Author Topic: Our Test Mule  (Read 57783 times)

Offline mizlplix

  • Administrator
  • Senior Member
  • ****
  • Posts: 1445
  • Country: us
  • "The more you look, the more you find."
    • The old  build thread....
Our Test Mule
« Reply #30 on: February 09, 2013, 02:35:28 AM »
___________________________________________________________________________ ________________________________
              Back in the saddle again

The contactor came in early and I got it installed.  The original
 contactor is good for 500 Amps continuous and 1,000 Amps
 momentary loads.  What it can not deal with is any sort of
current loads during opening and closing.  When it has any
load and it is closing, it tends to weld the contacts solid. On
an opening loaded condition, it arcs the points badly. (Even
 though it is in a vacuum)

There are several solutions to this.  One is built into the
Curtis controller as a "precharge" feature.  The second
is to provide a shunting resistor across the contactor
load posts (across the points). 

At rest, the car slowly drains the Large Capacitor bank
 inside the controller.  When the key switch is turned on,
 the current surge is when the caps are charging up. 
This is a problem with most all controllers.

The Curtis "Precharge" feature is a slight  .75-1 second
delay before commanding the contactor to close.  This
 is done by the controller as it handles the contactor directly. 

The precharge resistor allows current to "bleed" over
and into the caps before the key switch is turned on
and has no effect after the contactor is closed.  The
only caveat is that there is a constant pack drain
while parked, unless the main pack disconnect is opened.

My system was set up at a higher voltage than Curtis
 had recommended initially.  It provided a measure of
 extra zip and mileage, but it had several other side
effects.  The built in and not adjustable 1 second
precharge time was too short to fully charge the
caps at this voltage.  When the contactor was closing,
 there was considerable current flowing across the
contactor points and they welded closed. (Which set
 a contactor error in the controller.)

I had no extra at the time, So I "fixed it" by tapping it
 with a hammer until it broke apart and opened again.
  While it still worked after that, it still had point contact
 damage and threw Error messages about "points being
 oxidised"  or " too much point resistance" at times.

After installing the new contactor, I installed a bleed
 resistor which works by starting a precharge cycle
 after I close my main pack switch and adds about two
seconds to the one second internal delay.  According
to HPEV, this should do the job.  They were very helpful
 helping me with this issue and even recommended the
 correct value resistor to use.

This is unusual and not a "factory" fix.  Use at your own risk.

It seems to work fine for me.

You have the option of turning the precharge feature off
in the controller, but I did not hurt to just leave it on.

Miz 

1930 Ford Speedster, AC50, full manual powerglide, 6.14gears, 38-130AH CALBs.

Offline mizlplix

  • Administrator
  • Senior Member
  • ****
  • Posts: 1445
  • Country: us
  • "The more you look, the more you find."
    • The old  build thread....
Re: copper rotors
« Reply #31 on: February 09, 2013, 02:52:52 AM »
___________________________________________________________________________ ________________________
                     Further optimizing steps

Lessons learned:

1- The Curtis controller recommends the slip gain setting to use. 
I walked it up and down over and under that setting and always
seem to go back to it in the end.

2. The Curtis controller recommends the Field Weakening Base
Speed setting also. I have a new way to choose the correct F/W base speed.

My car with the new motor is much faster than the AC50 by a large margin in
 the 0- 2,500 RPM range.  After that the acceleration rate is
 almost the same, (but I am still working on it at this point).

My forward map Ki setting is critical in the initial launch of the car from a dead stop. 

I can not do more than 10-15 full power launches before recharging.  (Or it starts affecting my results.)

Later...Miz
 
1930 Ford Speedster, AC50, full manual powerglide, 6.14gears, 38-130AH CALBs.

Offline mizlplix

  • Administrator
  • Senior Member
  • ****
  • Posts: 1445
  • Country: us
  • "The more you look, the more you find."
    • The old  build thread....
Re: copper rotors
« Reply #32 on: February 11, 2013, 05:29:33 AM »
___________________________________________________________________________ __________________________
            Happy Monday morning to everyone!

Yesterday. I finally got through with the optimization tests.  The car takes off like a Bat-out-of-hell and accelerates smoothly up the RPM band and arrives at it's top RPM nicely with no jerking or jolts.  It behaves smoothly with any throttle input and is totally predictable.  It responds immediately without any delay on both accel and decell. 

Reverse was set up a little softer on the initial throttle application to give it a docile feel when parking in tight spots.

The part throttle cruise was decent also.  The programmer read out 75-85 amps while holding to 40 MPH.  If you recall, the AC50 was in the same region at 80-90 Amps in the same car, same road, same clear weather and same driver with the pack freshly topped off in both tests. (the variables were the same as humanly possible)

My zero - 60MPH time was at 9.8 seconds with the new motor while the AC50 was 12.1 seconds (using high gear only when accelerating).  The AC50 took 10.5 seconds when using low and high gears to do the same test.

There was no real weight change to speak of in the motor swap.  The weight of the AC50/powerglide unit was within 5 Lbs of the new motor weight.  Although the frictional drag in the powerglide was minimal, it might have added a second to the AC50's 0-60 time.  (Making the corrected time for the AC50 at 11.5 seconds if it were direct drive.)

Acceleration zero to 60MPH:
The  AC50=11.5 seconds
New motor=9.8 seconds

Cruise current: At 40 MPH:
The AC50=80-90 Amps   
New motor=75-85 Amps

Our original theory was that the AC50 left a little controller capacity on the table and our new motor would take advantage of that.  The end results seem to bear this out. 

But (there always seems to be a but), The new motor is falling a little short of the top RPM gestimate and my car would need a rear gear or rear tire change to suit me and get the overall performance I want. 

After "putting our heads together", we have decided to retest the motor on the static bench at the Tucson Motor shop.  Possibly revise the stator winding if deemed necessary before re-gearing the car.

So, off to Tucson I go.  I will post anything we find out today.

Miz
 
1930 Ford Speedster, AC50, full manual powerglide, 6.14gears, 38-130AH CALBs.

Offline mizlplix

  • Administrator
  • Senior Member
  • ****
  • Posts: 1445
  • Country: us
  • "The more you look, the more you find."
    • The old  build thread....
Re: copper rotors
« Reply #33 on: February 11, 2013, 03:23:27 PM »
___________________________________________________________________________ _________________________________
                                             Yahoo!

I just got back from Tucson.  From all we can  determine, my motor needs rewound for a lower voltage range.  Everything we have done up to now has been based on several assumptions.....One is wrong and another is questionable.  When we get solid answers, we will share them.

That should do several things, Allow a lower field weakening rate, lower the resistance through the stator and Raise the top RPM.

Ivan has it and is rewinding it now.  I should have it back in a week or so.

I am getting good at removing and installing it...LOL

Miz

1930 Ford Speedster, AC50, full manual powerglide, 6.14gears, 38-130AH CALBs.

Offline mizlplix

  • Administrator
  • Senior Member
  • ****
  • Posts: 1445
  • Country: us
  • "The more you look, the more you find."
    • The old  build thread....
Re: copper rotors
« Reply #34 on: February 14, 2013, 06:43:31 AM »
___________________________________________________________________________ _____________________________

I am predicting a friday date for reinstallation of the new rewind.

It should take about an hour.  Both Ivan and myself will be able to quickly compare and determine what our latest modifications will do.  Guesses, theory and formulas are all well and good, but actual performance, testing and concrete results have the final say.

We have been criticized by others on many other forums for not being qualified, not scientific, not being highly educated and even being a disservice to other builders by misleading them with our conclusions.

To that I would like to say, most progress in history was done by people like us.  At least we are trying instead of spending our time telling others "It can't be done." 

We are not selling information.  We are sharing freely.  We post failures as well as improvements alike. 

There is a large lack of understanding and direct experience in AC vehicle motors because those who do know, do not share. 

I am not talking about those "Internet Theorists" who can prove through miles of equations or linking to the work of others to try and prove a point. I am speaking of real, verifiable, hands on experience building and testing AC motors for EV use.   

Please excuse the small rant.  I am at my saturation point with internet nay-sayers (and even knowledgeable persons) who only use that knowledge to belittle our efforts and not ever to contribute. 

We will have some news (good or not-so-good) sometime friday evening.

Miz
1930 Ford Speedster, AC50, full manual powerglide, 6.14gears, 38-130AH CALBs.

Offline HighHopes

  • Senior Member
  • ****
  • Posts: 225
  • Country: ca
Our Test Mule
« Reply #35 on: February 14, 2013, 10:08:03 PM »
our fore fathers accomplished a LOT more than we have with far fewer tools and materials available.  we've got it made by comparison!  they had no computers, hardly any lab tools (except those they built themselves), difficult to source any sort of material, would take months to hear from the ONE or TWO colleagues around the world to collaborate with .. yet they changed the world. 

Never underestimate what one can do in his/her garage .. with limited knowledge and amateur tools. 

the only universal truth for discovery and wisdom is to realize that "mother nature has been most kind to you, she has given you everything - you just need to be clever enough to figure it out"   .. so ignore the nay sayers and lets go figure it out!

$0.02

Offline mizlplix

  • Administrator
  • Senior Member
  • ****
  • Posts: 1445
  • Country: us
  • "The more you look, the more you find."
    • The old  build thread....
Our Test Mule
« Reply #36 on: February 16, 2013, 03:49:58 PM »
___________________________________________________________________________ ________________________________

Ivan reports that he has the stator rewound and tied.  It has
just a little varnish on critical spots just to allow us to run it.

After some consultation, It will be wound for a much lower voltage,
 18 gauge, all in-hand, one turn on each pole for each phase.
(We have shortened the wire length to 30% of what it was while
 maintaining the same slot fill.)

If it responds like expected, it will come back out and have the
 full coating applied.  This will be the first one we have done with
 a total spray coating.  The other ones had a dipped and baked
traditional motor shop finish.    Our friend at the Tucson motor
shop thinks the newer synthetic sprays will probably be as good
or even better than the traditional dip they use. (and easier to apply)

By Sunday evening, I should have some results and hopefully
 good news too.

Miz
1930 Ford Speedster, AC50, full manual powerglide, 6.14gears, 38-130AH CALBs.

Offline few2many

  • Senior Member
  • ****
  • Posts: 128
  • Country: us
Our Test Mule
« Reply #37 on: February 17, 2013, 06:03:29 PM »
I definitely look forward to hearing how it goes! Are you worried about possibly overdoing the reduction in windings and voltage?
Will bringing down the wound-voltage extend the torque range further?

Offline few2many

  • Senior Member
  • ****
  • Posts: 128
  • Country: us
Our Test Mule
« Reply #38 on: February 17, 2013, 08:46:41 PM »


Please excuse the small rant.  I am at my saturation point with internet nay-sayers (and even knowledgeable persons) who only use that knowledge to belittle our efforts and not ever to contribute. 

We will have some news (good or not-so-good) sometime friday evening.

Miz

Rant understood. There are definitely a few, very arrogant, persons who are intelligently rude and condescending. They know what and how to say things to really insult, while staying within forum "guidelines".

Offline mizlplix

  • Administrator
  • Senior Member
  • ****
  • Posts: 1445
  • Country: us
  • "The more you look, the more you find."
    • The old  build thread....
Our Test Mule
« Reply #39 on: February 18, 2013, 05:28:12 AM »
___________________________________________________________________________ ________________________
                  Good news on Sunday!

Ivan arrived early on Sunday morning and we fully assembled the motor.

We are using only the spray on type of motor varnish this
 time to duplicate what a DIYer would have to do in their back
 yard.  Up to this point we were using the motor shop's varnish
 tank and oven to do the standard "bake on" finish.

Jim, the Tucson motor shop owner, said he thinks this would be
 just as good, but it is more costly for him to do that way. We
 wanted to test the new stuff if everyone will be using it for their
 builds.

The old winding was 12 - 18 gauge wires in hand with 2 turns around
 each pole.  The new wind has almost the same slot fill at 20 in hand
 but only one turn around each pole. (We used slightly heavier paper
 and had to drop a few wires.)  When doing the Delta connections
 and combining two of these 20 wire bundles, it makes three- 40
wire bundles.  Those are your motor leads.  They wind up about
equal to a #2 cable and can easily handle that 650 amp controller
maximum current. 

We performed the "Auto-run" portion of the optimization routine
again (This IS a new motor from the last one).  The F/W base
speed went from 770 RPM to 1050 RPM, the slip gain to 3.20
and the F/W went to zero for the starting setting to perform
the acceleration tests.  I reset the F/W percentage to the HPEV
setting of 20%  as my starting place also.

Right away, the motor reacted differently from the last time. 
The top RPMs went up from 3000 RPM's to 4000 RPM's.  (Which
was what we expected and was the whole point).  We traded
some bottom torque for an extended top RPM range.  I need
a 60 MPH car for my around town driving. I never drive
expressways, but I need good off-the-line jump (which the
AC50 lacked).

The AC50 in a direct drive application is good IF the car weighs
 no more than 2,000 LBS, the car is geared perfectly and the
driver is OK with a gentle initial take off.  But from 3000 RPM
up, it is a decent performer.

The new motor with the larger diameter stator, is capable of
much better off-the-line torque, but lacks the upper RPM the
AC50 had (Which I did not need for my purposes). 

It was my opinion that the AC50 left some controller capacity
on the table, so the new motor will be able to utilize all of it,
but it certainly would respond to a larger capacity controller
when one finally is available.

We drove around a little and played with the programmer,
but the serious optimizing needs done with just me in the
car and a fully charged pack, so it is the same as I previously
did. 

I will be busy today building a motor for an early Corvette
 and I will start my final optimization in earnest on Tuesday
 morning.

But, At this point, It looks like the journey is almost at an end.

Miz

1930 Ford Speedster, AC50, full manual powerglide, 6.14gears, 38-130AH CALBs.

Offline mizlplix

  • Administrator
  • Senior Member
  • ****
  • Posts: 1445
  • Country: us
  • "The more you look, the more you find."
    • The old  build thread....
Our Test Mule
« Reply #40 on: February 21, 2013, 01:13:58 PM »
___________________________________________________________________________ ___________________________
    Life interferes with living

I got to do a few runs Tuesday.  I need to frequently recharge the pack because every run I do is full throttle, all out,Tweaking the field weakening number in small increments.  I have to watch the SOC because it starts affecting the result and I must stop to top off the pack .

It stormed and rained Wednesday all day.

Thursday is really cold, but I expect to get in some time.

This portion requires patience and keeping good notes.  Each run makes the motor a little bit better. 

Miz
1930 Ford Speedster, AC50, full manual powerglide, 6.14gears, 38-130AH CALBs.

Offline mizlplix

  • Administrator
  • Senior Member
  • ****
  • Posts: 1445
  • Country: us
  • "The more you look, the more you find."
    • The old  build thread....
Our Test Mule
« Reply #41 on: February 27, 2013, 08:16:38 AM »
___________________________________________________________________________ _______________________________
                               OK, Where to start...

Each day, I wait until the morning chill begins to disappear before I do any testing.  I have the pack plugged in over night and it starts out at the same SOC and temperature every test period.

I divide my day into two test sessions.

I test by driving one zero to 60 sprint, then pull over and stop.  I compare my time (and feel) to the previous tests.  I then determine my next parameter change.   I repeat, and repeat until I get that first sonic alarm from my BMS that the pack has dipped below the 2.8 volt point during full load.  (When stopped  and under no load the pack is still above 3.0 volts) This does not really take long at all as most of the time is spent above 300 amps.

I return gently to the shop and plug the car back in.

I compare what I had learned and decide where I will start for the PM test session. Which I do as soon as the pack is fully charged to duplicate the AM session as completely as I am able.

This "optimization" process seems to be divided into three distinct sections.

1-The Curtis "auto-run" portion.
2- Changing parameters until the bset overall performance is achieved.
3- Changing parameters to maintain that level of performance while reducing the current draw.

It is not an intuitive process as some of the parameters are not explained very well and actually leave the wrong impression. 

I am now at the last phase of the process where I am fine tuning the controller for reduced pack draw.

My performance goals have been achieved but the rest needs further work.

Miz
 
 
1930 Ford Speedster, AC50, full manual powerglide, 6.14gears, 38-130AH CALBs.

Offline arlo1

  • Senior Member
  • ****
  • Posts: 80
Our Test Mule
« Reply #42 on: February 27, 2013, 09:14:58 AM »
I just read this whole thread.  Its very exciting.  What frame size is that motor?  What RPM was it rated for origanly?

Offline mizlplix

  • Administrator
  • Senior Member
  • ****
  • Posts: 1445
  • Country: us
  • "The more you look, the more you find."
    • The old  build thread....
Our Test Mule
« Reply #43 on: February 27, 2013, 07:34:07 PM »
There was no tag on the motor when Ivan got it from the motor shop junk pile.  It was disassembled and it took him a couple of hours to find all the correct parts.

But one just like it was a 20HP  3,750RPM   Dont know the frame...Maybe Ivan remembers.....?

These things can be outwardly the same, but can vary a LOT when disassembled. 

This one has an 8" Dia. twisted rotor and a 7.5" long, 48 slot stator and has a standard length housing.(maybe inverter duty)

One just like it had a 7.25" dia. rotor with a 5" long, 36 slot stator, but still has a standard length housing.(had a C face for a pump)

At the end of today, I have it set up pretty well and it drives really nicely in traffic.  I am still tweaking to bring the current draw down while maintaining the smooth acceleration and the 4,200 RPM top speed. (in this car =65MPH) 

(Note to self: even with the pack turned off, this controller will bite you....)

Miz
1930 Ford Speedster, AC50, full manual powerglide, 6.14gears, 38-130AH CALBs.

Offline arlo1

  • Senior Member
  • ****
  • Posts: 80
Our Test Mule
« Reply #44 on: February 27, 2013, 11:35:07 PM »
LOL yup I always discharge the caps in the controller before messing around.

Its interesting how this all works.  What would you say the max rpm you can safely spin your rotor?