Pusher Trailer Conversion

[Morgen 3Eman]

Well sometimes I can be dumb as a post.  The 4th termination could be made at the windings.  Getting old is not fun.....

But I still don't get the "less torque with delta" business.

TTFN,
Dennis

[truly_bent]

3Eman;

I did as you suggested and spent some time searching on Wye/Delta motor configurations. The best explanations I've found so far were in the RC community. Specifically, this Brian Mulder guy at http://www.southernsoaringclub.org.za/. In his Electric Motors - Part 2 he goes into some depth on windings configuration. Using his explanations as reference, I bounced around the MP data sheets, trying to make better sense of it all.

The question that kept coming up in my mind was, "Why would anybody configure for Wye when Delta seems to be better on most points?". Delta is more efficiency and has better top end rpm. Brian says that Wye has better torque characteristics, but our data sheets show that peak torque for the 16" only slightly lower (albeit at higher power consumption).

I think the missing ingredient is wire gauge. Thinner wire can't handle the amps as well and because Delta draws considerably more current it's most likely to be wired with a heavier gauge. Heavier gauge wire means fewer turns per tooth and fewer turns means less field flux. Less field flux equals less torque.

The higher the number of turns on a tooth, the greater the magnetic field produced for a given current.  Stronger magnet field results in more torque and lowers the RPM/Volt number.

I don't know if our 16" MP motor is wound with heavier gauge wire or not, but that could account for the reduced torque even though we see more power input to the system.

The 36V data sheets would seem to agree:
28 N.m = 750.64 watts (31.50A) for the 16"
28 N.m = 583.52 watts (23.66A) for the standard

Here, both motors are producing the same torque, but the 16" is costing us more in power. What we lose here in power is gained in speed because the 16" is rotating at 255 rpm versus 199 rpm for the standard.

[Bikemad]

The 4th termination could be made at the windings.

That's exactly where where it is, click on the image below to see it full size and you should be able to see where they terminate.

As I understand it, the MP is a basically a phase synchronous  or stepper motor, driven by a variable frequency drive.

As I understand it, the MP is a three phase AC motor where the timing and switching of the phases is controlled by the hall sensors, but the average value of the voltage (and accordingly the current) being delivered to the windings is actually controlled by Pulse-width modulation (PWM) which basically adjusts the "On" period of the duty cycle in relation to the throttle signal.

Alan
 

[Morgen 3Eman]

Hi Alan and Truly,

I really meant that I should hit Google, but thank you for your research.  I'll never get smart if I have you guys doing all the work!

"Stronger magnetic  field means more torque" was quoted, and so does a stronger current through a given winding.  Higher impressed voltage across a coil drives greater current through the coil, increasing the magnetic field, which means more torque, does it not? 

I'm an old gear head, and it looks funny to read a line where increased power is considered a bad thing......."What we lose in power we gain in speed" !  To me that should read " What we gain in power gives us more speed" is my preferred what to convey that data.

Alan, I never once considered that PWM was still in use for motor speed control!  That makes a great difference in how to look at the question.  Variable freq has so many advantages, but I guess PWM must be really cheap to build. 

Thanks for trying to help me get smarter, guys.

TTFN,
Dennis


 

 

[truly_bent]

3E;
Glad to be of service . I'm trying to make sense of this myself.

Alan;
Just to scratch my curiosity itch, can you find out if the 16" motor is wound with a different gauge of copper? You may have to go deep undercover for that one .

[Morgen 3Eman]

Perhaps the torque is affected by the fact that with delta wiring, when one winding is energized, the other two are also receiving power and generating  a magnetic field.    Wye only energizes two windings with voltage impressed across one phase.  I really need to find some time to research this.

TTFN,
Dennis

[truly_bent]

This Brian Mulder guy shows how that operates as well. He explains that if each winding has a resistance of 1.0 ohm, then Ohm's law proves the total resistance of the network to be 0.66 ohm.

As you suggest, the winding between the two conducting points has a strong magnetic field while the other two have lesser fields - probably on the order of 1/3 strength or thereabouts. All the windings therefore are contributing to the resulting torque.

I didn't realize this myself, until I read through his tutorials.

[Bikemad]

As far as I am aware, each phase wire connection is either switched High (battery +ve) or Low (battery -ve) in relation to the magnetic field adjacent to its respective hall sensor (using PWM to control the duration of the "On" period of the duty cycle).

To make it easier to understand, let's simply assume that the "On" period is 100% of the duty cycle (full throttle) so full battery voltage is being switched.

With star (wye) windings, all three windings will be energised at any one time, one in series with the two other paralleled windings.

Using High/Low switching with Delta windings means that only two windings can ever be energised at the same time, because the third winding will either be High at both ends or Low at both ends and will therefore have no potential difference (voltage) across it.

I found a great animation that demonstrates the basic operating principal of a three phase star wound motor that can also be manually overridden by moving your cursor over the boxes.

Alan
 

[Morgen 3Eman]

Well, how much dumber than a post can I eventually become?  Clearly you can get current flowing through all three windings of a wye, and only two windings of a Delta.    But there is actually a third value that a phase wire can attain, and that is open circuit. 

To bring a particular phase wire to battery +, the upper FET in its H switch is turned on.  To bring that wire to battery -. the lower FET switch is turned on.  With both FETs off, the wire is floating,  and no current flows through the associated winding in wye configuration.  So while the motor pulses are overlapping, current will flow through all three windings, and during the time a single phase is on only two windings get power.  And the other way round for Delta. 

At part throttle, the phase wires can be at open circuit a significant  portion of the time. 

TTFN,
Dennis

[truly_bent]

Well, it's all moot now anyway.

I opted not for the trailer, but to electrify just one of my bikes. They both have 26" rear wheels, so i've ordered a 26" MP II. Not sure which bike will end up "electrified" at this point, but that'll work itself out.

I guess it was the extra $500+ for the BOB Ibex that killed it. Still, the setup i'm going to end up with isn't too shabby to start. Plus the MP II there's:
- Lyen 184110 MIIR 65A controller (does everything except perk coffee apparently)
- 48V 20AH LiFePO4 battery
- Large screen Cycle Analyst
- Various other bits necessary to glue it all together

The controller and Cycle Analyst arrived today. All the other pieces should be in within a week or so. Because I wasn't sure what voltage I wanted to go with, Edward modified the controller to accept any battery pack from 36V to 88V. Absolute max voltage is 110 before the FETs fry.

Another recumbent phreak I met recently wants to build a AZ Cycle Bully as well. The pressure's off now so we're planning on making it a summer project - maybe finishing in August.

In the course of this discussion i've learned quite a bit about these brushless motors and want to thank those who contributed. I'll post a picture when one of the bikes is up and running.

[truly_bent]

It Lives!
True to my word, I am posting pics of my little project.

Just powered up today. A little trouble with the throttle connector but she's okay now. I've just finished downloading parameters into the controller and she's ready to rock.

For any of you who can't see too well, the donor bike is a Burley Canto recumbent.
- Cycle Analyst on the bar with cruise buttons and a 3 speed range selector.
- Brakes are linked to the controller for regen
- 6 speed gear cluster with new speed selectors
- Lyen 18 FET controller
- One torque arm installed (lost the bike stand in the process)

That huge pannier bag on the back contains a 48V 20AH (LiFeP04) battery, and a bunch of foam to keep it centered. I have another Arkel bag for the other side. I went with a removable bag for the battery so I could carry it and the charger separately.

Obviously, a little more work needs to be done to improve water resistance of the cable connectors. Actually, it's not as bad as it looks. Everything is pretty well isolated with heat shrink tubing. I'm looking into plastic junction boxes to find something suitable.