[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: (ET) motor ratings



Interesting comments and thoughts..... figured I'd add some more.
   When METI designed and built the "modern" Elec-Trak a few years ago we 
used a programmable GE golf cart controller from FSIP (Flight Systems in 
Reading PA).  It was designed by a friend of mine and I was able, with his 
help, to program the E-20 motor into it and it worked well.  METI sold 
some repower kits which were ready to go at the time.  METI closed down 
due to the 8 founders all being remotely located and the cost of trying to 
manufacture a great machine cheap enough for a limited market.  It may be 
starting up again by one of the founders in Minneapolis.
   At any rate, it is a good controller but did not regenerate but would 
field pulse to hold the tractor on a hill at standstill.  Once you hit the 
accelerator you were at the mercy of the gas pedal and brakes.... I've 
learned to deal with mine and have not had a single issue in 4 years of 
using it.  I can hook my laptop to it and watch volts/amps/field... 
etc..... but, to use it "wisely" you need a tach on the motor and be able 
to program it and I'm not sure that (programming capability) is available 
to the general public.  Remember, it is a high amp golf cart controller 
(36/48 volt too).  They program golf carts for hills or flat courses by 
varying current ramp slopes and limits.  GE/FSIP also makes EV controllers 
and a regenerating one but it is also high amp capability and overkill for 
an ET.  On the METI tractor we used a  motor which was larger and had much 
more torque but we also had a locking differential if you wanted to 
impress your friends with wheel stands or tractor pulls.
   Series motors lack finesse and control that shunt or compound motors 
have.  You get more use of a shunt/compound motor because of field control 
and field weakening for higher speed.  Controllers are different too for 
that reason.
   The simplicity of the original ET control to regenerate is harder to 
duplicate with a solid state control because you need a reverse bridge 
also.
FSIP has a ton of experience in EV controls and is a good source of 
information -  www.fsip.biz

...Walt

-----Original Message-----
From: Brian E. Haines [mailto:bhaines nycap rr com] 
Sent: Saturday, December 29, 2012 9:56 PM
To: elec-trak cosmos phy tufts edu
Subject: Re: (ET) motor ratings


Yeah, I too would agree with David's post.  An all around controller, that 
would work with any ET motor would be a bear to create.

If I were to make a few suggestions:

     1) Design a PWM controller for E15, and E20 class motors.  The 
controller would be identical between E15 / E20, and would cover a large 
subset of tractors.
     2) Integrate a 36VDC->12VDC circuity for the lights/lift/misc.  
Besides being really handy it would also help issues with taping mid 
string that could cause power to flow back into the controller in strange 
ways.
     3) Integrate a second PWM circuit for 36VDC-> 120VAC .  It wouldn't 
have to be that robust, a modified sine wave, with a max of 10a....That 
would be MIGHTY handy.

There have been posts of using expressPCB to create the first two (above). 
 I can find a link to a pure sine inverter design document(If anyone wants 
it).  It had the theory and design notes. Sadly my skills don't go quite 
that far.

> Someone with more chops may blast me out of the water on this one, but 
> I wonder if it might be possible to design a microprocessor controlled 
> controller for which the firmware was user-uploadable.  Then all the 
> manufacturer would have to do is develop a new field map or control 
> algorithm for each motor the users wanted.

A NAND flash could do that.
http://www.micron.com/products/nand-flash

Run a tiny embedded linux core and read/write the settings to your heart's 
content. :)  But that's the programmer in me talking.
B
On 12/29/2012 1:31 PM, David Roden wrote:
> On 29 Dec 2012 at 11:23, Bill Alburty wrote:
>
>> We need a controller which will work on all types of DC motors; 
>> namely, series wound, shunt wound, compound wound, and all the other 
>> combinations.
> I'm not an engineer, but my impression is that this is a pretty tall 
> order.
> Pairing a motor and controller is not a trivial process.
>
> In particular, separately excited motors need a customized field 
> current profile, different for each motor.  You may be able to find a 
> compromise that will work with several similar motors from the same 
> manufacturer (the Alltrax ET controller works OK with most ET 
> wound-field drive motors, AFAIK).  But I think that once you get into 
> more dissimilar motors, the controller's designer has to consider each 
> motor's behavior separately.
>
> Series motors seem to be more forgiving, but you can still run into 
> compatibilty problems, even beyond whether the controller can handle 
> enough current to run the motor in its intended use.
>
> For example, certain Curtis controllers need more inductance than some 
> Advanced DC motors provide.  Their current limiting doesn't work right 
> with those motors unless you add a series inductor.  This is why more 
> recent Curtis road EV controllers reduce their chopping frequency to 
> the audible range at low duty cycles - it gives the current limiting 
> circuitry more time to respond.
>
> Someone with more chops may blast me out of the water on this one, but 
> I wonder if it might be possible to design a microprocessor controlled 
> controller for which the firmware was user-uploadable.  Then all the 
> manufacturer would have to do is develop a new field map or control 
> algorithm for each motor the users wanted.
>
> However (here I am attacking my own idea), doing somethng like this in 
> a timely manner - so you weren't waiting months or years for your 
> firmware - would, again, be a nontrivial process.  It'd require a long 
> term commitment from the designers.  And at engineers' salaries, I 
> don't think it would be too compatible with a price that most of us 
> would want to pay.
>
> It would be great to have someone design such a controller as a 
> volunteer, but would he want to support it long term, generating yet 
> another profile for yet another user for years to come?  I think the 
> best chance for such an enterprise would be if we could find a large 
> community of qualified developers to work on the project, something 
> like some of the Linux variants have.  Maybe with open specs, we could 
> develop a user community to support such a controller.
>
> I question whether we have enough EEs in the ET world to make such a 
> project fly.  Ffor this to work, I think the controller would have to 
> also be applicable to other machines more common than the ET. Maybe golf 
> cars?
> NEVs?  What else?
>
> The market size is an ongoing problem with ETs, just as it is with 
> hobbyist road EVs.  I suspect that the reason we have the Alltrax ET 
> controller at all is that Steve Richardson owned an ET, worked for 
> Alltrax, and wanted a modern, customized controller for his tractor.  
> Given the market size, I can't imagine Damon making it a priority 
> otherwise.
>
> While it would be great to have a DIY / open source option, I would be 
> very impressed if a controller of any kind, DIY or otherwise, could be 
> made to support "all types of DC motors" - especially while costing in 
> the hundreds rather than the thousands of dollars.
>
>
> David Roden - Akron, Ohio, USA
>
> = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
> Note: mail sent to the "etpost" address will not reach me.  To send me 
> a private message, please use the address shown at the bottom of this 
> page : http://www.evdl.org/help/ = = = = = = = = = = = = = = = = = = = 
> = = = = = = = = = = = = = =
>
>
>
>
> _______________________________________________
> Elec-trak mailing list
> Elec-trak cosmos phy tufts edu
> https://cosmos.phy.tufts.edu/mailman/listinfo.cgi/elec-trak
>


_______________________________________________
Elec-trak mailing list
Elec-trak cosmos phy tufts edu
https://cosmos.phy.tufts.edu/mailman/listinfo.cgi/elec-trak