrible for milling work on a drill press.

With things being much more stable now, I get a far better finish.

The drill press quill is less than spectacular and has some amount of movement and slop in it. This allows mills to chatter and if that happens they can self destruct. This was a good mill until it started to chatter. That's $30 of worthless carbide!

I finished the ID with a 1/4" end mill to 15mm. The results are far superior to my previous attempt.

The sleeve wall now has .01mm variance and the finish in the ID is nearly perfect. With the improved stability, I can cut deeper too. The new one is 5/16" longer than the first attempt.

 

[tylerwatts]

That's impressive work with the equipment. Very good setup you've pulled together, well done.

Sent from my Moto G (5) Plus using Tapatalk

 

[ElectricGod]

That's impressive work with the equipment. Very good setup you've pulled together, well done.

Sent from my Moto G (5) Plus using Tapatalk

Thanks1

 

[ElectricGod]

I did this mod on the inrunner on this scooter a long time back. I brought out all 6 phase ends so that I could mess with WYE/delta switching. So far on this scooter I've been running it in WYE. Delta will increase the motor Kv from 64 to 109 or 1.7X.

For experimentation I was using a 40 amp 3 pole relay to do the switching, but the relay is too weak to run the motor under load and I really need a sealed relay that's filled with nitrogen. These types of relays are quite expensive so I never bothered making this happen.

Anyway, now I want to try it out in delta and of course I still don't have an appropriate relay. Last night I made these so that I can run it delta. I made three 14 awg copper Y things to go inside the 5.5mm bullets. Solder probably would have been OK, but I trust copper more. The motor phase ends are terminated in male 5.5mm bullets. Each of these has 2 female bullets to connect the phase ends together.

 

[Ianhill]

Be prepared for huge start up current, I like the connectors I done simular to parallel two cells.

If you dont have a temp probe in the windings now is the time to do it your going to be giving the motor a world of pain in high load situations like full throttle start on a steep incline bang delta start have that it won't be repeatable to much for sure.

 

[ElectricGod]

Be prepared for huge start up current, I like the connectors I done simular to parallel two cells.

If you dont have a temp probe in the windings now is the time to do it your going to be giving the motor a world of pain in high load situations like full throttle start on a steep incline bang delta start have that it won't be repeatable to much for sure.

I'll drop a tooth or 2 on the motor sprocket as well.
There's temp sensors on it and in it.

I already know this motor "likes" WYE better than delta. When I was bench testing it, it would build up heat in the lams more quickly unloaded in delta than in WYE. It does have .5mm lams. Also, I originally intended this scooter to run at 131v, but the motor was getting far too warm under no load at higher than 82v so I never went to 131v in the build. HLD inrunners are not exactly awesome motors. They do the job, but there are better options.

 

[Ianhill]

Motors with iron focusing the linkage tend to have voltage as its limiting factor unless it can control it's eddys in an impeccable fashion or the magnet ring is shite so they fling off instead.
Where as a dual halbach array ironless axial will have RPM as its limiting factor how fast can it spin before it self destructs and that's going to be right up there but the down side is less EMF so it can suck alot of juice unsupervised.
http://www.launchpnt.com/portfolio/transportation/halbach-electric-motor

I'll be honest the freerider has been the bike of my life it takes me anywhere the 12kw is only used at start up at speed even up inclines it stays below 8kw the only time 12kw gets pulled is stationary massive hill back wheel in grass so I get a huge power draw that calms once at 2 or 3mph.

Monitoring motor behaviour to extend range has been my lesson past few weeks I can get 20 miles of harsh riding at 10 mph going up large inclines with bogs and rutts etc or do the same 20 miles at flat out 35mph on dust tracks, I prefer the slower rides I get alot more time and have a more technical ride.

I've stalled on my scooter builds the sun came out.

 

[ElectricGod]

Motors with iron focusing the linkage tend to have voltage as its limiting factor unless it can control it's eddys in an impeccable fashion or the magnet ring is shite so they fling off instead.
Where as a dual halbach array ironless axial will have RPM as its limiting factor how fast can it spin before it self destructs and that's going to be right up there but the down side is less EMF so it can suck alot of juice unsupervised.
http://www.launchpnt.com/portfolio/transportation/halbach-electric-motor

I'll be honest the freerider has been the bike of my life it takes me anywhere the 12kw is only used at start up at speed even up inclines it stays below 8kw the only time 12kw gets pulled is stationary massive hill back wheel in grass so I get a huge power draw that calms once at 2 or 3mph.

Monitoring motor behaviour to extend range has been my lesson past few weeks I can get 20 miles of harsh riding at 10 mph going up large inclines with bogs and rutts etc or do the same 20 miles at flat out 35mph on dust tracks, I prefer the slower rides I get alot more time and have a more technical ride.

I've stalled on my scooter builds the sun came out.

Halbach can be applied to just about any motor. I've run across a few outrunners that were made using this magnet effect.

Ironless motors are cool, but rewinding one has got to be a PITA. I've looked at Launch Point before.

I know the HLD will run at 82v in Delta and not sling magnets. I bench tested it already and in real world under load, it will never see those RPM's anyway.

Stalled...that's becasue you are out riding the Kuberg!

 

[ElectricGod]

Now that I have the delta connectors made up, I could remove the WYE connector and take a picture of it.

Converting to delta was not hard. It took 2 tries at phase ends to get it right. It took longer to take off the various cowlings to get to the connections than anything else. While I was in there, I rerouted the phase wires to make them more accessible. I also changed out the 16T motor sprocket for 15T. I'd have gone to 14T, but I don't have one.

 

[ElectricGod]

It's all back together again and tested out. I'll get a couple of lower toothed sprockets to try out soon.

Tomorrow I'll see how it does on the 15T.

 

[ElectricGod]

I went for a short ride today. 15T wasn't close to enough. The motor is bogged down way too much and so acceleration is significantly weaker than before. When I put on the 15T last night, I was already 99.9% sure without doing any math that it would not be remotely enough to compensate for the increase of 1.7X more Kv. I've purchased a 13 and 14 tooth motor sprocket from Azusa Engineering and they will arrive in a few days. They won't be enough either on a 66T wheel sprocket. I'll need a larger wheel sprocket to make this work well.

Some basic math for calculating the correct ratio:

On WYE I was running 16T to 66T or 4.1:1 and topped out at 60mph. I want to still get to 60mph.

WYE: 64kv x 82v = 5248RPM
Delta: 109kv x 82v = 8938RPM

1 mile = 63360 inches
The back tire is 58 inches in circumference around the tire.

63360"/58" = 1092 tire rotations to go 1 mile.

60mph = 1 mile per minute (MPM)

1MPM x 1092 rotations = 1092 RPM at the wheel to get 60MPH.

8938 RPM/1092 RPM = 8.2:1

At 66T on the wheel I'd need an 8 tooth motor sprocket (66/8.2 = 8T) and I know for a fact that 12T to 18T is the range for 219 motor sprockets.
NOTE: Since sprockets only come in whole numbers of teeth, I round up or down to the next nearest tooth.

12T x 8.2 = 98T wheel sprocket
13T x 8.2 = 107T wheel sprocket
14T x 8.2 = 115T wheel sprocket
15T x 8.2 = 123T wheel sprocket

I have 88T, 108T and 117T wheel sprockets.
88T/8.2 = 11T (not possible)
108T/8.2 = 13T
117T/8.2 = 14T

I'll be using the 117T wheel sprocket since it gets me the most "range" up or down a few motor sprockets.

With the 15T on the motor now and the 117T wheel sprocket, that's 7.8:1. This might be OK.

Another way I could have done this is is take the gear ratio at WYE or 4.1:1 and multiply by 1.7 to get 7:1 for delta. Since what is on the scooter now was a best case scenario for the sprockets I had on hand at the time, the results multiplying by 1.7 may not be as good. Still, with these numbers, the 117T wheel sprocket gets me the best range for motor sprockets. Also, I want to get a bit more acceleration, which was a big reason for for switching to delta.

With those same sprockets and 7:1...
66T/7 = 9T (not possible)
88T/7 = 13T
108T/7 = 15T
117T/7 = 16T

https://azusaparts.com
IF anyone needs larger 219 sprockets, contact Asuza Engineering. They will make them for you at a very reasonable price. I got mine made with standard KART 6 bolt pattern. I got 2 108T and 2 117T made.

KART adapter vs 88T vs 108T:

KART adapter vs 88T vs 117T:

108T vs a 16" tall tire:

117T vs a 16" tall tire: My back tire is 19.5" tall. This won't be a problem.

 

[ElectricGod]

The 117T sprocket is in place now. I'd go test ride tonight, but it's snowing.

This is pretty much one of the smallest 219 sprockets available vs a very large 219 sprocket...66T vs 117T

It's big, but it fits. I kind of wish it had sections cut out of it to take away a bit of weight and to let you see through that giant sprocket.

With a very large sprocket like this, you discover if your alignment is really straight or not. I discovered I needed to move the sprocket inward slightly and that my back wheel was angled to the right a tiny bit.

 

[Ianhill]

You got snow we have had an early spring been around 15c all week not a drop a rain.

Looks awesome that's gonna chug like a gooden

 

[ElectricGod]

The snow melted off this afternoon so I went for a ride.

I'm possibly going back to WYE...

I think I have the phase pairing wrong and that may fix the motor issues. Under load it doesn't run like it should.
I still need to drop a tooth on the motor sprocket to get the gear ratio right.
Probably between these 2 items, it will be fine.

The 117T sprocket acts like a speaker. The scooter is way too noisy now. With the previous set up, you got some motor sound and otherwise it was really quiet. I know my chain alignment is dead on now so that's not it. The noise coming off that rear sprocket is pretty significant. I may need to pull it off and see about milling away portions of it so it doesn't act like a sound board.

 

[ElectricGod]

I took the scooter apart tonight, switched out the 15T driver for a 13T.

I also swapped the starts and ends of the phases for the other combination I haven't tried that runs the motor.

Running under no load, the back wheel seems slow. Swapping in a 14T or 15T motor sprocket is pretty easy so if the motor runs loads better now, that will be next.

I'll do a test ride tomorrow.

 

[ElectricGod]

Went for a test ride...
I successfully failed at making any more power or speed. This basically proves what I saw bench testing. The HLD inrunners don't do as well in delta as they do in WYE.

Switching the phase end combinations fixed the semi-smooth way the motor was running on the last test run, but it still under performs compared to WYE. The motor got hotter quicker, the controller got warmer and the scooter was slower in every way. I'm under geared right now by 1 tooth. It ought to have extra torque and it doesn't. I topped out at 35mph.

Conclusion: Just don't ever run an HLD inrunner wired delta.

Time to switch back to WYE and put the original gearing back in place.

Was it a waste of time? Nope...I learned a few things about this motor and I got to mess with my EV...so it was worth it to me.

 

[ElectricGod]

I still have the over sized sprocket on the back and it's geared less than optimally, but it has loads of torque right now. I'll go back to a more standard sized sprocket later. I'm charging right now so I can do a day ride tomorrow. My neighbor rode it tonight, he says it gets up speed faster than his 175hp bike does, but then I run out of motor RPM's at about 43mph.

This thing definitely needs the RV-120 on board.

 

[ElectricGod]

As usual...get it all running and then decide to mod it again. This scooter ahs never spent more than a few days working before I decide to spend hours and hours on some upgrade or change for it. I think once it has the RV-120-regular in it, that then I'll be ready to call it done.

I've pulled off the back wheel and gone back to the 66T wheel sprocket.
I've dropped the swing arms so I can get access to the space were the motor mounts so I can install the RV-120 in place of the HLD inrunner.
The RV-120 is 50kv and the HLD is 64kv so I'll need to do some re-gearing too.

 

[Dui ni shuo de dui]

If your only problem is top speed, can't you find a way to do flux weakening? I don't know what controller you're using, so maybe it's not possible on yours, but I think that's the only way you could have both torque and top speed without having to use a mechanical gearbox.

 

[ElectricGod]

If your only problem is top speed, can't you find a way to do flux weakening? I don't know what controller you're using, so maybe it's not possible on yours, but I think that's the only way you could have both torque and top speed without having to use a mechanical gearbox.

Flux weakening to be very effective needs lots of exposed iron on the armature. An IPM inrunner is best for FW. Otherwise FW has a small effect. The controller can do FW, but it's not on since it provides a small increase in motor RPM. FW increases phase current and reduces motor efficiency.

There's a third way to get lots of torque and top speed...a bigger motor...and as it turns out I have one to use.

 

[Dui ni shuo de dui]

Flux weakening to be very effective needs lots of exposed iron on the armature. An IPM inrunner is best for FW. Otherwise FW has a small effect. The controller can do FW, but it's not on since it provides a small increase in motor RPM. FW increases phase current and reduces motor efficiency.

Ah yes, I somehow forgot it was an inrunner...
Any idea how much would be the increase of rpm if you turned it on?

 

[ElectricGod]

Flux weakening to be very effective needs lots of exposed iron on the armature. An IPM inrunner is best for FW. Otherwise FW has a small effect. The controller can do FW, but it's not on since it provides a small increase in motor RPM. FW increases phase current and reduces motor efficiency.

Ah yes, I somehow forgot it was an inrunner...
Any idea how much would be the increase of rpm if you turned it on?

The outrunner has 50kv vs 64kv for the HLD inrunner. However the outrunner is a good bit more powerful. If all I did was get my top speed back at 60mph, I'd have a much stronger EV. That's all I'm shooting for. I don't need it to go faster than 60 mph....but that could be fun!

 

[kiwifiat]

If your only problem is top speed, can't you find a way to do flux weakening? I don't know what controller you're using, so maybe it's not possible on yours, but I think that's the only way you could have both torque and top speed without having to use a mechanical gearbox.

Flux weakening to be very effective needs lots of exposed iron on the armature. An IPM inrunner is best for FW. Otherwise FW has a small effect. The controller can do FW, but it's not on since it provides a small increase in motor RPM. FW increases phase current and reduces motor efficiency.

There's a third way to get lots of torque and top speed...a bigger motor...and as it turns out I have one to use.

With the Lebowski inverter FW works very well with both in SPM and IPM motors. In this video we see a 100% increase in Kv with FW.

[youtube]https://www.youtube.com/watch?v=57kRSoOrTxw?start=1018[/youtube]

On my scooter that is fitted with a gen1 QS273 motor on 18S I get a 91% increase in Kv with Lebowski's recommended setting of 70% FW.