EV Digest 6803
Topics covered in this issue include:
1) Best paint for Battery racks
by Joe Buford <[EMAIL PROTECTED]>
2) RE: 15k DIY Porsche. 20k Installed.
by [EMAIL PROTECTED]
3) Re: Please help, Aircraft Generator starting torque issue.
by "[EMAIL PROTECTED]" <[EMAIL PROTECTED]>
4) Re: 15k DIY Porsche. 20k Installed.
by GWMobile <[EMAIL PROTECTED]>
5) Re: 42 more to go!
by "Jorg Brown" <[EMAIL PROTECTED]>
6) Re: FW: electric riding mower
by Tim Humphrey <[EMAIL PROTECTED]>
7) Re: Main Contactor Ebay Search Criteria
by Lee Hart <[EMAIL PROTECTED]>
8) Re: Best paint for Battery racks
by Mark Ward <[EMAIL PROTECTED]>
9) Re: Triangle wave generator
by tt2tjw <[EMAIL PROTECTED]>
10) Re: How regen works
by tt2tjw <[EMAIL PROTECTED]>
11) Re: Triangle wave generator
by Danny Miller <[EMAIL PROTECTED]>
12) Re: 15k DIY Porsche. 20k Installed.
by [EMAIL PROTECTED]
13) Re: Triangle wave generator
by tt2tjw <[EMAIL PROTECTED]>
14) Re: Triangle wave generator
by Rod Hower <[EMAIL PROTECTED]>
15) Re: Best paint for Battery racks
by Michael Barkley <[EMAIL PROTECTED]>
16) Re: Triangle wave generator
by "(-Phil-)" <[EMAIL PROTECTED]>
17) Building a Variable Voltage, High Current Batt Charger
by "Scott Littledike" <[EMAIL PROTECTED]>
--- Begin Message ---
What would be the best paint to paint the battery
racks after cleaning and striping them?
Thanks
Joe
____________________________________________________________________________________Choose
the right car based on your needs. Check out Yahoo! Autos new Car Finder tool.
http://autos.yahoo.com/carfinder/
--- End Message ---
--- Begin Message ---
Michael> while i like the 914, i've been wondering if the boxster might
Michael> be a better, more modern equivalent. what does the list think,
Michael> too heavy?
A Boxster's curb weight looks to be around 2850 lbs. The 914 is around 2150
lbs. GVWR is probably more important for this application. GVWR for a 914
is a touch under 2700 lbs. A little googling didn't turn up a GVWR for a
Boxster.
You are going to take more weight off of a Boxster when you dump the ICE,
maybe 100-150 lbs so you'll have a bit more leeway when tossing batteries
back into the chassis. Should you need them I suspect the parts necessary
to handle the higher weight of the batteries (911 front struts, springs,
brakes, etc) will be cheaper for a 914.
--
Skip Montanaro - [EMAIL PROTECTED] - http://www.webfast.com/~skip/
--- End Message ---
--- Begin Message ---
HI-
It's been a few years but Fred Saxby set up my G-32
when we were racing boats.He disconected a small wire
(about 12 or 14 gage) from one of the brush holders. I
believe it was color coded red. He just taped it off
back out of the way. I rember the low torque in reverse
problem when one of the racers tried to ues a G-32
in a boat with an opposite turning prop.
FT.
> [Original Message]
> From: Lee Hart <[EMAIL PROTECTED]>
> To: <ev@listproc.sjsu.edu>
> Date: 5/22/2007 10:14:34 AM
> Subject: Re: Please help, Aircraft Generator starting torque issue.
>
> Michael Barkley wrote:
> > I'm using an Aircraft Generator J29 to be exact, it's
> > not a combo, starter/generator and therefore it seems
> > to be lacking in starting torque as a motor...
> > Please, if anyone can help me with a solution, to my
> > torque problem from a deadstop, I sure would like to,
> > off list talk with you.
>
> I agree that this is on-topic, as a number of people have used, or are
> interested in using these old beasts.
>
> I used a Jack & Heinz 30v 500a aircraft generator in my first EV, too.
> Torque was lower than a series motor, but I made up for it by hitting it
> with 1000+ amps. I used a contactor controller (Zillas had an even
> longer deliver time in those days) :-) Tire burnouts were no problem;
> I'd lose traction before failing to climb any hill.
>
> Both the generators and the starter/generators have series windings. The
> generators have only two large terminals, and a small series field which
> is internally connected as a compensation winding (to boost voltage
> under load so the output voltage does not sag).
>
> The starter/generators have three large terminals, and a larger series
> field. When you connect to the outer two large terminals, you are
> getting the armature and series field in series.
>
> It is important to run it in the right direction. The correct direction
> of rotation was stamped right on mine. Since you can't reverse the
> phasing of the series field, the correct direction will also reveal
> itself if you power the outer two of the three large terminals.
>
> The shunt field has one end internally connected to the armature. This
> also forces it to turn in only the correct direction (assuming the field
> is powered with a supply of the same polarity as the armature). You can
> easily disconnect this internal connection, so both sides of the shunt
> field are available externally. This allows you to get electrical
> reverse *BUT* I found that torque in reverse is significantly lower
> because the two fields are now opposing rather than aiding each other.
>
> What are you using for your field supply? If you simply switch it to a
> constant voltage of less than 30v, you will get a low-speed moderately
> high-torque motor. To increase speed, you must reduce field current; but
> that will also reduce torque.
>
> I found that you can supply a significantly higher shunt field voltage
> to get significantly higher torque -- but only for a very short time
> before the field winding overheats. I had a 72v pack, and could connect
> the field to 72v for perhaps 30 seconds max. This produced drastically
> higher torque; enough to slip the clutch or spin the tires on dry
pavement.
>
> Hope this helps!
> --
> Patent examiner to applicant: "No, your application seem to be legal.
> I'm just having trouble classifying it as intellectual property."
> -- Frank and Ernest
> --
> Lee A. Hart, 814 8th Ave N, Sartell MN 56377, leeahart_at_earthlink.net
>
--- End Message ---
--- Begin Message ---
I used to own a 914 and they are great fun however there is one serious
consideration to think about with them.
Remember unless the goal is a science experiment you probably want a
USEABLE electric car in the end. That means some storage space and maybe
more than two seats.
Alfter all is said and done you want a car that is functional. The only
saving grace for a gas powered 914 which has NO back or space at all
behind the front seats is the front and rear trunk (because the engine
is right behind the seats so you have both a big front and back trunk)
but when you add batteries in those trunks you no longer have that
space.
If you want an electric go cart for fun then a 914 is a great electric
car.
If you want a functional car that you can even put a few grocery bags in
then an electric 914 is useless.
It will have NO storage space. And only one other seat.
It will seriously have NO Storage space. People really don't understand
it until they use one for a while. There are NO nooks and crannies
behind the seat - just a sheer firewall.
Something to keep in mind.
You would hate to spend all the time and money to make an electric only
to realize you don't have a very day to day functional vehicle when you
are finished.
You won't save gas or the environement if it is a car you really won't
drive.
On Tue, 22 May 2007 11:05 am, Michael Wendell wrote:
I was just looking at the www.electroautomotive.com site
that link is http://www.electroauto.com
...and their AC kit is around 15k with batteries. After all
the talk of what's available I'm begining to think that is a
bargain. If you are a miser it will go 150 miles on the flat.
i'm a lurker, still in the dreaming stage with this, but i didn't
realize
their kits had that kind of range. i'm hoping to have the time to
seriously
begin tinkering this fall.
the 914 seems to be a somewhat popular conversion, and electro's kit
may
have something to do with that.
while i like the 914, i've been wondering if the boxster might be a
better,
more modern equivalent. what does the list think, too heavy? i've seen
beat
up boxsters selling for reasonable prices on ebay, and i'd imagine that
parting out the ICE equipment would yield a nice payback.
thanks, i'll go back to lurking.
m.
Michael Wendell
Web/Graphics Guy
Speedgoat Bicycles
724.238.7181
[EMAIL PROTECTED]
www.GlobalBoiling.com for daily images about hurricanes, globalwarming
and the melting poles.
www.ElectricQuakes.com daily solar and earthquake images.
--- End Message ---
--- Begin Message ---
100K and you're thinking you're going to set a record?? Think again.
You can see the records by going to youtube, clicking on the videos
tab on top, making sure the Browse on the left is set to "Most
Viewed", and making sure Time is set to "All Tine".
The winner, at 49 million views, is Evolution of Dance.
http://www.youtube.com/watch?v=dMH0bHeiRNg
If you want to crack the top 10, it's pretty clear that you need
music, dancing, or at least a laughing baby.
On 5/21/07, Mike Willmon <[EMAIL PROTECTED]> wrote:
I barely got to it when I read this and you were already 76 over 100k.
Anyone know how many views would be a record for You Tube?
Mike,
Anchorage, Ak.
> -----Original Message-----
> From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]
> Behalf Of Bill Dube
> Sent: Sunday, May 20, 2007 8:44 PM
> To: ev@listproc.sjsu.edu
> Subject: 42 more to go!
>
>
> If 42 more people look at the KillaCycle Las Vegas "plasma" video, it
> will go over 100,000 views.
> http://www.youtube.com/watch?v=3dRpAZci9m0
> The Firebird video is over 94,000 views as well, so a few thousand
> more will push it over 100,000
> http://www.youtube.com/watch?v=GDHJNG2PngQ
>
> Bill Dube'
>
>
--- End Message ---
--- Begin Message ---
www.elec-trak.org
>
>
> -----Original Message-----
> From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On
> Behalf Of Jeff Major
> Sent: Tuesday, May 22, 2007 11:40 AM
> To: ev@listproc.sjsu.edu
> Subject: Re: electric riding mower
>
>
> Hi midiguy732,
>
> You got a better name? Anyway, I think electric
> riding mowers (or just the tractor part) are a lot of
> fun. I have a 30 year old(?) Sears tractor which
> originally had a 10 to 12 hp engine. Bought it 15
> years ago without engine or deck for like $20. Had
> decent tires and seat. Put in an old 6" dia 24 volt
> floor sweeper motor and a 48 volt Crutis 275 amp
> controller and 4 garden tractor batteries. Kept the 3
> spd transmission (or transaxle). Works great. I
> don't mow with it, but about everything else. Lawn
> sweeping puts the biggest load on it, about 40 amps.
> So I cannot do that very long, but who wants to?
>
> So as far as what size motor to use, for
> traction(propulsion) a 6 to 6.6" dia motor with a 275
> amp limit at 36 or 48 volts is probably more than
> enough. I can do Wheelies if I fool the controller.
> Don't have many hills, but never met a ramp I could
> not do even pulling a half ton of firewood or dirt or
> rock.
>
> Blade motors are a bit trickey. You probably are
> going have trouble with that. Direct drive (blade
> attached to the motor shaft) can be difficult and if
> you hit a rock and bend the shaft, the motor is
> screwed. You might consider using the belt-pulley
> system and a single motor to drive both blades. A
> 6.6" dia motor would probably work here as well. I
> did a motor drive for a guy with a big 3 blade deck
> using a single 7.2" motor. Works great.
>
> Another thing I'd suggest, look up Elec Trak mowers by
> GE from the early 70's. You can find them sometimes
> on eBay, or parts or manuals.
>
> Good luck and have fun.
>
> Jeff
>
>
>
> --- midiguy732 <[EMAIL PROTECTED]> wrote:
>
>> > Anyone used a bully electric mower? I'm thinking
>> about using one as
>> the
>> > basis for a R/C electric mower. (There, it's back
>> on topic, sortof)
>>
>> No, but I am considering converting my gas powered
>> murray to a fully
>> electric mower. I did the robot-mower thing
>> already, but with an old
>> riding mower deck repowered with a B&S engine simply
>> because of parts
>> availability.
>>
>> When I joined this list a few weeks ago, I posted
>> about this but the
>> post didn't make it through for some reason. I'll
>> repost.
>>
>>
>> Hello folks,
>>
>> I recently joined this list because electric
>> vehicles have always
>> fascinated me, even though there are some drawbacks
>> as compared to
>> gasoline/diesel vehicles, mostly relating to range.
>>
>> Before I embark into a more expensive project like a
>> vehicle conversion,
>> I was pondering taking a "baby step" and converting
>> an old riding mower
>> to fully electric, mostly as a learning tool about
>> the process,
>> fabrication required, and enhance my somewhat basic
>> of electrics.
>>
>> After researching fairly extensively (on and off for
>> approximately a
>> month or so) I'm left with a few questions, which I
>> wish to pose to the
>> membership.
>>
>> The riding mower has a 12HP B&S engine, with two
>> pulleys. One to drive
>> the transmission, one to drive the dual 22" mower
>> blades. Obviously I
>> could do this conversion the "easy way" and purchase
>> a large motor and
>> extend the shaft to drive the two belt systems,
>> leaving the motor to run
>> at 3000-3600 RPM and call it a day.
>>
>> It's unclear to me how to properly size a motor to
>> replace a gasoline
>> engine. I understand gasoline engines are rated at
>> max RPM and peak HP
>> and torque figures are given whereas electric motors
>> are rated with
>> their continuous duty hp/torque figures. Are there
>> any mathematics or
>> at least a "rule of thumb" that might offer some
>> clarity in sizing?
>>
>> Another option I thought about was to purchase three
>> smaller motors.
>> Two motors that would drive the two blades on the
>> mower deck, and one
>> motor to drive the transmission directly. The
>> advantage I see with this
>> method would be to fully eliminate the two clutch
>> mechanisms which are
>> simply pulleys that move, tightening the two belts.
>> These mechanisms
>> are troublesome at best, and wear the belts out
>> fairly quickly (I find
>> flipping the mower on its side to replace them
>> annoying, but that's me!)
>> I could instead use two contactors or relays - one
>> switch/button to
>> actuate the drive motor, one to actuate the mower
>> deck.
>>
>> So I guess my second question is also about sizing.
>> How much HP or
>> wattage would an electric motor need to be to mow
>> grass with a razor
>> sharp 22" blade?
>>
>> It seems logical to me to size the motor(s)
>> correctly first based on HP
>> and torque requirements, then use those figures to
>> determine wiring,
>> controller, and battery sizing.
>>
>> Because the mower has a 5-sp transmission, I would
>> imagine I don't need
>> to install a variable speed controller though maybe
>> for the drive motor
>> if I go with separate motors. I'd also like to
>> retain the transmission
>> because there are cases where having a low first and
>> second gear has
>> been useful (I've towed 1500lbs of patio blocks in a
>> trailer with this
>> mower).
>>
>> I see no reason for the mower deck to have any
>> controller other than
>> current limitation to protect the motors. On/Off
>> would be more than
>> adequate.
>>
>> Comments, thoughts, flames welcomed.
>>
>>
>
>
>
>
> ____________________________________________________________________________
> ________
> No need to miss a message. Get email on-th
--
Stay Charged!
Hump
I-5, Blossvale NY
--- End Message ---
--- Begin Message ---
Steve Condie wrote:
> how hard is it to add blowout magnets?
Not easy when the contactor wasn't designed for it.
The fundamental problem with this contactor is that it only has a single
contact, like a relay; not two like a high-power contactor. Thus, it
doesn't have the spacing between the off-state contacts to sustain high
voltage. The contacts are physically large (good for high current), but
bad for adding blowout magnets (because stronger ones are required to
span the large horizontal distance).
It is designed from the ground up to switch high current at LOW
voltages. I think the only practical way to use it in a high-voltage EV
would be with several in series, all switched at the same time.
--
Ring the bells that still can ring
Forget the perfect offering
There is a crack in everything
That's how the light gets in -- Leonard Cohen
--
Lee A. Hart, 814 8th Ave N, Sartell MN 56377, leeahart_at_earthlink.net
--- End Message ---
--- Begin Message ---
Try duplicolor truck bed coating available at Walmart or Autozone. I used it
in my battery case and parts exposed to elements. Very durable like Rhino coat
which will also work.
---- Joe Buford <[EMAIL PROTECTED]> wrote:
>
> What would be the best paint to paint the battery
> racks after cleaning and striping them?
> Thanks
> Joe
>
>
>
> ____________________________________________________________________________________Choose
> the right car based on your needs. Check out Yahoo! Autos new Car Finder
> tool.
> http://autos.yahoo.com/carfinder/
>
--- End Message ---
--- Begin Message ---
Danny Miller wrote:
> It is far easier to design a system- and do it right- with a uC.
That was the impression I came to after reading many application notes
and reference designs (from IR, POWEREX, SEMIKRON, FAIRCHILD etc) for
intelligent IGBT modules, and IGBT drivers, they all seem to be designed
around a microcontroller.
Chet Fields wrote:
> it may not be a good idea as it may lock up and cause the MOSFETs to
be stuck on
the IGBT drivers I have looked at have integrated protection to deal
with this situation, I was assuming that in a mass produced driver this
function would be implemented correctly and reliably.
Danny Miller wrote:
That's pretty close to nonsense. IIRC the person claiming this has no
experience with microcontrollers.
A uC is a fine way to generate a PWM signal. It is far easier to
design a system- and do it right- with a uC. There's an awful lot to
designing an analog controller. The analog controller has a lot more
that can go wrong with it. You need a lot of inside information on
how each type of component may fail to start to work the problems out.
It's important to understand however that being able to generate a PWM
is only a small part of the design problem for an EV controller. The
output transistors are large and expensive to obtain and very tricky
to mount where you can meet their extreme thermal, mechanical, and
electrical needs. The transistor driver, where we take 0 to 5v or
whatever PWM and turn it into a transistor drive with the appropriate
voltage level and best waveform shape is tricky too.
Danny
Chet Fields wrote:
I had thought first of using a microcontroller, but then good advice
from the list suggested that
it may not be a good idea as it may lock up and cause the MOSFETs to
be stuck on. I don't really
know of a good way to protect from that happening except for maybe
some fuses or something.
I did think that a microcontroller like a PIC would give me alot of
flexibility though. I am a
programmer professionally so I would definitely be more in my element
instead of designing these
discrete analog circuits, although I have to admit it has been and
still looks like a lot of fun
if I can get it working.
I thought of starting with the analog and then moving later (perhaps)
into the PIC if I can come
up with the proper scheme to protect against a dangerous situation
with a lock up.
Chet
--- tt2tjw <[EMAIL PROTECTED]> wrote:
Chet,
is there any reason not to use a microcontroller with PWM outputs?
Chet Fields wrote:
I am trying to generate the triangle wave for the PWM circuit and I
am using the good ol' 741
Op
____________________________________________________________________________________
Looking for earth-friendly autos? Browse Top Cars by "Green Rating"
at Yahoo! Autos' Green Center.
http://autos.yahoo.com/green_center/
--- End Message ---
--- Begin Message ---
Thanks Phil,
this sent me in the right direction, "boost converter" on wikipedia,
I have learnt that;
back emf=V=dLambda/dt=L(di/dt) + i (dL/dt)
and since L is nearly constant
V=L(di/dt)
so if the switch is on for D.T seconds and off for (1-D).T seconds in
every cycle and if the switching frequency is high enough to prevent all
the energy stored in the inductor from transferring in one part of the
cycle, then;
when the switch is closed
deltaI=V.D/L
when the switch is opened
deltaI=deltaV(1-D)/L
and so
deltaV=V.D/(1-D)
or
D=1-(V generated in armature) / (desired V to charge batteries)
I guess that (provided the switching frequency is high enough) the above
is a good estimate of the required value of D to charge the batteries.
Probably there are other variables too and also V generated in the
armature will be changing. So, the charging current should be monitored,
if too high then increase D if too low then decrease D.
Have I got the right idea?
(-Phil-) wrote:
If you use a switch (chopper) to "short" the armature momentarily,
then the current that was flowing there has no where to go all the
sudden, and begins to build up as a magnetic field in the armature.
This is essentially "storing" the energy as inductance. Then the
switch opens, and the magnetic field collapses, it's lines of force
cutting back across the windings (just the opposite of what was
happening an instant ago.) It induces a high voltage in the coils and
that turns the freewheeling diode on and dumps the now (much higher
voltage) into the battery bank. The switching is done very fast, so
the magnetic field never reaches saturation.
This is the same principle used in a "boost converter" but rather than
using a separate inductor, we just use the rather large (and very
handy) one that's in the motor.
-Phil
----- Original Message ----- From: "tt2tjw" <[EMAIL PROTECTED]>
To: <ev@listproc.sjsu.edu>
Sent: Friday, May 18, 2007 10:54 AM
Subject: Re: SepEx Regen Chopper - Help!
Chet
Citroen sepex controllers are arranged like this,
__________
| | |
/ /\ |
_______| | ---- +
| | | - -
motor / /\ |
| | | |
-----------------
"Go" chopper on the high side. "Regen" chopper on the low side
As far as I understand you don't need the regen chopper untill you
are below full field full voltage RPM, above that ROM current will
flow through the free wheel diode into the battery.
Like you I still don't understand how regen occurs at low RPM, I have
been told it is dependent on the inductance of the armature, I
haven't had time to read up about inductance but will let you know if
I do,.... or hopefully someone more knowledgeable will reply first.
Personally I'm not that bothered about regen at low RPM but I'm not
sure if there will be a problem for the motor if it isn't implemented.
I have a manual which describes the principle but not the finer
details of my controller, its in French, I am emailing it to you.
I can also send some pictures of the internals of the controller, if
that will help you, but not immediately.
go well,
Tom Ward.
Chet Fields wrote:
Even after reviewing all the Regen info I could find in both the
archives and the web in general I
was hoping someone could help clarify a couple of specific points on
it. Keep in mind I am talking
about a brushed DC SepEx motor.
_____________________________________________
| | | |
| 'Regen' | 'Go' | / \
--- Chopper / Freewheel _____ /Motor\
- / Diode / \ \ /
--- | | \ /
- | | |
| |__________________|__________|
| | | | 'Regen' | |
| Freewheel? | 'Go' / | Diode _____
Chopper /
| / \ | | | |
| | |
______________|_____________________________|
Assuming that the choppers are PWM driven MOSFETs during 'motoring'
current will flow from the
battery through the motor and the 'Go' chopper when on and through
the 'Go' freewheel diode when
off. The battery voltage is always greater than the motor voltage in
this scenario so therefore
current flows in this direction. (I'm talking positive to negative,
conventional? flow) So up to
this point it's pretty clear to me.
However, when we leave the 'Go' chopper off (open) and I apply full
field current this is where it
gets fuzzy. What exactly is happening to the armature windings? They
are now rotating through the
fields' flux and therefore should be creating a voltage across the
windings correct? However,
because they are not hooked to anything there is no current flow.
And because there is no current
there is no 'back' torque or stopping power?
And the voltage that is produced is proportional to the RPM correct?
Which for this particular
motor would be full pack voltage at base speed, right? Or, if there
was any field weakening the
correspondingly lower voltage? In other words, with field weakening
the RPMs would need to rise to
match the RPMs that the motor would spin at full pack voltage with
that particular field strength?
So in order to get current to actually flow back into the battery I
need a greater voltage, and I
will only get that if the motor is rotating at greater than base
speed right? And below base speed
there is no way I could get any regen unless somehow I can boost the
voltage. This is where I need
some help. If we were to 'chop' the regen, as the regen mosfet is
closed this basically shorts the
motor windings allowing current to flow (ramp?) through the windings
and the regen mosfet. Then
when the mosfet is switched off, because of the inductance of the
armature windings the current
continues to flow as the field collapses. But is this voltage
proportional to the duty cycle and
actually higher than the battery pack voltage at some duty cycle
less than 100% and RPM below base
speed? Is this how you can get regen at lower speeds?
And if this is all true, then how can you control the voltage
produced in the off cycle? Do you
simply need to reduce the duty cycle or do you also need to adjust
the field current?
A final question, should I be worried about generating too high a
voltage and protect the
batteries and capicitors (omitted from the diagram) somehow with
like a high power zener (if there
is such a thing)?
I will have 108V nominal battery pack. Is there such a thing as a
135V or so Zener? How much power
would it need to dissipate? Or, should I try to simply keep the
voltage down with the duty cycle?
Thanks for any help,
Chet Fields
____________________________________________________________________________________Be
a better Globetrotter. Get better travel answers from someone who
knows. Yahoo! Answers - Check it out.
http://answers.yahoo.com/dir/?link=list&sid=396545469
--- End Message ---
--- Begin Message ---
You can design a lot of great features into a uC-driven motor controller
too. It's simple and easy to make them effective.
Still, a reminder- additional features are of limited use. The primary
problem is still the power stage and its driver. Having a PWM generator
10x more advanced than an analog one won't do much to reduce the
problem. Well, advanced thermal shutdown procedures, testing the
ON-state voltage drop, and overcurrent detection can reduce the demands
on the transistors, but generally not a remarkably high reduction, if any.
Danny
tt2tjw wrote:
Danny Miller wrote:
> It is far easier to design a system- and do it right- with a uC.
That was the impression I came to after reading many application notes
and reference designs (from IR, POWEREX, SEMIKRON, FAIRCHILD etc) for
intelligent IGBT modules, and IGBT drivers, they all seem to be
designed around a microcontroller.
Chet Fields wrote:
> it may not be a good idea as it may lock up and cause the MOSFETs to
be stuck on
the IGBT drivers I have looked at have integrated protection to deal
with this situation, I was assuming that in a mass produced driver
this function would be implemented correctly and reliably.
Danny Miller wrote:
That's pretty close to nonsense. IIRC the person claiming this has
no experience with microcontrollers.
A uC is a fine way to generate a PWM signal. It is far easier to
design a system- and do it right- with a uC. There's an awful lot to
designing an analog controller. The analog controller has a lot more
that can go wrong with it. You need a lot of inside information on
how each type of component may fail to start to work the problems out.
It's important to understand however that being able to generate a
PWM is only a small part of the design problem for an EV controller.
The output transistors are large and expensive to obtain and very
tricky to mount where you can meet their extreme thermal, mechanical,
and electrical needs. The transistor driver, where we take 0 to 5v
or whatever PWM and turn it into a transistor drive with the
appropriate voltage level and best waveform shape is tricky too.
Danny
Chet Fields wrote:
I had thought first of using a microcontroller, but then good advice
from the list suggested that
it may not be a good idea as it may lock up and cause the MOSFETs to
be stuck on. I don't really
know of a good way to protect from that happening except for maybe
some fuses or something.
I did think that a microcontroller like a PIC would give me alot of
flexibility though. I am a
programmer professionally so I would definitely be more in my
element instead of designing these
discrete analog circuits, although I have to admit it has been and
still looks like a lot of fun
if I can get it working.
I thought of starting with the analog and then moving later
(perhaps) into the PIC if I can come
up with the proper scheme to protect against a dangerous situation
with a lock up.
Chet
--- tt2tjw <[EMAIL PROTECTED]> wrote:
Chet,
is there any reason not to use a microcontroller with PWM outputs?
Chet Fields wrote:
I am trying to generate the triangle wave for the PWM circuit and
I am using the good ol' 741
Op
____________________________________________________________________________________
Looking for earth-friendly autos? Browse Top Cars by "Green Rating"
at Yahoo! Autos' Green Center.
http://autos.yahoo.com/green_center/
--- End Message ---
--- Begin Message ---
* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
* ---REMAINDER OF MESSAGE TRUNCATED--- *
* This post contains a forbidden message format *
* (such as an attached file, a v-card, HTML formatting) *
* Lists at sjsu.edu only accept PLAIN TEXT *
* If your postings display this message your mail program *
* is not set to send PLAIN TEXT ONLY and needs adjusting *
* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
--- End Message ---
--- Begin Message ---
Danny,
by power stage do you mean busbars , cooling, mountings ?
by driver do you mean the IGBT driver? If a driver board such as BG2A
from powerex is used does that solve the "driver" problem or is the
"driver" problem more than just this?
Tom Ward
Danny Miller wrote:
You can design a lot of great features into a uC-driven motor
controller too. It's simple and easy to make them effective.
Still, a reminder- additional features are of limited use. The
primary problem is still the power stage and its driver. Having a PWM
generator 10x more advanced than an analog one won't do much to reduce
the problem. Well, advanced thermal shutdown procedures, testing the
ON-state voltage drop, and overcurrent detection can reduce the
demands on the transistors, but generally not a remarkably high
reduction, if any.
Danny
tt2tjw wrote:
Danny Miller wrote:
> It is far easier to design a system- and do it right- with a uC.
That was the impression I came to after reading many application
notes and reference designs (from IR, POWEREX, SEMIKRON, FAIRCHILD
etc) for intelligent IGBT modules, and IGBT drivers, they all seem to
be designed around a microcontroller.
Chet Fields wrote:
> it may not be a good idea as it may lock up and cause the MOSFETs
to be stuck on
the IGBT drivers I have looked at have integrated protection to deal
with this situation, I was assuming that in a mass produced driver
this function would be implemented correctly and reliably.
Danny Miller wrote:
That's pretty close to nonsense. IIRC the person claiming this has
no experience with microcontrollers.
A uC is a fine way to generate a PWM signal. It is far easier to
design a system- and do it right- with a uC. There's an awful lot
to designing an analog controller. The analog controller has a lot
more that can go wrong with it. You need a lot of inside
information on how each type of component may fail to start to work
the problems out.
It's important to understand however that being able to generate a
PWM is only a small part of the design problem for an EV
controller. The output transistors are large and expensive to
obtain and very tricky to mount where you can meet their extreme
thermal, mechanical, and electrical needs. The transistor driver,
where we take 0 to 5v or whatever PWM and turn it into a transistor
drive with the appropriate voltage level and best waveform shape is
tricky too.
Danny
Chet Fields wrote:
I had thought first of using a microcontroller, but then good
advice from the list suggested that
it may not be a good idea as it may lock up and cause the MOSFETs
to be stuck on. I don't really
know of a good way to protect from that happening except for maybe
some fuses or something.
I did think that a microcontroller like a PIC would give me alot of
flexibility though. I am a
programmer professionally so I would definitely be more in my
element instead of designing these
discrete analog circuits, although I have to admit it has been and
still looks like a lot of fun
if I can get it working.
I thought of starting with the analog and then moving later
(perhaps) into the PIC if I can come
up with the proper scheme to protect against a dangerous situation
with a lock up.
Chet
--- tt2tjw <[EMAIL PROTECTED]> wrote:
Chet,
is there any reason not to use a microcontroller with PWM outputs?
Chet Fields wrote:
I am trying to generate the triangle wave for the PWM circuit and
I am using the good ol' 741
Op
____________________________________________________________________________________
Looking for earth-friendly autos? Browse Top Cars by "Green Rating"
at Yahoo! Autos' Green Center.
http://autos.yahoo.com/green_center/
--- End Message ---
--- Begin Message ---
Chet,
I just did a simple brush motor control with an 8 pin
Microchip MCU (12F683) that cost about $1.00 at
digikey. If you want the code and circuit I'll send
it to you. The 12F683 is somewhat limited for motor
control, I just did it as a side project at work.
My preferred device is a dsPIC30F2010 that cost about
$8 at digikey (I'm paying MUCH less than this at work
in volume). I don't have brush motor code for this
one, but it would be simple to change my brushless DC
code to do brush motor.
This route gives you much more flexibility after the
hardware is built (assuming it works). More up-front
learning is required, but is definitely worth the
effort.
Rod
--- tt2tjw <[EMAIL PROTECTED]> wrote:
> Chet,
>
> is there any reason not to use a microcontroller
> with PWM outputs?
>
> Chet Fields wrote:
> >
> > I am trying to generate the triangle wave for the
> PWM circuit and I am using the good ol' 741 Op
> >
> >
> >
>
>
--- End Message ---
--- Begin Message ---
I painted everything that could possibly come into
contact with battery acid, with "Spray on Bedliner"
material. Now you can purchase it in a spray can for
about $8 at WallyWorld, and most auto parts stores.
We used it at a metal pickling plant on all exposed
metal that we didn't want the sulphuric acid to
attack. We had tested it's durability by coating a
piece of steel, and dipping it in the 10% solution of
sulphuric acid for over a month, and it did not get to
the metal.
--- Joe Buford <[EMAIL PROTECTED]> wrote:
>
> What would be the best paint to paint the battery
> racks after cleaning and striping them?
> Thanks
> Joe
>
Michael Barkley
www.texomaev.com
--- End Message ---
--- Begin Message ---
Most microcontrollers have watchdogs. Lockups just don't happen unless you
subject them to electrically hostile environments. (provided you don't have
buggy software, that is!)
Best advice is to use an off-the-shelf MOSFET driver chip. Many of them do
all the dirty work for you, and if you are building a T-Bridge, then they
have protection from accidental shoot-through, and the needed dead-band gap.
When you start out, build a test bed with a small motor and a
current-limited supply. Run the MCU off a totally separate power supply.
Then get your software right. Once that's done, you can scale up the design
by adding more MOSFETs in parallel. Remember to make sure your gate drive
circuitry is up to the task. MOSFET gates have high relative capacitance,
so to switch them fast enough to not be a problem, sometimes that involves
several peak amps! (don't switch TOO fast though!)
It's also easier to start off with a relatively low switching frequency. It
gets harder to maintain efficiency at higher frequencies, but optimally you
want the switching to be above human hearing range otherwise it will be
annoying. (I run mine at 18khz)
Once you've got the "hang" of everything, you can slowly increase
frequencies and watch your waveforms and heat.
-Phil
----- Original Message -----
From: "Chet Fields" <[EMAIL PROTECTED]>
To: <ev@listproc.sjsu.edu>
Sent: Tuesday, May 22, 2007 10:40 AM
Subject: Re: Triangle wave generator
I had thought first of using a microcontroller, but then good advice from
the list suggested that
it may not be a good idea as it may lock up and cause the MOSFETs to be
stuck on. I don't really
know of a good way to protect from that happening except for maybe some
fuses or something.
I did think that a microcontroller like a PIC would give me alot of
flexibility though. I am a
programmer professionally so I would definitely be more in my element
instead of designing these
discrete analog circuits, although I have to admit it has been and still
looks like a lot of fun
if I can get it working.
I thought of starting with the analog and then moving later (perhaps) into
the PIC if I can come
up with the proper scheme to protect against a dangerous situation with a
lock up.
Chet
--- tt2tjw <[EMAIL PROTECTED]> wrote:
Chet,
is there any reason not to use a microcontroller with PWM outputs?
Chet Fields wrote:
>
> I am trying to generate the triangle wave for the PWM circuit and I am
> using the good ol' 741
Op
>
>
>
____________________________________________________________________________________
Looking for earth-friendly autos?
Browse Top Cars by "Green Rating" at Yahoo! Autos' Green Center.
http://autos.yahoo.com/green_center/
--- End Message ---
--- Begin Message ---
Hi there,
I was just wondering how complex it is to build a charger for an EV, with
lithium batteries in mind. Taking A123 cells as an example, they should be
charged at 3.6V per cell, at a maximum of 3A until the current drops off.
However, if you look at a pack like bill dubes killacycle pack, there are 9
batteries in parallel, and 110 rows of these!
Looking at the 9 batteries in parallel, you must supply 3.6V at 27A to
charge each cell at 3A.. if you have a 150V system, thats 27A at 150V. As if
that wasnt difficult enough, a BMS should be used, which may involve
changing the output voltage to below 150V. so, thats a variable voltage
charger, 0 - 150V, at 27A!
How is this normally done? An appropriate transformer would need a 4kVA
rating, which would be a very large transformer? and even then, you would
not be able to change voltage, unless you use a giant potential divider (two
resistors in series, from +ve to gnd, with a tap between them).
suitable variacs are expensive and very large. plus, there is the difficulty
of handling such a high power supply - how do you rectify and smooth it etc
- normal components cant be used.
Or, do you make lots of 3.6V 27A chargers and connect them to each
individual "bank"?
chargers such as the PFC20 dont look very large, and you can simply change
the voltage with a trimmer pot...
If anyone can shed some light on how the ac voltage is reduced, rectified
and smoothed, whilst being completely variable, i would be most grateful!
Many Thanks
Scott
_________________________________________________________________
Could you be the guest MSN Movies presenter? Click Here to Audition
http://www.lightscameraaudition.co.uk
--- End Message ---
