e-Buggy: Alternator Conversion

olaf-lampe said:
IIRC the timing advance is depending on momentary RPM and load: the higher the RPM the more timing advance is necessary for the best output.
I believe the sensored infineons may have an auto-advance algorithm, because I felt some small vibrations comming from the motor under certain conditions and after a while they're gone. Seems it takes a bit time to find the right 'commutation angle'

The worst thing is too much advance, it's heating up the motor and controller.
-Olaf

Just finished reading the paper from Krumpi which give me wild ideas, given a non auto advancing controller like mine, I think we could have a very advanced fix hall sensor and separate controller to introduce timing delays, which then could dynamically changing the delays based on RPM 8)
 
I did some "research", too. Seams that advancing does the same thing as field weakening, but all of them suggest that field weakening or advancing should be done to reach speed above nameplate value, if maximal torque isn't needed. The principle of advancing is - it increases "opposite" BEMF so field decreases and speed rise, same like with the field weakening, but I think FW is much simpler. They say that with FW can increase speed 3 times! Of course, if torque is satisfactory. But, all of them state that FW should be done when voltage reaches max, or in our case 100% duty cycle. So, we should use FW only to reach higher speed then we can without that? I can't certify that, but can you widodo drive some speed with full field current for the test purposes, and then drive the same speed with FW, measuring current drawn from battery, if you drive field and stator from the same? I'm interested in power economy, if we decrease field, stator current rises and speed too, but can we pull less current under 100% (e.g. 50%) duty cycle for the same speed? :?: :roll:
 
I did some more testing this morning, my alt gets really hot when run no load, almost to hot to toutch. I use It at 40v, so maybe there is too much losses in the the iron from high speed? If it is the problem, I will use more field to keep rpm low, but this reduces eficiency in the motor :| at 12v field it takes 250w (whithout field power), and at 5v it takes 120w. I'm more after high torque than high rpm, and anyway my controller loose sync at high speed, I need to find a tackymeter to test that...
 
Windtalker, I found that my Rotor saturates at around 5-6 volt. Hence, I supposed after that you will only gets temp increase rather than flux increase. Still your experiment with 40V supply really draw my interest (My ESC only supports 4S), Please do update us on the top RPM for a 5V rotor, and 40V stator if you could have access to tachometer :D.

anyway, is there some constraint which nullify a reduction gear for higher torque in your case ?
 
I plan to use this alternator on a velomobile, whith a moped wheel on the rear. It would be easy and convigent to use a single stage gearing, and mopeds wheels usualy use 52thooth or so sprokets, 11t front, and they are about 22"-24" od.
 
At 48V, field connected to separate 12V bat, my setup draws 3.05A. Axle braked with pliers over leather glove (to protect surface) draws not more than 8A, I can't stop it no matter how hard I press. Well, it's on bare axle, 17mm, I know that torque is directly proportional to radius. I have to find another way to load motor. After 10 minutes interrupted testing, rotor gets warm, stator is cold, but it's mainly no-load run. With saturated rotor, it turns at 5000 rpm.
 
krumpi said:
At 48V, field connected to separate 12V bat, my setup draws 3.05A. Axle braked with pliers over leather glove (to protect surface) draws not more than 8A, I can't stop it no matter how hard I press. Well, it's on bare axle, 17mm, I know that torque is directly proportional to radius. I have to find another way to load motor. After 10 minutes interrupted testing, rotor gets warm, stator is cold, but it's mainly no-load run. With saturated rotor, it turns at 5000 rpm.

8A is .5HP and will push a bike very nicely on the flat. I find the work you fellows are doing on the altermotor very exciting. It may not be 99% efficient, but it is one hell of a bang for the buck. If you can design an inexpensive controller that will reduce the field V as the rpm of the rotor increases, you will revolutionize the low cost bike motor. Just the manual pot on the field, is great for a cart or ATV.
Keep going! :D :mrgreen: :D
 
Last night (on work :D ) I've found interesting patent related to automatic field weakening, using discrete electronic parts (it's from 1980.!). Think that widodo is using something similar, but with processor.
 

Attachments

  • US4206389.pdf
    281.7 KB · Views: 415
krumpi said:
Last night (on work :D ) I've found interesting patent related to automatic field weakening, using discrete electronic parts (it's from 1980.!). Think that widodo is using something similar, but with processor.

:shock: Wow, Unfortunatelly, I was still 10 years old that days and the Internet was only available back in 1990 here in Indonesia. Anyway, had 0 knowledge about BLDC until I've read Ron van Sommerens posts and links, Great stuff & Great Guru, I suppose there is where all the ideas coming from, my credit to him.

Will have our AGM coming within weeks, hence will have to ceased out my hobby for a while.. :( , appreciate if everybody could keep the thread alive with any conversion progress.. :wink:
 
You are 1year younger then I am! :eek: I thought about the principle of field weakening used in that patent - compare BEMF and referent value, and if difference is small, field will be weakened, but if difference isn't so small, field will be increased to gain required torque.
 
Any updates on the subject? I played more with my alt, and going from delta to y dropped the no load to 90w. rpm at 40v is 2500, tested with the CA speedo. Anybody managed to fit halls to alternators? Because sensorless start-up is no good for a 40kg vehicle!
 
Funny , I was just about to give this alternator topic a kick myself :D because it was awfully quiet here...

My own starting of the project is going slower than expected. Tight on money , summer aproaching , bills to pay . Money , money , money that's what it's all about.......
Finally I couldn't take it anymore and ordered the bare minimum of components 3 days ago to get me going atleast and try to build something primitive that will turn the alternator !

Microcontroller , 4 fets (3 for stator , 1 for field) 4 optocoupler gate drives hcpl3120 and some small stuff...

Now I should be able to make something rev with some very conservative pwm settings but will try to build a microcontroller controlled Fet driver for the Field varying voltage , that will certainly be of interest to people here :)
Years of no BASIC programming , see if I can still do it haha.

Progress report soon......
 
Oh well forum-bandwidth isn't costing me any money , I'll shove a big boot under it and kick it up yet again.

Widodo is probably still waiting for the AGM to arrive .... :wink:
zenon , salty9 , Harold in CR , krumpi , dajsinjo , magnusvr , bzhwindtalker :

Any results , updates , wants or needs or expectations in the alternator stuff department ? What do you think is the biggest obstacle for any practical usage of the Alter-motor ? I think it looks like the controller to drive this thing....
Well we are working on that , hopefully we will have an open-source CHEAP , simple to build , copied or purchased controller soon .

My own efforts at the moment : gate drivers are back-ordered ...no gate drivers no driving Alter-motor...I am busy studying programming the microcontroller for now.

>Widodo , if you happen to read this some time , how does your programmed commutation cycle work fom A to B ? (see my other post)
 
bzhwindtalker said:
... Anybody managed to fit halls to alternators? Because sensorless start-up is no good for a 40kg vehicle! ..

I did, while it was on the advance side and on a fixed location, my Car ESC still working OK, thought you could clearly see that it now running warmer. You could see how it was done from previous page

Fligh high said:
Widodo is probably still waiting for the AGM to arrive .... :wink:

Survived in our AGM, but our CFO wasn't, so I did ends up with both operation and finance in my hand :x :x , lots of things need to be settled within a month or two, hence had to put my new BOSCH 120A on hold until at least end of july.

Congratulation on you first step, will try to help you as far as I could. I haven't got access to my PC for the ESC logic which I will be glad to share it up with you. Nevertheless, for the commutation switch configuration, I'm using a look up table and used hall A,B,C as a three bit index on which configuration had to be used. Due to limited PWM output in AVR8535, I only used PWM on the High-Side, while the low side was strictly turned on by direct TTL output.
Will try to elaborate more of this in your thread. :D
 
Only thing I can say is, I have a small Alternator (35A) and bought some hall Sensors. Went to the local Hobby Shop, when I was in Fl., and the guy knew nothing about what I asked. Prices of the controllers went from $15.00 to $125.00. Not knowing exactly what I needed, I decided to just sit tight and see what others come up with.

Got 500 watts of Solar Cells to build into a panel or 2 or 3, so, money for that and batteries has me tapped for now.

I build furniture and do other odd things for people here, and want to start building a long cargo type Bike. Building frames would be my niche. I have all the machines I need, sitting in storage in Fl. Just need to get it all shipped down. John in CR and I have discussed maybe doing things together might be a good start. See where that leads us. Altermotors seem to be good torque motors, so, pushing families around in a Cargo Bike might be just what is needed here.

Yeah, It doesn't bode well on my part, but, with the Worlds economics in the toilet, I'm concentrating on survival mode, for when it all comes crashing down. At least we will eat, have battery power in the house, and hopefully not be over run by the masses that don't believe things are as bad as what I think they are.
 
hey widodo, I saw your sensored set-up, but I know nothing about sensored rc esc usage, and i thougth that they needed only 1 hall sensor... I will study your set-up and maybe buy some halls.
 
After some more testing, I think I will ditch the Alternator idea:
First motor on the trike did not worck at all (Burning about 500w without any noticable speed !)
2gnd motor (delta driven) produced SOME power at 1000w in but again most of that turned into heat.
On the two motors, the controller had some pain to keep sync, and I needed to ramp up gaz really slowly, otherwhise it just jitters and make heat...
So I think I will go ahead and just spend some $$$ and get a big hubbie to get my trike rolling on E-power :p
 
Just this morning, picked up some new info on placing Pill Magnets on the face of the S rotor side, so the Halls can be set off by them. Looks like this is finally the way to go. No gouging out the laminations and getting the timing exactly equal.

Info is over on the RC Forum.
 
Any progress anyone? :)

Zenon
 
I have just joined and I am facinated by this post.
Can someone pls help me wit a step by step procedure for a nippon denso alternator conversion?
Pls note that I am a layman and I do not understand much of the nerd information. So pls make it as detailed as possible
Thanx
 
I'm beginning my own alternator-motor conversion project and very much appreciate all the information everyone has contributed here!

I have a junkyard alternator and a cheap Ebay ESC on it's way, and a fair amount of microcontroller and electronics experience, so hopefully I'll be able to contribute some value here soon.

My primary objective (other than learning about these things) is to build a drivetrain for an e-bike something along the lines of the one described in this Instructible: http://www.instructables.com/id/48V-Electric-Flat-Tracker/


jjg
 
Back
Top