EV Digest 3350
Topics covered in this issue include:
1) Why not Truck Reduction Gears?
by Aaron Birenboim <[EMAIL PROTECTED]>
2) Re: Did I torch my new old EV meter?
by Lee Hart <[EMAIL PROTECTED]>
3) OT:RE: More T-rex-Godzilla Nonsense
by David Brandt <[EMAIL PROTECTED]>
4) Re: Sparrow reborn
by Rich Rudman <[EMAIL PROTECTED]>
5) Re: AC/DC hybrid drive combo?
by Rich Rudman <[EMAIL PROTECTED]>
6) Re: Sparrow reborn
by Rich Rudman <[EMAIL PROTECTED]>
7) Re: AC/DC hybrid drive combo?
by Rich Rudman <[EMAIL PROTECTED]>
8) Advanced Automotive Battery Conference
by "John Bryan" <[EMAIL PROTECTED]>
9) Transmission efficiency (Re: The Wheel)
by "Michael Hills" <[EMAIL PROTECTED]>
10) ASCII Schematic of hi-lo controller idea
by [EMAIL PROTECTED]
11) Re: ASCII Schematic of hi-lo controller idea
by Michael Hoskinson <[EMAIL PROTECTED]>
12) Re: Sparrow reborn
by Victor Tikhonov <[EMAIL PROTECTED]>
13) Re: Modifying my Sparrow motor
by Michael Hoskinson <[EMAIL PROTECTED]>
14) RE: Transmission efficiency (Re: The Wheel)
by David Brandt <[EMAIL PROTECTED]>
15) Re: anyone EVER fried a zilla?
by Otmar <[EMAIL PROTECTED]>
16) Re: Finally back online, newbie question @ 'hi-lo' controller
by Ben Apollonio <[EMAIL PROTECTED]>
17) Re: AC/DC hybrid drive combo?
by Victor Tikhonov <[EMAIL PROTECTED]>
--- Begin Message ---
I expect that people here have tried about everything,
so I'm assuming that truck reduction gears and/or overdrive
differentials have been dismissed for some reason.
I'm just curious about why?
If all we need is some reduction gearing for an EV, why
not use standard truck parts? Are they too heavy and innefficient?
Why not use reduction gear sets or overdrives as a simple
2-speed transmission, since most EVs (even DC) only really need
2 gears?
--
Aaron Birenboim | This space available!
Albuquerque, NM |
aaron_at_birenboim.com |
>http://aaron.boim.com |
--- End Message ---
--- Begin Message ---
Roy LeMeur wrote:
>> Hey! you guys are really giving me a hard time on this one. :-)
Joe Smalley wrote:
> It was the "SHOW ME ONE EXAMPLE" type of challenge that motivates
> people to think.
Same for me. I basically agree with your point; that stranded wire
should be used if there is any chance that the wire will flex or bend.
But the way you phrased your question encouraged me to think of
counterexamples -- and thinking is always good!
--
"Never doubt that a small group of committed people can change the
world. Indeed, it's the only thing that ever has!" -- Margaret Meade
--
Lee A. Hart 814 8th Ave N Sartell MN 56377 leeahart_at_earthlink.net
--- End Message ---
--- Begin Message ---
Godzilla is also *much* larger than a T-rex.
T-rexes...er..T-rex'es...um...T-rex's...no, that's not right...
Tyrannosaurs (there we go!) were at most a few stories tall. But in all the
movies, Godzilla is seen towering over bridges and low buildings, and
getting his legs tangled in high-tension lines. Of course, the size has
varied with the movies, but it averages out around 15 stories. Tall enough
to squash a T-rex underfoot if he's not careful (which he usually isn't).
> -----Original Message-----
> From: John Wayland [mailto:[EMAIL PROTECTED]
> Sent: Monday, February 09, 2004 9:25 PM
> To: [EMAIL PROTECTED]
> Subject: More T-rex-Godzilla Nonsense
>
> Hello to All,
>
> Getting way off topic, but what fun.....
>
> Roland Wiench wrote:
>
> > Hello John and all,
> >
> > This is dinosaur country. The name of this creature is a which
> > Japenese call is a Gojulas Dinosaur type (RBOZ-003), also
> call Terox and Zoidzilla.
> > The technical name is Tyrannasaurus.
> >
> > There are children books that talk about a dinosaur named
> King Zilla
> > which is a T-Rex.
>
> Wrong! Take a look at a typical T-Rex....yes, a similar
> 'walks upright' profile to Godzilla, but also way different.
> The T-Rex was a real dinosaur, that by all accounts, had
> long, powerful hind legs, an aggressive skull shape with a
> longish snout, eyes more on the sides, and a smooth back.
> Godzilla is a fictional, Japanese monster, with short, stubby
> rear legs, a more human-like skull shape with human like eye
> placement and oversize gorrila-like nostrils, and most
> different....large armor plates radiating down the back, much
> like a Stegosaurus...all of these monster attributes, are
> very un-T-Rex.
>
> By the way, T-Rex's full technical name is Tyrannosaurus Rex,
> and even if you live in dinosaur country, you won't ever find
> skeletal remains of a Godzilla!
>
> See Ya.....John Wayland
>
--- End Message ---
--- Begin Message ---
Victor Tikhonov wrote:
>
> Rich Rudman wrote:
>
> > The "motor problem" was Corbin never set them up right.
> > The Ac drive in a Sparrow is NOT the quick cure... it's the expensive
> > cure that will have alot of issues.
> > Wait until they have a real road test of the AC drives.
> > We all said a Ac drive was the way to go but it cost twice the list
> > price of the Sparrow.
> > I am pretty sure that has not changed much.
> > time will tell.
>
> I just wonder what kind of AC drive cost $24k (list price of Sparrow
> is $12k, is it?). Even best OEM drives are less expensive than that.
>
> Any info on this Rich? Corbin may be ripped off shopping in wrong
> place or for the wrong thing. (That is, unless $24k include charging/
> BMS/all integrated hardware and, may be the batteries
>
> --
> Victor
> '91 ACRX - something different
Yea Victor it was a ACP drive.
We had a good laugh about it.
--
Rich Rudman
Manzanita Micro
www.manzanitamicro.com
1-360-297-7383,Cell 1-360-620-6266
--- End Message ---
--- Begin Message ---
Victor Tikhonov wrote:
>
> Joe Smalley wrote:
> >
> > Rich called me to respond to this.
> >
> > Comments inserted...
>
>
> Good, we agree on everything. So I dare to think I know how an
> induction motor works :-)
>
> --
> Victor
> '91 ACRX - something different
I still don't think you understand that the 6Khz PWM Freq is almost Not
a item in this argument.
That's all I said. Well ment to anyways.
I also think Joe missed the slip factor that Lee nailed to the wall.
Freely spining at 1800 rpm with no load and ZERO slip, the motor
basicly falls asleep.
All of us have run a AC motor un loaded at Grid voltages and at 60 hz.
Since your AC drive motors are designed to run from 0 to 12000 RPM,
That means it can run just fine at 1800 rpm. Which is 60Hz, Which is
Grid.
Since your motors also are good for 312+ battery volts, your motors
also good for 120 VAC work.
So both points are within the design envelope... It will work.
But having both 60Hz and 120 volts across the motor at the same time...
That may be a different story.
My theory is that with zero slip.... the unloaded motor senario...
Nothing happens, the motor is in sync with the line Freq, and all you
have is windage losses and magnitization losses. Motor spins at 1800 rpm
and we all get very bored. The motor voltage can be jacked all over the
place, and very little change in current happens.
As long as the motor is in sync with the driven Frequency.
Wanna come up here with one and Watch?? I will be REALLY carfull. OK I
will do it with my AC motors first.
--
Rich Rudman
Manzanita Micro
www.manzanitamicro.com
1-360-297-7383,Cell 1-360-620-6266
--- End Message ---
--- Begin Message ---
Doug Weathers wrote:
>
> On Monday, February 9, 2004, at 11:56 PM, Roderick Wilde wrote:
>
> >
> > Doug, If you are going to use it at 120 volts or above, please advance
> > the timing. I beg you.
> >
> > Roderick
>
> Come on, Rod, get up off of your knees. You're embarrassing me. :)
>
> I'm thinking either 14 or 16 Optimas, depending on being able to wedge
> them into the Ghia nicely. That's 168 or 192 volts, so I need to
> advance the timing on my Sparrow motor.
>
> So the next question is, how do I do this? I've seen a couple of
> messages go by that mentioned what to do, but they didn't make much
> sense to me. Can anyone point me to some explicit directions,
> including things like torque numbers for bolting it back together?
>
> Alan Batie and Otmar have posted some pictures - it looks pretty scary,
> what with taking the motor apart and drilling new holes in the casing.
> Is that really what's required?
>
> Call me silly, but it seems that you should be able to loosen
> something, twist something else, then tighten the first thing down
> again. You don't need to drill new holes in the engine block of an ICE
> to adjust the timing.
>
> AC motors look better all the time. Maybe on the next EV.
>
> Later,
>
> Doug
> --
> Doug Weathers
> Bend, OR, USA
> http://learn-something.homedns.org:8100/weblog/
The up coming Gone Poatal show will show me dialing in 13.6 deg of
advance on 2 stock 8 inch "Sparrow Spec" motors. AvDC has a deal on them
for volume suppliers.
They are single ended long shaft motors. With NON timed cases. AvDC made
a TON of them, and well with Sparrow by Corbin out of the picture, they
have Stock. So we all get to see it in color, soon I hope.
But it's simple take 4 bolts out rotate the brush end bell about .750
of a inch mark, Drill and tap. Done. Do the trig to get the actual
advance that you need.
Street drives should NOT be taken over 10 Deg advance.
--
Rich Rudman
Manzanita Micro
www.manzanitamicro.com
1-360-297-7383,Cell 1-360-620-6266
--- End Message ---
--- Begin Message ---
Lee Hart wrote:
>
> This thread is wandering a bit, so I'll try to 'weave' it back together
> again. :-)
>
> To me, the bottom line is not, "Will it work". I am confident that you
> *can* run an AC traction motor on plain old 120vac 60 Hz power with
> relatively few extra parts. The motor will run, will produce a
> significant fraction of its rated power, and it won't overheat or be
> otherwise damaged.
>
> The *real* question is, "Will it work well enough to bother with?" It
> may require a big variac and big capacitor bank; that's already the
> beginnings of a crude high-power charger. It might also require a fairly
> high-power AC receptacle and cord to plug it in. And, it might require
> extra switches or contactors to reconnect the motors between driving and
> charging modes.
> --
> "Never doubt that a small group of committed people can change the
> world. Indeed, it's the only thing that ever has!" -- Margaret Meade
> --
> Lee A. Hart 814 8th Ave N Sartell MN 56377 leeahart_at_earthlink.net
OK Lee I am going to try to memorize this, since it's what I knew, but
lost the math on.
Thanks for you clear thinking, I get lost in the fight and forget
details.
OK Victor... now you know your motor won't even break a sweat at 120VAC
and 60Hz.
As I predicted.
--
Rich Rudman
Manzanita Micro
www.manzanitamicro.com
1-360-297-7383,Cell 1-360-620-6266
--- End Message ---
--- Begin Message ---
Come and join us from June 1-4, 2004 in the beautiful city of San Francisco for
another highly
charged ADVANCED AUTOMOTIVE BATTERY CONFERENCE.
We are now soliciting Poster Presentations on the following topics:
* Advances in High-power Li Ion Batteries and Components
* Advances in High-power Lead Acid Technology Advances in Ultracapacitor Design and
Components
* Power Distribution and Battery Management
* Other Battery Systems
* Advances in Fuel Cell Design and Components
* Power Sources for Specialty Vehicles
There will be two dedicated 75-minute poster sessions during the 2nd day of the
conference.
Present your most recent developments to the AABC audience of top power system
technologists from
all major automotive companies and power source developers. To learn more about plans
for AABC-04,
including instructions on submitting a poster paper, and for online AABC-04
registration, please
visit: www.advancedautobat.com. Please note that poster papers will also be included
in the
conference proceedings, which are distributed to all participants at the beginning of
the event.
We are also accepting reservations for exhibits. Please email us for more information
and a floor
plan.
Poster papers will be accepted based on relevance of content on a
first-submitted, first-reviewed basis. Space is limited, so please submit well in
advance.
Please note that panel presentations are by invitation only.
Waiting to hear from you and to be of service
The AAB team
--- End Message ---
--- Begin Message ---
I guess I wasn't too surprised that no one commented on the post below since
I think we've seen this (and others like it) before, and wheel motors seem
to get dismissed pretty quickly here because of the unsprung weight issues
and the fact that this list is mostly focused on things that people on this
list can actually buy and use.
Anyway, while looking at this site, I was appalled by the fact that they
can't even do an efficiency calculation correctly. Then I realized that
they were using a comma instead of a decimal point. :) Some of the
assumptions they make still look a little iffy, and it's clear they are
trying to push their solution, but one of the main points that they are
trying to get across got me thinking.
They claim that a standard gear train (transmission and differential)
consumes about half of the energy produced by an ICE, or by the electric
motor in an electric vehicle with a standard drive train. Of course my
immediate reaction was "no way" since I "know" they are more than 90%
efficient. But the rated efficiency for the transmission is at some torque
and RPM. Most of the time the car is cruising at a reduced load. What is
the efficiency at that load? One iffy assumption they made was that the
power consumed by the transmission is proportional to RPM and does not
depend _at all_ on the amount of torque being transferred. That's how they
calculate a 50% power consumption (see "Example - a conventionally geared
electric bus?" on the web page listed in the original email). It does make
sense though that the power transmission efficiency would decrease at loads
less than rated torque.
Not long ago, someone made a comment saying that a fixed ratio transmission
was definitely less efficient. Unfortunately I couldn't find the post in
the archives. If I remember correctly (I could be imagining things here),
it was Otmar and he said that since he has been leaving CAPOPE in 3rd gear,
he's been getting much worse overall efficiency. This shows that the motors
are less efficient with a fixed ratio, but to say that the car would be less
efficient requires the assumption that removing the transmission and setting
the ratio in the differential to be equivalent to 3rd gear wouldn't provide
a significant efficiency boost. It seems quite likely that using the
transmission during acceleration, when it's being used at close to rated
capacity and therefore 90+% efficiency, would improve efficiency, but if
it's adding a significant amount of drag at cruise speed, a few miles of
that could waste any energy saved during acceleration.
Also, in the back of my mind I have a vague recollection of hearing that the
Dualin' 7 was actually pretty efficient as long as Rich wasn't laying
rubber. Is that true?
It seems that in the past people have gotten noticeable range improvements
by switching lubricants in their transmission. If the transmission is
taking less than 10% of the power, changing the oil would have to make a
large difference in the transmission to make a noticeable difference in
overall range, and I have a hard time believing that one oil could be _that_
much better than another.
Thoughts?
Of course none of the examples I've given are substantial (based on
remembered second hand info and speculation). Any idea where one could find
actual data on the efficiency of typical automotive drive trains at various
loads and RPMs?
-Michael Hills
Date: Tue, 10 Feb 2004 07:21:44 -0800 (PST)
From: Rod Hower <[EMAIL PROTECTED]>
Subject: The Wheel
http://www.e-traction.com/TheWheel.htm
120kW wheel for busses/forklifts.
Perhaps 2 on the back of CE would be nice.
_________________________________________________________________
Plan your next US getaway to one of the super destinations here.
http://special.msn.com/local/hotdestinations.armx
--- End Message ---
--- Begin Message ---
* LP8.2: HTML/Attachments detected, removed from message *
--- End Message ---
--- Begin Message ---
I think it would look better if you sent it as plain text.
:)
Mike Hoskinson
[EMAIL PROTECTED] wrote:
* LP8.2: HTML/Attachments detected, removed from message *
...
Whaddaya'll think?
____________________________________ ________
|
/ _|__
| . R1______ ___________________/ :
| |
|____/ | | |
|
: | 96 |
____|_________ __|____|_______ | | | :R2
| v
|
| 9" ADC Motor | |1221c Controller | | 96 | | :
|___|
|_____________| | | v |
| :
______|
| | |___|
:/
|____________|_________________|_______ _/
--- End Message ---
--- Begin Message ---
David Roden wrote:
> I always thought that Corbin should have had a chat with Josef Brusa. His
> inverter designs (AMC-320/325) are close to a decade old now. Not only are
> they mature and proven, but much of the engineering costs should be well in
> the past. With a commitment to purchase enough of them, I wonder if Brusa
> might be willing to negotiate a good price. I could be ^way^ off base with
> that notion, though.
A little off, not way off :-)
AMC325 is no longer in production. That is if you want one or two
it may be in stock If you want 100, I believe no way. But, if you want
50,000, they may restart production, even though there are better
designs right now.
Also, as I mentioned numerous times, the price of silicon in
the AC drive is small fraction ot its cost. If you take one apart,
the loose parts, bought separately will perhaps cost 15% more than
the parts for a DC controller of comparable power.
--
Victor
'91 ACRX - something different
--- End Message ---
--- Begin Message ---
Shortening shaft of the Sparrow motor (if that is all you need to do
to get everything to fit) can be easily and very neatly done by
yourself. Spin the motor on 12VDC and hold a hack saw blade to the
spinning shaft at the appropriate spot. You might want to build a
barrier to keep the filings from going inside the motor or bearings,
but the method makes a nice smooth cut.
If not taking into account the shaft length is the only bad thing you
do on this project, you could be in line for an award for fewest
problems ever.
Mike Hoskinson
Edmonton
David Brandt wrote:
You need a spacer ring that matches the front of the motor and the hole
pattern on the adapter plate. Most adapter sets include these. It is a
very simple part, because the outer profile is unimportant, so it is usually
made from roud bar or mechanical tubing. It shouldn't cost you much at all.
-----Original Message-----
From: Doug Weathers [mailto:[EMAIL PROTECTED]
Sent: Tuesday, February 10, 2004 11:24 PM
To: [EMAIL PROTECTED]
Subject: Modifying my Sparrow motor
Hi all,
I bought my Sparrow motor from Bob, then turned around and
bought my motor adapter from Mike, telling him it was a
regular 8inch ADC motor.
My bad - turns out the Sparrow's motor shaft is longer than
the regular motor's, and so Mike's adapter leaves the
flywheel too far forward.
...
--- End Message ---
--- Begin Message ---
In general, a system of gears to change speed/torque is very efficient, and
as far as I know, that is roughly constant at all speeds.
Where you would see the largest single component of energy wasted in a
transmission is in moving the fluid. Keep in mind that you are lubricating
the gears as they turn (possibly by the transmission taking some of your
power for a pump, possibly by splash lubrication). When the gears mesh, the
free thickness fluid film is squished down to a lower thickness due to the
contact pressure between the gears, which pumps some of the oil out. So
your transmission essentially has a parasitic hydraulic pump built in. This
happens in all transmissions. This is why changing to a different fluid
can have such a (relatively) large impact.
These losses DO change with RPM, as you get more resistance when pumping a
fluid faster.
Automatic transmissions have much larger hydraulic losses involved, since
they use fluid power to shift and work valves, to say nothing of the torque
converter.
That said, the overall impact of the transmission hydraulic losses are
relatively small, as evidenced by the gains in range of around 5-10% (plus
or minus, YMMV) experienced by EV'ers that have switched to various
synthetic fluids.
> -----Original Message-----
> From: Michael Hills [mailto:[EMAIL PROTECTED]
> Sent: Wednesday, February 11, 2004 1:42 PM
> To: [EMAIL PROTECTED]
> Subject: Transmission efficiency (Re: The Wheel)
>
> I guess I wasn't too surprised that no one commented on the
> post below since I think we've seen this (and others like it)
> before, and wheel motors seem to get dismissed pretty quickly
> here because of the unsprung weight issues and the fact that
> this list is mostly focused on things that people on this
> list can actually buy and use.
>
> Anyway, while looking at this site, I was appalled by the
> fact that they can't even do an efficiency calculation
> correctly. Then I realized that they were using a comma
> instead of a decimal point. :) Some of the assumptions they
> make still look a little iffy, and it's clear they are trying
> to push their solution, but one of the main points that they
> are trying to get across got me thinking.
>
> They claim that a standard gear train (transmission and
> differential) consumes about half of the energy produced by
> an ICE, or by the electric motor in an electric vehicle with
> a standard drive train. Of course my immediate reaction was
> "no way" since I "know" they are more than 90% efficient.
> But the rated efficiency for the transmission is at some
> torque and RPM. Most of the time the car is cruising at a
> reduced load. What is the efficiency at that load? One iffy
> assumption they made was that the power consumed by the
> transmission is proportional to RPM and does not depend _at
> all_ on the amount of torque being transferred. That's how
> they calculate a 50% power consumption (see "Example - a
> conventionally geared electric bus?" on the web page listed
> in the original email). It does make sense though that the
> power transmission efficiency would decrease at loads less
> than rated torque.
>
> Not long ago, someone made a comment saying that a fixed
> ratio transmission was definitely less efficient.
> Unfortunately I couldn't find the post in the archives. If I
> remember correctly (I could be imagining things here), it was
> Otmar and he said that since he has been leaving CAPOPE in
> 3rd gear, he's been getting much worse overall efficiency.
> This shows that the motors are less efficient with a fixed
> ratio, but to say that the car would be less efficient
> requires the assumption that removing the transmission and
> setting the ratio in the differential to be equivalent to 3rd
> gear wouldn't provide a significant efficiency boost. It
> seems quite likely that using the transmission during
> acceleration, when it's being used at close to rated capacity
> and therefore 90+% efficiency, would improve efficiency, but
> if it's adding a significant amount of drag at cruise speed,
> a few miles of that could waste any energy saved during acceleration.
>
> Also, in the back of my mind I have a vague recollection of
> hearing that the Dualin' 7 was actually pretty efficient as
> long as Rich wasn't laying rubber. Is that true?
>
> It seems that in the past people have gotten noticeable range
> improvements by switching lubricants in their transmission.
> If the transmission is taking less than 10% of the power,
> changing the oil would have to make a large difference in the
> transmission to make a noticeable difference in overall
> range, and I have a hard time believing that one oil could be
> _that_ much better than another.
>
> Thoughts?
>
> Of course none of the examples I've given are substantial
> (based on remembered second hand info and speculation). Any
> idea where one could find actual data on the efficiency of
> typical automotive drive trains at various loads and RPMs?
>
> -Michael Hills
>
>
> >Date: Tue, 10 Feb 2004 07:21:44 -0800 (PST)
> >From: Rod Hower <[EMAIL PROTECTED]>
> >Subject: The Wheel
> >
> >http://www.e-traction.com/TheWheel.htm
> >
> >120kW wheel for busses/forklifts.
> >
> >Perhaps 2 on the back of CE would be nice.
>
> _________________________________________________________________
> Plan your next US getaway to one of the super destinations here.
> http://special.msn.com/local/hotdestinations.armx
>
--- End Message ---
--- Begin Message ---
Seth Murray wrote:
has anyone EVER fried a zilla? and what makes the zilla so much more
robust than a DCP anyway? I imagine obviously different/more
FETs/IGBTs, but are we talking different transistor packages as well,
like to247 vs isotop or anything? better heatsinking? greatly
overrated diodes etc? better current monitoring? electrically curious,
seth
--
QUESTION INTERNAL COMBUSTION
'72 Datsun 240Z Conversion
http://users.wpi.edu/~sethm/
I will let Otmar take that one... I know but...I have the chip on my
shoulder....I need to give it a rest.
Come on Otmar...critical review of a obsolete out of production
design....
Hey Seth and Rich and other interested ones,
Yes, people have fried Zillas.
So much has been learned over the years that it's hard to remember
what happened when, especially since it's been so long since the
smoke came out of a Zilla outside of experiments in my shop. Someone
else was asking this recently so I'll wrack my computer a bit. I
think that three Zillas have let the smoke out. They were all of the
older design that used less powerful parts, all this happened back in
the last century. (It's kinda fun to say that)
The new models have had some noise related faults that cause them to
shut down, but fortunately none of the new series have left the smoke
out. I think I have those bugs worked out now, but only time will
tell.
Of the three that blew, they were all before I upgraded to the more
rugged IGBTs and diodes.
1) The first one the college students hooked the battery up
backwards. In addition they neglected to follow the instructions on
precharging and checking voltage before turning it on. That was back
in 1997. All of you who are wondering why it now requires a
integrated precharger and Hairball interface know one of the many
reasons why. I can't blame that on the controller.
2) Rod blew a fresh 1400A unit back in 1999. I believe he was
climbing a long hill in Jerome AZ in the Land Rover when it went into
thermal cutback. He continued to push it. The controller had cut
back to 200 amps because the heatsink was so hot. This was all
happening due to the fact that he had not yet hooked up the cooling
system (or was it that he had forgotten to fill it?) Rod can be quite
a component tester! The Zilla should have survived even that, but the
next time he started it up it went pop and wouldn't move. The diodes
had overheated and shorted out when they cooled down again. Maybe Rod
can fill in more details. I fixed that under warranty since it had
some odd corrosion on the control board that may have contributed to
the problem. It seems that the boards had not been properly cleaned
before the conformal coating step. That was unfortunate because it
was hard to tell if it was a design or manufacturing problem.
3) The last one that blew was Berube's old unit. After 4 years of
running he had noticed that the current limit was getting irregular
and so I told him to send it in. He said he would "be careful" and
keep driving it. Yeah right! Anyway, it blew within a few months and
got rebuilt as well as upgraded to a then current design. It had a
number of old issues that I had since improved upon, so it was no big
surprise that it failed.
About that time I was spending a couple years revamping the Zilla to
the Z2K series. This one was a major redesign that used more powerful
IGBTs and Diodes, as well as thicker copper and easier to assemble
construction. The Z1K was designed in as a subset from the start.
Where the old Zilla required ice cooling water to reach a maximum of
1800 amps, the new Z2K series was designed to put out 2000 Amps with
ambient temperature water even in the heat of Phoenix. Along with
the Z2K series came the Hairball and Precharger which enabled a host
of safety and performance features.
If you like the new Zilla offering, you can all thank Rick Woodbury
and his Tango for bringing it back from a near retirement. Rick
needed a reliable controller with many features to make it safe for
regular street use. His prepayment on a number of units helped fund
the redevelopment of the Zilla to make the line what it is now.
As to what makes them reliable, there's only one thing I know of that
has to happen.
The silicon that makes up the active material of the IGBTs and Diodes
must not get too hot.
From this simple statement come lots of problems and design issues.
There are many ways to insure the parts stay at safe temperatures,
and different people will do it differently. Also, people put
differing values on various aspects of controller design. This is
because there are enough variables that it is hard to quantify them
all scientifically. I'm sure that some parts of the Zilla are way
over-designed while others are probably closer to the limits. The
hard part of this all is knowing just what is and isn't important.
I've focused on a wide variety of aspects as I became aware of them
and thought them important. A few that I do differently than the old
DCP design are these:
I belive it is essential to have good fast reliable current limit
feedback loops on the motor loop of the controller. This makes sense
to me since the heat created by the controller is largely a factor of
motor loop current. I do all my limit math based on motor current, so
I have to be able to control that. Even with a simple shunt circuit,
I tweaked it for years before I was finally happy with the current
sense signal. The noise that the controller produces makes it hard to
measure any low level signal.
I think it is crucial to have the power parts running at close to the
same temperature and voltage drop. How close is a critical question.
I always buy enough power parts for at least ten controllers before I
hand sort them into sets. That way I can match them very tightly. The
backs of the parts where the heat and power is conducted are also
surfaced prior to assembly. When I do assemble them I use a special
thermal grease mixture which I came up with myself. I would never
mount a power part on a electrical insulating pad in a massively
paralleled setup like a controller, I feel the risk of thermal
runaway is too high. These ideas are the result of many months of
math and testing. As an example I measured the thermal and electrical
resistance of the parts to the bus bars that they are attached to
with various surfaces and grease compositions. These ideas also
probably involve a lot of superstition. There are many details that I
am careful about during assembly that may not have any effect on
lifetime at all. I could test and try to determine how sloppy each
step could be, but mostly I'm not willing to take the time for that.
For these low volumes it's easier to just keep doing what works. I
establish a routine for the assembly even when I don't know if the
details would be significant. That way I hope I can reproduce designs
well, and if there is a problem down the road it's easier to know
what caused it.
Then there is always the easy solution. Use more parts, and more
heatsink. This costs money, but I find it pays off in reliability. I
think I have a good ratio of silicon vs rating. Even if I do have to
sacrifice markup to do it. In addition all my heatsinks are made of
copper. I think it would be hard to keep many parts at similar
temperatures using aluminum heatsinks. I also like the high specific
heat capacity of copper. Per volume it's about twice that of aluminum
and this allows me to make a smaller controller that can still go
down the drag strip while only cooling between runs. On the downside
copper is expensive to buy and hard to machine, such is life when you
are trying to optimize things. :-)
So, those are the main things. There are of course hundreds of other
problems that can destroy controllers. Gate drives are basically a
mini controller in themselves. The Z2K gate drives go from zero to 70
amps 30,000 times a second. And they have to get the signal to all
the IGBTs within nanoseconds of each other. Code has to work well,
power supplies are critical and so on.
Are the Zillas reliable now? I sure hope so. As the years go by I
feel better and better about them. Yet you never know when someone
will use one in some different way and bring up a problem that hasn't
been seen before.
Have fun!
--
-Otmar-
http://www.CafeElectric.com/ Home of the Zilla.
http://www.evcl.com/914 My electric 914
--- End Message ---
--- Begin Message ---
Who's been there & done that, or- what am I missing, besides the
potential to
let out some Curtis smoke?
The potential to weld your contactor? From there maybe battery smoke,
and you might even have bits of motor escape if you have to throw out
the clutch to stop the car. Then again, I believe John Wayland
did/does use a Bubba contactor to short the batteries across the motor
for an extra speed boost going down the track. What are you planning
on for batteries/contactors?
- Ben
--- End Message ---
--- Begin Message ---
Lee, thanks for thorough analysis! All theory is correct, and I learned
something new. But I can almost guarantee you, you will not see such
a setup practically implemented. All things considered, I's just a
plain bad idea.
Victor
Lee Hart wrote:
>
> This thread is wandering a bit, so I'll try to 'weave' it back together
> again. :-)
>
> It begins with the (slighly mad) idea that one has an EV with an AC
> traction motor/controller *and* a second DC traction motor/controller,
> making it an AC/DC hybrid. It thus has an AC drive for efficiency when
> cruising, and a DC drive for fast accelleration.
>
> Now, if the big AC and DC motors are coupled to each other and uncoupled
> from the wheels, they could perhaps be used as a motor-generator for
> battery charging. This could provide a high-power charger for very
> little additional cost and weight (since the big expensive parts are
> already in the car).
>
> Victor contends that this won't work (isn't practical) because the AC
> traction motor is built for high-frequency AC operation. It is built for
> something like 6 KHz, so at 60 Hz you'd have to derate it 100:1, which
> doesn't give you enough power to do anything useful.
>
> I contend that an AC traction motor is actually built to run well on
> anything from 20-200 Hz. The 6 KHz is only the PWM frequency, and not
> what the motor actually sees.
> _ _ _ _ _ _
> AC_______| | | |___/\/\/\__________| | | |__
> hot L1 R1 | | L3 |
> |_ | |
> _| > >
> L2 _| > R2 > R3
> _| > >
> | | |
> AC_________________________|____|___________|
> neutral
>
> Above is the equivalent circuit of an AC induction motor (view with
> fixed-width font such as Courier). Only a single phase is shown. Each of
> the values in the circuit corresponds to the following:
>
> L1 = stator winding inductance (inductance with rotor removed)
> R1 = stator winding resistance
> L2 = magnetizing inductance (inductance with rotor spinning at
> synchronous speed)
> R2 = core loss equivalent resistance
> L3 = rotor winding inductance (scaled to 1:1 turns ratio with stator)
> R3 = rotor winding resistance (scaled to 1:1 turns ratio with stator)
> plus the effective resistance of the mechanical load
>
> Victor gave figures for L1 and R1 for one of his Siemens motors. They
> were indeed very small; something like R1=0.01 ohms and L1=200uH. But
> before we can predict what will happen if you connected such a motor to
> 120vac 60 Hz, we have to know the values of R2, R3, L2, and L3.
>
> First, let's assume the rotor is stalled. Then R3 is simply the rotor
> winding resistance, which is very low (just one shorted turn). We scale
> it according to the turns ratio between the stator and rotor. A typical
> motor might have 10 turns on the stator. Like a transformer, the
> resistance scales as the square of the turns ratio. Thus, R3 is 100
> times the actual rotor winding resistance. With most induction motors,
> R3 will wind up being roughly equal to R1 i.e. R3=0.01 ohms.
>
> Estimating L3 is much harder. There are large differences in rotor
> inductance due to the type of laminations, the depth of the slots the
> rotor bars are placed in, size of the air gap, and other factors. The
> only way to know for sure is to look it up in the motor's engineering
> specifications, or measure it in actual test.
>
> The test proceedure is fairly straightforward. You lock the rotor so it
> can't turn. You apply AC voltage with a variac, and adjust it for full
> rated winding current. Then measure the stator voltage, current, and
> phase angle. From these you can calculate the total values of L1+L3 and
> R1+R3. Since L1 and R1 can be measured with the rotor removed, you can
> therefore find L3 and R3.
>
> But it is reasonable to assume that L3 is similar to L1. If so, then
> L3=200uH. Knowing L1, L3, R1, and R3 lets us calculate the current if we
> were to apply 120vac 60 Hz:
>
> X = 2 pi f L = 2 x 3.14 x 60hz x 400uH = 0.15 ohms
> R = R1 + R3 = 0.01 + 0.01 = 0.02 ohms
> Z = sqrt(R^2 +X^2) = sqrt(0.02^2 + 0.15^2) = 0.15 ohms
> I = E/Z = 120vac / 0.15ohms = 789 amps
>
> So yes indeed; the STALLED ROTOR current is very high!
>
> But, now let's assume there is no load on the motor. R3 includes the
> equivalent resistance represented by the mechanical load as well. The
> mechanical losses (friction and windage) are very small; like 3% of the
> motor's horsepower rating. You didn't say what the rated HP of this
> motor was; let's say it was 20 HP. Then R3 represents 20hp x 0.03 =
> 0.6hp x 746w/hp = 448 watts. At 120v, R3 = 120v^2 / 448w = 32 ohms.
>
> R3 is so large compared to R1, L1 and L3 that it totally dominates. The
> *in-phase* resistive AC current due to the frictional losses from the
> unloaded rotor is only I = 448w / 120vac = 3.7 amps!
>
> But we have left off L2 and R2. R2 is easier to estimate; it is the
> equivalent resistance caused by core losses; the things that make the
> core heat up. A high performance traction motor goes out of its way to
> use very good laminations and construction techniques to keep these
> losses low. So let's say they are only 3%. Thus, they account for
> another 32 ohm resistor, and another 3.7 amps of stator current.
>
> Now the big one; L2. This is the magnetization inductance; the apparent
> inductance the motor has if the rotor is spun at synchronous speed. For
> induction motors, L2 is on the order of 10-50 times the value of L1. You
> can measure it by letting the motor run no-load, which makes R3 so large
> that the reactive current thru L3+R3 is negligible. The resulting
> reactive stator current is due to the sum of L1+L2.
>
> L2 is what allows induction motors to be run from a fixed-voltage source
> regardless of load. For best efficiency at light load, you *should*
> apply a low AC voltage -- this is what the inverters in an EV AC drive
> do. But you can also deliberately apply full voltage to an unloaded
> motor -- this is what happens with normal induction motors on the AC
> line. The current is high; but it is REACTIVE current, almost 90 degrees
> out of phase. Thus, it does not produce heating or losses (except for
> incidental losses in the resistance due to the high circulating
> current).
>
> OK, let's suppose L2 = 20 x L1 = 10 x 200uH = 4mH. What is the AC
> current at 120vac 60 Hz with no load on the motor?
>
> X = 2 pi f (L1+L2) = 2 x 3.14 x 60hz x (4mH + 200uH) = 1.58 ohms
> I(x) = 120vac / 1.58ohms = 76 amps
> R = R1 + (R2 | R3) = 0.01 + (32 | 32) = 16.01 ohms
> I(r) = 120vac / 16ohms = 7.5 amps
>
> In effect, the L and R are in parallel, because L1 and R1 are negligible
> compared to L2, R2, and R3. So the total AC current is
>
> I = sqrt(76^2 + 7.5^2) = 77 amps. We have a huge current, but the
> resistive (loss-producing) portion is very small. The load is almost
> purely inductive. This is exactly what you see with an unloaded
> induction motor with excessive excitation voltage.
>
> As you add load (i.e. load the AC motor with the DC motor as a generator
> to charge the batteries, the resistive portion of the current increases.
> The inductive portion actually drops, as a voltage divider is formed by
> L1+R1 and L3+R3. But it will never get any better than a power factor of
> about 0.8-0.9 (characteristic of any induction motor).
>
> So, our questions become:
>
> 1. Can this motor stand 77 amps of continuous winding current? If not,
> then you would need an autotransformer or other means to step down
> the 120vac 60 Hz to reduce it to an acceptable level.
>
> 2. The power factor stinks at light load. You're not paying for the
> power, but you'd need a hefty circuit and wiring (like 100-amp)
> to supply it.
>
> 3. Can we load it heavily enough to get the power factor reasonable?
> Or will we need to add capacitors or other means to get the power
> factor down. My guess is that we'd want to use a big capacitor to
> run one phase of the motor (as shown in the circuit diagram in a
> previous email) both to provide the 3rd phase and to improve the
> power factor.
>
> To me, the bottom line is not, "Will it work". I am confident that you
> *can* run an AC traction motor on plain old 120vac 60 Hz power with
> relatively few extra parts. The motor will run, will produce a
> significant fraction of its rated power, and it won't overheat or be
> otherwise damaged.
>
> The *real* question is, "Will it work well enough to bother with?" It
> may require a big variac and big capacitor bank; that's already the
> beginnings of a crude high-power charger. It might also require a fairly
> high-power AC receptacle and cord to plug it in. And, it might require
> extra switches or contactors to reconnect the motors between driving and
> charging modes.
> --
> "Never doubt that a small group of committed people can change the
> world. Indeed, it's the only thing that ever has!" -- Margaret Meade
> --
> Lee A. Hart 814 8th Ave N Sartell MN 56377 leeahart_at_earthlink.net
--
Victor
'91 ACRX - something different
--- End Message ---
