EV Digest 5063
Topics covered in this issue include:
1) Re: Matt's 240SX Electric in the top 15!
by "Lawrence Rhodes" <[EMAIL PROTECTED]>
2) Re: I want to build a PWM DC motor controller
by Cory Cross <[EMAIL PROTECTED]>
3) Federal EV tax credit
by [EMAIL PROTECTED]
4) Drag240SX.com site
by "Matthew D. Graham" <[EMAIL PROTECTED]>
5) Re: I want to build a PWM DC motor controller
by "Stefan T. Peters" <[EMAIL PROTECTED]>
6) Re: I want to build a PWM DC motor controller
by "Arthur W. Matteson" <[EMAIL PROTECTED]>
7) Re: Federal EV tax credit
by Danny Miller <[EMAIL PROTECTED]>
8) Re: Is it bad to spin backwards?
by "Mark Hanson" <[EMAIL PROTECTED]>
9) Re: I want to build a PWM DC motor controller
by "Stefan T. Peters" <[EMAIL PROTECTED]>
10) Re: Fw: Geo Metro EV on a budget
by "jmygann" <[EMAIL PROTECTED]>
11) RE: refractometer vs hydrometer...
by "Roger Stockton" <[EMAIL PROTECTED]>
12) Re: I want to build a PWM DC motor controller
by "Stefan T. Peters" <[EMAIL PROTECTED]>
13) Re: Is it bad to spin backwards?
by Jim Husted <[EMAIL PROTECTED]>
14) RE: I want to build a PWM DC motor controller
by Cor van de Water <[EMAIL PROTECTED]>
15) Re: Federal EV tax credit
by Neon John <[EMAIL PROTECTED]>
16) RE: Still needing prof. battery advice. Solutions:
by "Roger Stockton" <[EMAIL PROTECTED]>
17) Re: Federal EV tax credit
by "Stefan T. Peters" <[EMAIL PROTECTED]>
18) Re: Federal EV tax credit
by Eric Poulsen <[EMAIL PROTECTED]>
19) Re: I want to build a PWM DC motor controller
by "Stefan T. Peters" <[EMAIL PROTECTED]>
--- Begin Message ---
It is now one vote from second place.
If you can, please go to
http://www.dragtimes.com/index.php
and vote for the Datsun 1200 Electric which is currently in 4th
place. You can find it in the bottom right corner frame of the site.
You may vote once per day, every day.
Lawrence Rhodes.....
--- End Message ---
--- Begin Message ---
Danny Miller wrote:
...
So consider a 10 milliohms total resistance in some interconnects. In
the PWM case we lose 100 watts (I^2*R) during the "on" period, since
it's a 50% duty, that's 50W average. But delivering the same 50 amps
constant is only 25W average loss. The one caveat here though is that
the losses in additional interconnects, MOSFET rds-on, and diode drops
in the specified BattPack schematic appears far larger than any gains
from reducing the PWM peak-to-peak difference. Relying on those diodes
to bypass 5 batteries would cost around 250W, not even considering
MOSFET rds-on.
The motor current is a constant 50A +/- a small ripple, since the motor
acts as an inductor. Thus, even with PWM, the loss will be 25W. However,
like you said, the diodes will waste a lot of power. With just one
battery, 1.2V x 7 off-units loses 8.4V, which is 17.5% of the power from
a 48V battery (or 70% of a 12V battery!). Maybe have a contactor engaged
over the diodes unless they may be used next?
At top speed, the diodes won't be used at all, only the low-loss
MOSFETs, so it would be more efficient at this point than an IGBT
controller.
Cory Cross
--- End Message ---
--- Begin Message ---
According to the following page, only new EVs that "have never been used as a
non-electric vehicle" are eligible for the tax credit.
http://www.fueleconomy.gov/feg/tax_afv.shtml#ev
Does anybody have a defensible way of claiming the credit for a conversion?
Cheers,
Jay Donnaway
www.karmanneclectric.blogspot.com
--- End Message ---
--- Begin Message ---
Sweet!
I'm so glad I went back to look over the hundreds of posts from the last
couple weeks while I was away. Cor, that's a great find, and I'll definitely
post a record on the drag240sx.com site. Joule Injected's first run ever
puts it in the top 50 list of 240SX times, and I'm hoping to move it up a
spot or ten over the next two weeks!
I also noticed that there were many people stoking the voting fire during
that time period. That is really fantastic--thanks guys! It's helped to
inspire me to contact local and national media (just clubs I belong to so
far) to help get the word out.
I know not everyone on the list is as one-track-minded when it comes to high
performance EVs. But I believe these examples will help to improve the
general public's perception of electric vehicles. If nothing else, it will
serve as a talking point, just to get people to consider an alternative to
driving a gas car every day. The reception I got from the gas and (yes)
diesel racers at Moroso last month really helped drive home the fact that
for the most part, people *can* be open-minded. I never expected the kind of
support they showed for me, but like Rod said before, they really are an
accepting bunch of tinkerers, just trying to improve their cars and the
sport.
Matt Graham
300V "Joule Injected" Nissan
http://www.jouleinjected.com
Hobe Sound, FL
-----Original Message-----
From: Cor van de Water [mailto:[EMAIL PROTECTED]
Sent: Saturday, December 31, 2005 1:49 AM
To: [email protected]
Subject: RE: Zombie at DragTimes Web Site - voting
John,
Leafed through the Dragtimes site but could not find it.
Do you have the link page?
BTW: Matt can post his car also on http://www.drag240sx.com/
Cor van de Water
Systems Architect
Email: [EMAIL PROTECTED] Private: http://www.cvandewater.com
Skype: cor_van_de_water IM: [EMAIL PROTECTED]
Tel: +1 408 542 5225 VoIP: +31 20 3987567 FWD# 25925
Fax: +1 408 731 3675 eFAX: +31-87-784-1130
Proxim Wireless Networks eFAX: +1-501-641-8576
Take your network further http://www.proxim.com
-----Original Message-----
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]
Behalf Of John Wayland
Sent: Friday, December 30, 2005 8:01 PM
To: [email protected]
Subject: Re: Zombie at DragTimes Web Site - voting
Hello to All,
Rod Hower wrote:
>You can vote once per day, so I voted again today and also voted at my
>friends computer.
>Rod
>
>
>
Thanks everyone for voting, last count it was up to 123 votes, and an
electric car is # 57 of the top 100 favorite time slips! Let's do it again
tomorrow and build it up higher. Tell your friends with computers, too.
I think we've gotten the attention of those who run the site, as I got a
nice email from Brooks, telling me he had put up links to a Zombie video and
the Plasma Boy racing site...cool!
See Ya....John Wayland
--- End Message ---
--- Begin Message ---
Danny Miller wrote:
Resistance appears greater with PWM. Say the load requires an average
voltage of half the pack voltage at 50 amps average current and thus
we use a 50% duty cycle. So the batteries put out basically 100 amps
half the time, whereas a single pack would deliver 50 amps all the time.
I think that you can almost make that issue go away with the same way
that PWM-based controllers do: stiffening caps. But the difference is
that these would be low voltage/very high capacitance. Nice thing is
that there is a whole industry built around those, so they are easy to
find/not too expensive (around $50/Farad at 40-50VDC). Another nice
thing is that they are an *optional* add-on, so if you don't want to
spend the extra $200-$400, you don't have to, and it will all work fine.
So consider a 10 milliohms total resistance in some interconnects. In
the PWM case we lose 100 watts (I^2*R) during the "on" period, since
it's a 50% duty, that's 50W average. But delivering the same 50 amps
constant is only 25W average loss. The one caveat here though is that
the losses in additional interconnects, MOSFET rds-on, and diode drops
in the specified BattPack schematic appears far larger than any gains
from reducing the PWM peak-to-peak difference. Relying on those
diodes to bypass 5 batteries would cost around 250W, not even
considering MOSFET rds-on.
The Rds(on) of the MOSFETs is ~4 milliOhms, but can be brought down to 3
with additional gate voltage over 5 (add a transistor/resistor). There
are four in parallel in each Power Module. The diodes are silicon power
diodes with a max forward voltage drop of 1.35V @ 400A, the max forward
power loss is ~1W/A with DC use, and there is one per MOSFET (cost
decision). I would hazard to guess that the bypass diodes will always be
the greatest loss in a BatPack setup. Let's say that "cruising" uses
100A @ 48V (50% PWM duty cycle on a 96V pack drawing 50 battery amps):
= 4 modules on
= 4 modules off
= 16 diodes bypassing @ 25W each (100A/4 diodes parallel)
= 400W bypass diode loss <- usable but eck!
I knew that using the $1.20 diodes would be less efficient. I'm just
trying for a "lowest common denominator" first, then sprucing it up
later. For $10.20, you can get a 1.35V @ 5000A diode which shows 0.7W/A
losses. You only need one per Power Module, and that would be:
= 4 diodes bypassing @ 70W each
= 280W bypass diode loss <- quite livable
Now the Power Modules I'm making right now (the cheap $40-$70 ones)
could handle 24V tops I'd guess based on the 40V MOSFET clamp. At least
that is what I'm going to test them to. They should be quite comfortable
at 18V. So If you want better efficiency, you could run less of them
with three 6V series batts per module. So you could run 6 @ 18V. Then
under the same conditions (4.8Kw load), you would be running 88A @ 54V
(half throttle again). This would give you:
= 3 diodes bypassing @ 61.6W each
= 184.8W bypass diode loss <- getting better!
I wonder what a longer string would be? 4.8Kw load with eight Power
Modules at 18V = 72V @ 67A:
= 4 diodes bypassing @ 46.9W each
= 187.6W bypass diode loss <- about the same...
This is one reason why I added the series/parallel switching. At half
throttle and less, you would be in "low voltage/high PAM" mode. This way
there are less (or none) diodes bypassing. So at 100A @ 48V in "low" mode:
= 0 diodes bypassing @ 0W each
= 0W bypass diode loss <- bingo, max efficiency!
It's kinda like the effect of a lockup torque converter. At certain
common conditions (cruising for example), it would "lockup" the system,
eliminating or reducing any transmission/switching losses it can. The
bonus is that you have double the amps available (800) when you are
likely to need it: slower motor RPM, but only have to pay for 400A of
silicon. ;)
--
Stefan T. Peters
--- End Message ---
--- Begin Message ---
> That makes 500A at 1 hundredth of a second = 0.5F. I can get a hold of
> those fine, then my question is back to the original one: How much would
> it help to decrease the spurious load on the bats, and hopefully
> increase the life?
It will change the Peukert effect:
http://www.amplepower.com/pwrnews/beer/
Let's say you need 100A for 0.0002 seconds and 0A for 0.0008 seconds.
(This is one-fifth throttle, not one-fourth.) Your average current will
be 20A in this case. Let's also assume that this draw happens for one
hour continuously. For an 'n' of 1.1 you need 27Ah of capacity if a
constant 20A were applied, but with ripple this becomes about 32Ah. In
the "ripple" car you can only go 84% as far if you drive it until it
dies (and don't let it recover).
> Is it just the motor that is "singing"? BTW, that TV flyback frequency
> has always bugged the bejeezus out of me. That supersonic whistling that
> TVs make whenever they are on, right? I bought an LCD monitor & TV as
> soon as I could because of that...
I can't stand the TV frequency either...although like most my CRT runs
at 67kHz and an LCD wouldn't help there.
> The suggested rate of the PAM is 100 - 1000Hz. I know the 1000Hz would
> be quite an annoying hum. Do you think the lower frequency would be a
> better starting point? Most people working in large building have
> 10-100Hz vibrations around them constantly. That doesn't seem to bug
> very many.
When I changed my EV from 5kHz to 20kHz I was much more pleased. You
may want to go the other direction since you're on the low end of
hearing. An Internet search reveals the following:
http://www.audioholics.com/techtips/roomacoustics/HumanHearingAmplitude.php
>From the graphs it can be determined that it takes 20dB more power to
just barely hear 100Hz compared with 1000Hz. It would probably be
better in your case to use 100Hz but that increases capacitor size and
may be worse for vibrations inside the vehicle.
Solving for a 10V capacitor drop seems a bit excessive. This voltage
drop should be less than the drop that occurs due to full current across
the battery. I find that my 12V, 110Ah batteries have about 0.01 ohms.
For 100Hz you would definitely need a very high capacitance to filter
the current. In comparison, my 20kHz EV design will use about 120uF for
150A. This is allowing 8V of ripple which is 2% of my bus. For this
same ripple you'd need 0.2F at 100A and 1.0F at 500A. At this point it
might be worth it to do a simulation - or to disregard the filtering
idea completely. Keep in mind that symmetric PWM decreases the
effective switching frequency by a factor of two (capacitor is twice as
large).
- Arthur
http://mywebpages.comcast.net/awmatt
--- End Message ---
--- Begin Message ---
Reclaimed title? You could argue it's not the same vehicle, the
original was just "parts".
Danny
[EMAIL PROTECTED] wrote:
According to the following page, only new EVs that "have never been used as a
non-electric vehicle" are eligible for the tax credit.
http://www.fueleconomy.gov/feg/tax_afv.shtml#ev
Does anybody have a defensible way of claiming the credit for a conversion?
Cheers,
Jay Donnaway
www.karmanneclectric.blogspot.com
--- End Message ---
--- Begin Message ---
Thanks Jim for your help, your right it doesn't matter which direction with
angled brushes according to Craig at Warfield engineering headquarters
1-800-435-9346. The local Warfield guy here (in Roanoke, VA) was wrong. Of
course time will tell that the machine shop is hooking it up to go CCW
instead of CW in the forward direction. You put my mind at ease when you
said that angled brush motors such as GE & Prestolite have replaced ADC
vertical brushes in different directional appiliacions. I think my cheese
wedge (commuta-city-car) went "backwards" pushing against the brushes
instead of pulling down in the pointed direction.
Thanks, Mark
----- Original Message -----
From: "Mark Hanson" <[EMAIL PROTECTED]>
To: "Jim Husted" <[EMAIL PROTECTED]>; <[email protected]>
Sent: Friday, January 06, 2006 10:16 AM
Subject: Re: Is it bad to spin backwards?
> The Warfield guy here who runs the motor shop said I would burn up brushes
> if I ran backwards (pushing up on the brushes). The brushes are angled
for
> a clockwise direction he told me and forward motion should be in the
> downward angled direction of the brushes he said. (This was also the guy
> who said the motor will "Splode" at 2X voltage). Anyway the machine shop
is
> having fits & caniptions going off the geared end but I don't want to
> constantly be replacing brushes or damage the commutator though. Do you
> know anone who consistently runs a motor in reverse and hasn't any
problems?
> Thanks, Mark
> ----- Original Message -----
> From: "Jim Husted" <[EMAIL PROTECTED]>
> To: <[email protected]>
> Sent: Thursday, January 05, 2006 9:52 PM
> Subject: Re: Is it bad to spin backwards?
>
>
> > Hey Mark
> >
> > Angled brushes are used by a veriety of manufatuerers. GE and
> Prestolite are the two most noted. They angle the attack to increase the
> footprint of the brush without having to pay for a larger brush. Now I
> don't know whether there is any differance with being over-volted, but if
it
> has 4 terminals it is designed to run in both directions. As far as the
> bevel do you mean the curve on the brush where it hits the comm? If so,
yes
> having as much of the brush touching the comm is what you want and is why
it
> has that curve. I believe you will have no issue running the motor the
> direction you require.
> > Hope this helps
> > Cya
> > Jim Husted
> > Hi-Torque Electric
> >
> > Mark Hanson <[EMAIL PROTECTED]> wrote:
> > Hi, it's me again. I'm getting my motor shaft machined to adapt with a
> slip yoke drive shaft to the differential for direct drive on a Cushman.
The
> shaft end they have to attach to will make the motor spin in reverse or
> pushing the brushes up. The other motor end has a beveled gear on it which
> the machine shop said they couldn't attach to. The desired direction I was
> told by Warfield was to go in the pointed downward direction of the
brushes.
> I thought my Cummuta-Car spun in reverse, but can't remember now and I
think
> Hondas do also.
> >
> > Are the brushes on a DC series wound motor beveled to increase surface
> area or for direction? Any motor guru's know or had experience with
> operating long term in reverse?
> > Is there an easy way to reverse the brush assembly if I need to?
> >
> > Thanks, Mark
> >
> >
> >
> >
> >
> > ---------------------------------
> > Yahoo! DSL Something to write home about. Just $16.99/mo. or less
> >
>
--- End Message ---
--- Begin Message ---
Cory Cross wrote:
Danny Miller wrote:
...
With just one battery, 1.2V x 7 off-units loses 8.4V, which is 17.5%
of the power from a 48V battery (or 70% of a 12V battery!). Maybe have
a contactor engaged over the diodes unless they may be used next?
Indeed, let's hope you don't have to drive around at 12V very long! If
you're going to do that, might as well use a DPDT contactor for each
controller, and loose the MOSFET and diode. Clickity-clack,
clickity-clack goes the car down the road. (Don't laugh too hard, the
software & network actually supports that)
Basically, any way you can think of to contribute voltage or bypass
yourself that can be controlled with a logic signal will work. Only
thing that would need to be adjusted is the timing. The Power Modules
are in two parts - Supervisor Board (micro with interface, sensor, and
isolation circuits, connects to data bus only), and the Power Board
(MOSFETs, bypass diode, connects to power bus only). A single 5V
isolated output line is running between them.
At top speed, the diodes won't be used at all, only the low-loss
MOSFETs, so it would be more efficient at this point than an IGBT
controller.
You have stumbled upon the true power of the Subnet Switch, young
padwan. Maximum efficiency at half speed. If you scale the setup
properly and pick the gears you drive in wisely, you can get over 80% of
your driving in those two "max eff bands". It may be the most expensive
component (at ~$100-$200), and it is totally optional, but I would
*highly* recommend using it.
--
Stefan T. Peters
--- End Message ---
--- Begin Message ---
So I got the Metro going ....
APPROX. - 43mph ....4th gear 210amps on the flat.....48 volts
Is this in the ball park ??
Only have a digital multi meter. 53.4 volts starting (is that a full
charge ?)
--- End Message ---
--- Begin Message ---
Dave [mailto:[EMAIL PROTECTED] wrote:
> Yeah I know it's probably anal but I'm a newbie and this is a
> brand new pack and I'm still getting familar with it and
> battery maintenance in general. The first dozen cycles I
> checked the SG after charge and after dis-charge just so I
> could see what was going on as quickly as possible.
I thought so, but didn't want to say [type] it. ;^>
> >(Stratification) This is true, however, if you take your sg readings
> >shortly after the batteries finish charging...
> >
> >
> But if you do that you are measuring a surface charge that will
> dissapate resulting in a higher than actual reading. ???
I haven't noticed any such effect in my testing, however, my
refractometer only resolves s.g. to 0.005, and I don't worry
particularly much about a difference of 0.005 or less between readings.
Perhaps a battery expert such as Nawaz can confirm/refute this for us,
however, my experience suggests that surface charge may be largely or
entirely a voltage phenomenon as I have not noticed any evidence of s.g.
varying significantly between readings taken an hour or two after the
end of charge or 12hrs or more after the end of charge. If you have
noticed such effects, it may be that what you are really noticing is the
effect of temperature on the uncompensated s.g. reading as the battery
(likely) cools over the several hours following the end of charge.
It is commonly recommended that voltage readings be taken many hours
after the end of charge when using them to estimate the state of charge,
or at the very least to take the readings some consistent amount of time
after the end of charge so that the readings have consistent meaning.
My own testing suggests that the voltage settles very nearly to its
final value within a few hours of the end of charge, so I don't worry
particularly much about this.
Cheers,
Roger.
--- End Message ---
--- Begin Message ---
Arthur W. Matteson wrote:
When I changed my EV from 5kHz to 20kHz I was much more pleased. You
may want to go the other direction since you're on the low end of
hearing. An Internet search reveals the following:
http://www.audioholics.com/techtips/roomacoustics/HumanHearingAmplitude.php
>From the graphs it can be determined that it takes 20dB more power to
just barely hear 100Hz compared with 1000Hz. It would probably be
better in your case to use 100Hz but that increases capacitor size and
may be worse for vibrations inside the vehicle.
Solving for a 10V capacitor drop seems a bit excessive. This voltage
drop should be less than the drop that occurs due to full current across
the battery. I find that my 12V, 110Ah batteries have about 0.01 ohms.
For 100Hz you would definitely need a very high capacitance to filter
the current. In comparison, my 20kHz EV design will use about 120uF for
150A. This is allowing 8V of ripple which is 2% of my bus. For this
same ripple you'd need 0.2F at 100A and 1.0F at 500A. At this point it
might be worth it to do a simulation - or to disregard the filtering
idea completely. Keep in mind that symmetric PWM decreases the
effective switching frequency by a factor of two (capacitor is twice as
large).
- Arthur
http://mywebpages.comcast.net/awmatt
This whole conversation has clinched the use of "power subnets" with a
supervised switch for me.
If you scale the pack to provide ~70MPH in your vehicle at 100% voltage,
then you also would have ~35MPH at 100% PAM (voltage) with half the
battery amps.
You can adjust these speeds up and down a bit with the proper use of
gears while driving. This way during most of your driving (either
30-40MPH on main through-fares, or 60-80MPH on the highway/freeway) you
would be drawing around the same battery current as a good PWM
controller. Any "stiffening" would only help in the low PAM periods
(0-5MPH and 40-45MPH in the above example). But I've noticed that I push
the throttle to half way when accelerating off a stoplight, then back
off a bit when I reach my cruise speed (raise PAM gradually to 100% -
48VDC within a sec, then hold until desired speed is reached).
Wouldn't the way a series DC motor works (no constant RPM to voltage
ratio) cooperate well with this typical use?
--
Stefan T. Peters
--- End Message ---
--- Begin Message ---
Hey Mark
Thanks for the follow-up. Nice to know I get some things right, hehehe. I
looked up about a dozen or so Prestolite pump motors and found there are
actually more that run against the brush angle than run with it. This is just
one more example of how misinformation is spread. Best of luck with your
prodject.
Cya
Jim Husted
Hi-Torque Electric
Mark wrote:
Thanks Jim for your help, your right it doesn't matter which direction with
angled brushes according to Craig at Warfield engineering headquarters
1-800-435-9346. The local Warfield guy here (in Roanoke, VA) was wrong. Of
course time will tell that the machine shop is hooking it up to go CCW
instead of CW in the forward direction. You put my mind at ease when you
said that angled brush motors such as GE & Prestolite have replaced ADC
vertical brushes in different directional appiliacions. I think my cheese
wedge (commuta-city-car) went "backwards" pushing against the brushes
instead of pulling down in the pointed direction.
Thanks, Mark
----- Original Message -----
From: "Mark Hanson"
To: "Jim Husted" ;
Sent: Friday, January 06, 2006 10:16 AM
Subject: Re: Is it bad to spin backwards?
> The Warfield guy here who runs the motor shop said I would burn up brushes
> if I ran backwards (pushing up on the brushes). The brushes are angled
for
> a clockwise direction he told me and forward motion should be in the
> downward angled direction of the brushes he said. (This was also the guy
> who said the motor will "Splode" at 2X voltage). Anyway the machine shop
is
> having fits & caniptions going off the geared end but I don't want to
> constantly be replacing brushes or damage the commutator though. Do you
> know anone who consistently runs a motor in reverse and hasn't any
problems?
> Thanks, Mark
> ----- Original Message -----
> From: "Jim Husted"
> To:
> Sent: Thursday, January 05, 2006 9:52 PM
> Subject: Re: Is it bad to spin backwards?
>
>
> > Hey Mark
> >
> > Angled brushes are used by a veriety of manufatuerers. GE and
> Prestolite are the two most noted. They angle the attack to increase the
> footprint of the brush without having to pay for a larger brush. Now I
> don't know whether there is any differance with being over-volted, but if
it
> has 4 terminals it is designed to run in both directions. As far as the
> bevel do you mean the curve on the brush where it hits the comm? If so,
yes
> having as much of the brush touching the comm is what you want and is why
it
> has that curve. I believe you will have no issue running the motor the
> direction you require.
> > Hope this helps
> > Cya
> > Jim Husted
> > Hi-Torque Electric
> >
> > Mark Hanson wrote:
> > Hi, it's me again. I'm getting my motor shaft machined to adapt with a
> slip yoke drive shaft to the differential for direct drive on a Cushman.
The
> shaft end they have to attach to will make the motor spin in reverse or
> pushing the brushes up. The other motor end has a beveled gear on it which
> the machine shop said they couldn't attach to. The desired direction I was
> told by Warfield was to go in the pointed downward direction of the
brushes.
> I thought my Cummuta-Car spun in reverse, but can't remember now and I
think
> Hondas do also.
> >
> > Are the brushes on a DC series wound motor beveled to increase surface
> area or for direction? Any motor guru's know or had experience with
> operating long term in reverse?
> > Is there an easy way to reverse the brush assembly if I need to?
> >
> > Thanks, Mark
> >
> >
> >
> >
> >
> > ---------------------------------
> > Yahoo! DSL Something to write home about. Just $16.99/mo. or less
> >
>
---------------------------------
Yahoo! DSL Something to write home about. Just $16.99/mo. or less
--- End Message ---
--- Begin Message ---
Stefan,
Main reason for the caps are to take a lot of the impedance
of the batteries and the wiring out of the motor-feeding loop.
This can impact battery life, but it is not the primary reason.
For example with a Zilla 2k (pretty extreme, but just to drive
the point home) you can generate 2000A motor current.
If you pulse your batteries at 2000A then they will likely have
a problem, but also your efficiency is not very good:
Say you have 25x 12V batteries for 300V, each battery has an
internal resistance of 4mOhm for a total of 100mOhm.
Even ignoring the wiring resistance (which should be very low
if done well) the batteries will sag by 2000A x 100mOhm = 200V
so the input to the controller will be only 100V when there is
no capacitor to deliver the motor current.
Draw from the batteries is an instantaneous 2000A x 300V =
600kW, of which 400kW is reduced to heat inside the batteries
and 200kW is delivered to the controller. (67% loss)
Now consider the math with a very large capacitor:
Say you are running 25% duty cycle in your controller, so an
ideal controller would deliver 2000A when getting an average of
500A from the batteries (motor voltage will be 1/4 of the
battery voltage)
now the same batteries will sag 500A x 100mOhm = 50V so the
controller gets 250V in and delivers an average 60V to the motor.
Doing the power math: 500A x 300V = 150kW draw from the batteries
continuously while the capacitor delivers 2000-500 = 1500A during
the periods that the motor gets power from the controller.
The heat dissipated in the batteries: 500A x 50V = 25kW (17% loss)
The peak power out of the capacitor: 250V x 1500A = 375kW
Instantaneous power into the motor: 2000A x 250V = 500kW, but only
25% duty cycle, so the motor receives 0.25 x 500kW continuous,
which is exactly the 125kW that the batteries are supplying to the
(ideal) controller.
In practice the controller and capacitor will have some losses,
but usually much lower than battery losses, so it makes sense to
"get the batteries out of the loop" and average the battery current
by using input capacitors in the controller.
BTW: also in the first example (without cap) can the motor receive
an average 125kW, but the low input voltage of 100V will cause the
duty cycle to rise to 62.5% (125kW/200kW instantaneous) which
means that the batteries will need to supply the 2000A all this
time, for an average of 0.625 x 2000A = 1250A instead of 500A.
This means that the average power draw from the battery is now
300V x 1250A = 375kW instead of 150kW as in the second example.
You see how the reduction in efficiency eats into your range and
I have not even considered Peukert or murdering your batteries
at the kind of abuse of pulsing them with 2000A.
Just the change in battery power will cut your range to 40%
(375kW/150kW = 2.5 times higher power, so the available energy
in the battery lasts 1/2.5 = 0.4 x the original time)
Hope this clarifies why all controllers have input capacitors,
which is the reason you need a "slow-start" or pre-charge, as
the capacitor voltage will be 0 when you start your car and if
you would simply close your main contactor between batteries
and controller, it will see a short pulse of 300V / 100mOhm =
3000A. That will hammer your batteries, contactor and capacitor
to an early dead. The pre-charge contactor will connect a
series resistor between battery and capacitor, typically 50 Ohm
or so. This causes a current of max 6A to flow for a few seconds
and since the time constant of this circuit is low (25ms for
50 Ohm & 500 uF) the capacitor will be completely charged after
a few seconds and the main contactor will close between
a 300V battery and an (almost) 300V capacitor.
This prevents burn marks on the main contactor, which would
increase the resistance in the main drive current flow.
NOTE that if your input capacitor increases to 0.5F then either
you need to permanently attach this capacitor to your batteries
and switch your contactor between this set and the controller,
or accept that a 50 Ohm resistor causes a time constant of 25s
which means that you will need to wait more than a minute
before your main contactor should be closed.
You can speed this up with lower resistance, but take into
account that a 5 Ohm resistor will generate an instantaneous
60A x 300V = 18kW of heat, which will destroy all but the
most rugged (and bulky) power resistors, likely you will need
at least 10 parallel resistors of 50 Ohm 100Watt, which are
the type of resistors used today for this job.
I think Victor and a few others have a parallel string of
capacitors across their battery, because most Li chemistry
tends to have high internal resistance (causing sag) and
not able to take high peak currents very well (fast takeoff)
Regards,
Cor van de Water
Systems Architect
Email: [EMAIL PROTECTED] Private: http://www.cvandewater.com
Skype: cor_van_de_water IM: [EMAIL PROTECTED]
Tel: +1 408 542 5225 VoIP: +31 20 3987567 FWD# 25925
Fax: +1 408 731 3675 eFAX: +31-87-784-1130
Proxim Wireless Networks eFAX: +1-501-641-8576
Take your network further http://www.proxim.com
-----Original Message-----
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]
Behalf Of Stefan T. Peters
Sent: Friday, January 06, 2006 10:53 AM
To: [email protected]
Subject: Re: I want to build a PWM DC motor controller
Cor van de Water wrote:
> Formula to determine Delta V on a capacitor: I x t = V x C
> (Current times Time equals Voltage times Capacitance)
> An example:
> I = 100A
> t = 50 us (0.00005 s = 5 x 10^-5)
> Now if we want to keep the voltage-swing on the capacitor under 10V
> how large a capacitor do we need? (Ignore series resistance for now)
> Enter the known values in the formula we get:
> 100 x 5 x 10^-5 = 10 x C
> 5 x 10^-3 = 10 x C
> 5 x 10^-4 = C
> C = 500 uF.
>
That makes 500A at 1 hundredth of a second = 0.5F. I can get a hold of
those fine, then my question is back to the original one: How much would
it help to decrease the spurious load on the bats, and hopefully
increase the life?
> If you go to the low frequencies you suggest, the capacitors become
> extremely bulky to supply so much current for that "long" and also
> you will have very annoying hums and whistles in audible frequencies.
> That is the reason Otmar chose the TV line flyback frequency: it is
> so often around us that most people have a hearing "notch" at this
> frequency, in addition to it being almost above the audible band
> of frequencies. No whines and whistles.
>
Is it just the motor that is "singing"? BTW, that TV flyback frequency
has always bugged the bejeezus out of me. That supersonic whistling that
TVs make whenever they are on, right? I bought an LCD monitor & TV as
soon as I could because of that...
The suggested rate of the PAM is 100 - 1000Hz. I know the 1000Hz would
be quite an annoying hum. Do you think the lower frequency would be a
better starting point? Most people working in large building have
10-100Hz vibrations around them constantly. That doesn't seem to bug
very many.
--
Stefan T. Peters
--- End Message ---
--- Begin Message ---
Go right ahead and argue that. Everyone ought to go through a tax
audit once in their lives.....
Is it just me or is this attitude getting out of control, that any
sort of cheating, any means to accomplish the ends is OK.
Sad commentary on the state of things.
John
On Fri, 06 Jan 2006 13:41:29 -0600, Danny Miller
<[EMAIL PROTECTED]> wrote:
>Reclaimed title? You could argue it's not the same vehicle, the
>original was just "parts".
>
>Danny
>
>[EMAIL PROTECTED] wrote:
>
>>According to the following page, only new EVs that "have never been used as a
>>non-electric vehicle" are eligible for the tax credit.
>>http://www.fueleconomy.gov/feg/tax_afv.shtml#ev
>>Does anybody have a defensible way of claiming the credit for a conversion?
>>
>>Cheers,
>>Jay Donnaway
>>www.karmanneclectric.blogspot.com
>>
>>
>>
>>
>
---
John De Armond
See my website for my current email address
http://www.johngsbbq.com
Cleveland, Occupied TN
A foolish consistency is the hobgoblin of little minds.-Ralph Waldo Emerson
--- End Message ---
--- Begin Message ---
Bob Bath [mailto:[EMAIL PROTECTED] wrote:
> Here are the options I'm going with right now:
>
> 1) I'm realizing now that lower current is better;
> less resistance, so the charger will let them take the
> voltage higher. Will attempt to charge the batts. at
> 4A for most of the day.
Huh?
The resistance does not change, but at lower current the voltage drop
across the interconnects and internal resistance will be less. If you
are running the charger in constant voltage mode, this means that as the
current drops the voltage at the batteries *increases* to be more nearly
equal to the voltage that the charger is set to.
If you are running in constant current mode, which your message
suggests, then the voltage the batteries will get to depends only on the
batteries themselves, the voltage limit you set the charger to, and the
current limit you set the charger to. The higher the current limit, the
higher the peak voltage that the batteries will reach: period. When the
voltage stops rising at a given charge current, one of two things has
happened: either you have hit the charger's voltage limit (crank it up
further; if the voltage rises, then that was the problem), or the
batteries are fully charged.
Your charger should be fully capable of taking those batteries to just
about any voltage you want at any current you want, subject mostly to
the limitations imposed by the outlet you have plugged into. This is
the real strength of the PFCxx charger.
3-4A is about what a relatively healthy US8VGC will taper to when held
at the proper absorption voltage; it is too low to do a proper equalise,
IMHO. I believe Nawaz recommends about 2x this current for equalising.
> 2) Will add a battery additive (per suggestion) from
> JC Whitney, as I _must_ get the sulfate crystals off
> of 2, 3, 6, and 8.
Good luck, but I fear you are about to trash otherwise relatively decent
batteries. A few cells with s.g. around 1260-1265 is not, IMHO,
justification for poisoning them with snake oil. Run a capacity test on
them and see how good or bad they really are.
Why do you believe these batteries to be sulphated?
A typical symptom of sulphated batteries is that they exhibit *higher*
internal resistance than healthy batteries, and therefore show unusually
low voltages when loaded and unusually high voltages when being charged.
The batteries you have singled out have some of the lowest on charge
voltages of all your batteries, which contradicts your sulphation
hypothesis.
> 3) Will purchase a 12V charger and charge these guys
> independently of the rest of the pack. Will have to monitor
> carefully, as there are no separate chargers for 8V
> batteries. I could drop the voltage down with a couple of
> diodes, though.
You don't need an 8V charger; just get a 12V charger or power supply
that can deliver about 10A max and that will start into an 8V battery.
Hook it up to the suspect battery and leave it charge until hourly
voltage readings indicate that the voltage has quit rising (2-3
consecutive readings are near identical).
In fact, your PFCxx ought to be able to do this task, shouldn't it?
> Thanks as always, and wish me luck...
Good luck.
Please do *not* apply any additive to your batteries yet. If you need
professional battery advice, contact Nawaz Qureshi
<[EMAIL PROTECTED]> as he is the authority on your batteries.
If you charge the individual batteries until their voltage has stopped
rising for at least an hour, then they are fully charged; any cells
which still show low s.g. simply will not come up any higher. For
whatever, reason (unequal electrolyte levels, sulphation, age, etc.),
this is now the fully charged s.g. for those cells.
Don't lose sleep over it unless you do a capacity test on those
batteries and find that their capacity is enough lower than the rest of
the pack to make the car unusable. In this case, go ahead and try your
snake oil, but try it only in one battery until you have verified that
it either results in an improvement or is benign; this way if it worsens
the battery (e.g. s.g. goes up but capacity drops) you only have one
battery to replace instead of several before the EV is usable again.
Remember, s.g. indicates the state of charge of the cell/battery, but
does *not* indicate its capacity; you may find snake oil that can prop
up the s.g readings but has no effect on usable capacity, or indeed even
worsens the capacity.
One of the best things you can do for your US8VGCs is to hit them with
the highest initial charge current your outlet will allow. It may be
too late to realise any dramatic improvements from your present pack,
however, regularly hitting US Battery product with the recommended C/10
(or better) initial charge current will result in the best life and
performance from them.
Cheers,
Roger.
--- End Message ---
--- Begin Message ---
Neon John wrote:
Go right ahead and argue that. Everyone ought to go through a tax
audit once in their lives.....
Can be a very very bad thing, trust me.
Is it just me or is this attitude getting out of control, that any
sort of cheating, any means to accomplish the ends is OK.
Sad commentary on the state of things.
I think maybe it's just the same 'ol "sticking it to the man"
motivation, shrouded in eco-concern... depends on your current opinion
of our government whether or not that's a bad thing.
What I want to know is when I can apply for some of that federal grant
money for EV conversions - they pay for (most of) the parts, you provide
the labor ;)
--
Stefan T. Peters
--- End Message ---
--- Begin Message ---
From the page:
What is a qualifying electric vehicle?
To be eligible for the tax credit, a vehicle must meet the following
requirements
* It must be powered primarily by an electric motor drawing current
from rechargeable batteries, fuel cells, or other portable sources
of electrical current.
* It must have at least four wheels and be manufactured primarily
for use on public streets, roads, and highways (vehicles
exclusively used on rail or rails do not qualify).
* It must meet all federal and state emissions requirements.
* It has never been used as a nonelectric vehicle.
* You must purchase the vehicle new and for your own use, not for
resale
* You must drive it mostly in the United States.
* Government agencies, tax exempt organizations, and foreign
entities are not eligible.
Must be "purchased new." Translation: no home-built or conversion.
Also, no from-scratch motorcycles (in theory easier than a from-scratch
car), as they don't have four wheels.
[EMAIL PROTECTED] wrote:
According to the following page, only new EVs that "have never been used as a
non-electric vehicle" are eligible for the tax credit.
http://www.fueleconomy.gov/feg/tax_afv.shtml#ev
Does anybody have a defensible way of claiming the credit for a conversion?
Cheers,
Jay Donnaway
www.karmanneclectric.blogspot.com
--- End Message ---
--- Begin Message ---
Cor van de Water wrote:
NOTE that if your input capacitor increases to 0.5F then either
you need to permanently attach this capacitor to your batteries
and switch your contactor between this set and the controller,
Thanks for all the good info!
Given that there is one simple "controller" per battery, and they are
manipulated in concert to provide PAM to the motor, the cap will always
remain connected to the battery. Hence my use of the term "stiffening
capacitor" - it effectively creates a stiffer battery, like Victor and
the others are doing. You would need to charge it before installing the
battery, or risk melting the contacts on the cap.
Now given that this setup will be almost full on during typical driving
(on for 0.009 secs, off for 0.001 - maybe just leave 'em all on), the
caps would just be for low PAM modes (0-10MPH). The PAM frequency is
100Hz, the switching speed of the transistors is 1KHz. During low PAM
periods, a particular battery would be switched on part of the time
(equivalent to the PAM voltage/pack voltage, with two pack
configurations of X and X/2 volts).
I have noticed that even though PAM predates PWM, it can be a bit more
complicated. There are three fundamental variables (amplitude, pulse
width & frequency) instead of just the latter two. Now if you set pulse
width to 100% and use a fixed frequency, it becomes much simpler. I'm
thinking of support both modes in software, actually.
--
Stefan T. Peters
--- End Message ---
