EV Digest 6053
Topics covered in this issue include:
1) Re: GoWheel.com & EV-Battery.com
by "Charles Whalen" <[EMAIL PROTECTED]>
2) Re: Who did kill the electric car?
by "David Roden" <[EMAIL PROTECTED]>
3) Re: EV controllers? the 5th option...
by mike golub <[EMAIL PROTECTED]>
4) RE: EV controllers? the 5th option...
by Cor van de Water <[EMAIL PROTECTED]>
5) Re: Who did kill the electric car?
by "Lawrence Rhodes" <[EMAIL PROTECTED]>
6) RE: Who did kill the electric car?
by Cor van de Water <[EMAIL PROTECTED]>
7) Re: Current multiplication, was: EV controllers? Practical parame
ters
by Cory Cross <[EMAIL PROTECTED]>
8) Formula
by "Don" <[EMAIL PROTECTED]>
9) Bradley GTE Project
by "Don" <[EMAIL PROTECTED]>
10) Re: Testing 123
by Jude Anthony <[EMAIL PROTECTED]>
11) Re: Formula
by "Roland Wiench" <[EMAIL PROTECTED]>
12) Re: Formula
by "Don" <[EMAIL PROTECTED]>
--- Begin Message ---
Thanks for this explanation, Jukka. In light of this, I wonder why we are
torturing ourselves (at least it would be torture for me, although probably
more like a fun and exciting challenge for people like Jukka and Jay) to try
to parallel all of these little 2.3Ah cells when A123 is supposedly coming
out with large-format (100-200Ah) cells in the first quarter of '07,
according to a friend of mine in the EV business who has been speaking with
the company.
Wouldn't it be a quantum order of magnitude easier to do a much simpler BMS
for 100 3.3V/150Ah cells connected together in a series string of single
cells than to try to deal with all the complexity and challenges of
designing and engineering a BMS for 100 3.3V blocks connected in series
where each block is a single-layer 8x8 tray of 64 3.3V/2.3Ah cells all
tab-welded together in parallel (with all of the attendant problems that
Jukka describes below)?
And I'm not quite sure how one is going to do those regular Sunday afternoon
cell swaps that Jukka was talking about when the darn things are all
tab-welded together like that.
Can you really match cell impedances and capacities close enough when
assembling each 8x8 tray of 64 3.3V/2.3Ah cells such that they will stay
entirely in balance (or at least close enough not to cause any problems of
the type Jukka describes below) over the claimed 15-year calendar life of
these cells without having BMS control circuitry to the cell level (but
rather merely to the block level)?
Alternatively, is there any practical way to actually design engineer the
BMS all the way down to the individual 3.3V/2.3Ah cell level within each
block of 64 cells in parallel? And how exactly would you do that with each
cell tab-welded to its adjacent cell, all packed tightly together, with each
cell touching its adjacent cell?
Here is how Hybrids Plus in Boulder, Colorado (whose battery consultant is
Bill Dube, I believe) does it with their assembled pack of A123 cells for
their Prius PHEV conversion kit:
http://hybrids-plus.com/pmwiki/uploads/Ext/bmsinstall2.jpg
But it looks from the picture like they've only got one BMS regulator module
for each group of 10 paralleled cells in blocks of what appear to be 20
cells in parallel. They certainly haven't taken the BMS all the way down to
the individual cell level. One might be able to do so if one were to use
those longer pre-welded tabs that come with A123's engineering sample kits
(that A123 sells on its website), which you either bolt or rivet together to
connect adjacent cells, because that leaves enough space between cells that
you could fit tiny PCBs (of the type shown in the linked photo above) in
there for each individual cell. But then of course your volumetric energy
density for the pack gets shot and goes to hell because of all that empty
space in there (which may be better for cooling effect, but I believe Jay
said that these cells don't need it).
So my guess would be that there probably isn't any practical way to
physically/topologically engineer the BMS all the way down to the individual
cell level in blocks of paralleled cells when the cells are packed tightly
together touching each other to give the best possible volumetric energy
density for the assembled pack. (And remember that the A123 cells are
already less energy dense than other li-ions to begin with due to their
LiFePO4 chemistry, which sacrifices a bit of energy density for much greater
safety and stability, so you really don't want to have to open up a bunch of
additional space in there, as then you might as well just use much cheaper
NiCd or NiMH in that case.)
... All of which brings me back to the question I started with: Why are we
making all of this extraordinary effort, jumping through hoops, with what
would appear to me to be over-engineering a solution to a problem that will
be rendered moot and easily solved by A123 itself when it comes out with
large-format cells next quarter? Wouldn't the path of least resistance,
both literally and figuratively, be to just wait until next quarter for the
large-format cells?
Thanks for any thoughts/insight on this.
Charles Whalen
On Friday, October 20, 2006 10:28 AM, Jukka Jarvinen wrote:
In the tests I've done I found that parallering the cells you actually
force the cells to same voltage. At least it's trying it's best with
rotational currents (dunno the real term in English) to get them in
"balance".
But as it's known Lithium cells rarely can have SOC determination with
voltage. cells will maintain the voltage based balance and the weakest
(less capacity one but not necessary the highest impedance) paralled cell
will fail first. It may even short. With welded poles it's married to 7
others. Worst case is that 7 times best current at the present voltage
level will be transformed to heat in the shorted cell. It may not short
completely but it will draw paralled cells to near 0 V.
Now... If that would have been Cobolt cell.. thermal runaway @ 180
Celsius, Manganese usually over 200 Celsius and Ferrites well over 300
Celsius. I think A123 and Valence has about the same limit. Can't say.
They have not been so interested to sell me any cells.. Wonder why...
So question is that can the paralled cells contribute enough power to heat
shorted cell enough. If your answer is yes.. problems ahead. If No... You
should be safe but frequent cell swapping should be scheduled for sunday
afternoons.
The single cell solution up to 10 000 ah from TS has same problem. That is
if you paraller the cells. But as single cell inside one case with same
electrolyte compartment.. It's another story. Such problem does not exist.
Careful management of cells even in this 8 paralled cell design is best
way to go. you can get away with most problems by sorting the cells by
impedance and capacity.. Whatta F#€ .. AM I helping possible rivals here..
damn I must be stupid :)
-Jukka
p.s- cause first!
[EMAIL PROTECTED] kirjoitti:
Our feelings are that if a single cell goes bad, we
will refurb the whole pack, rotating remainder cells
of any block into our own toys, testing, destruction,
etc. If too many occurances of failures, we revise
the BMS to be more comprehensive. We're already
planning for more BMS features. It's only a matter of
time, design, money, real world results, and history. We will move
forward, hopefully staying ahead of the
needs and wants of our customers. We have a new plan
for easier cell replacement too that we are testing. We'll see.
Jay
--- Charles Whalen <[EMAIL PROTECTED]> wrote:
Jukka,
Could you please elaborate on this and explain why
you feel this is the
case?
Is it -- (if I understand Jay's description
correctly) -- because their BMS
apparently only goes to the block level (each block
being 8 cells in
parallel) but not to the individual cell level
within blocks, such that
there is apparently no way to bypass charging
current around individual
cells (that have already reached their top-of-charge
voltage before others
in the block) but rather only around an entire
8-cell-in-parallel block en
masse, and thus would seem to be making an
assumption and leap of faith that
all 8 cells in each block will track absolutely
dead-even with no variation
in voltage over the supposedly 15-year calendar life
claimed for these
cells?
I too was wondering about this, so I'm glad you
raised the question.
Another question in my mind is due to the fact (as I
understand it from
Jay's description) that the 8 cells in parallel in
each block are all
(high-current) tab-welded together, it would appear
that when the weakest
cell in the block utimately degrades to the point of
needing replacement,
(which, again, would seem to be made more likely at
some point, or hastened,
by the lack of BMS control circuitry to the
individual cell level), it will
therefore require replacing, at minimum, the entire
8-cell block (since
they're all tab-welded together) rather than merely
the one single errant
cell.
Any thoughts on that from either Jukka, Jay, or Bill
Dube?
Thanks,
Charles Whalen
On Thursday, October 19, 2006 3:39 PM, Jukka
Jarvinen wrote:
Sounds quite solid for me. you can get safe pack
by this way. But I'm
afraid after certain point and much before
intented a cell replacements
will be reality and frequent. Straining the
"weakest link" in every cycle
will bring it down and fast.
-Jukka
[EMAIL PROTECTED] kirjoitti:
Mark and all,
EV-Battery.com responding per question below:
--- Mark Fowler <[EMAIL PROTECTED]>
wrote:
Can you give us some more details about these
packs?
We will post results and photos at
www.EV-Battery.com
as we receive and analyze results, probably
updating
the websites about every five days.
What does the BMS do?
(From simple V monitoring to fully temp
compensated
V and I monitoring with over-charge regulation
and over-discharge
protection...)
Our first batteries are the 12 volt gasser
batteries
that require different BMS from stacked EV packs.
The
bigger switch-out packs (79.2V) will have BMS
monitors
that monitor voltage on each stack of eight
paralleled
cells. If this stack voltage gets too high, then
it
places a 3.3 ohm resistor across the stack,
discharging the excess energy.
Can the BMS communicate with an external (to the
batt) device?
(Like a charger, fuel gauge, or logging
computer)
Three different sets of alarms are combined into
�slow-down� or �stop� alarm signals.
First, if
the
voltage of a stack climbs dangerously high, an
alarm
is indicated which should be used to throttle the
charger down to about one amp. A second alarm is
sent
if the stack voltage falls dangerously low,
indicating
the stack is out of charge and the load should be
reduced or stopped. A third alarm source is the
three
temperature sensing thermistors in each module.
Thresholds can be set for
each of the BMS module�s
three thermistors. If any temperatures get above
their individual threshold, the alarm will again
be
signaled. All these alarms are indicated by
turning
on an opto-isolator on the BMS module. Because
they're optoisolated, all the BMS modules in a
pack
can
have their alarm signals wire-ORed in parallel to
create one pack alarm signal.
Do you have photos of a completed pack?
(Lots of 3D mock-ups in your photo gallery)
There are already some photos of our first gasser
(12
volt motorcycle batteries and 12 volt car batts)
already there. Keep in mind that these are
intended
to be gasser replacement batteries targeted to
the
custom car and motorcycle show vehicles. Users
are
trying to stay away from acid damage to their
expensive paint jobs and vehicles that sit for
months
(sometimes years) without driving. Many vehicles
are
primarily artistic and concept vehicles. Many of
them
would like to replace a 45 pound bulky battery
with a
1 pound motorcycle battery or a 5 pound auto
battery
that can hide safely anywhere in any position.
The
motorcycle builders are talking about putting
batteries inside their frames like a flashlight.
The
car guys are talking about a recessed floorboard
mount
with water cooler style lid. They are interested
because there is no bulging, venting, toxic
fumes,
explosive, or fire dangers.
We also realize that gasser batteries are not the
focus of the EV List. So now, we have
commitments to
supply a few race cars (including Le Mans), and
other
traditional gassers with these first batteries.
After
gasser success, we will sell them to people that
use
them in electric vehicles, BUT we are going to be
very
careful that we only supply educated builders
with
logical applications within reasonable
specifications.
We are not ready to promote mainstream EVers to
use
our batteries (as series or parallel packs) until
we
are sure of performance and safety. We will not
make
the switch-out battery pack & case available to
the
general public until it is ready. We only have
so
much manufacturing ability and will restrain
ourselves
from claims of a universal solution.
How are the individual cells interconnected?
=== message truncated ===
--- End Message ---
--- Begin Message ---
On 22 Oct 2006 at 0:35, Christopher Zach wrote:
> Before they made the EV1, they made the Dolphin series of cars.
That's not the way I understand it.
When I investigated purchasing one of the USE Prisms many years ago
(probably around 1997 or 1998), it was my understanding that GM did not
build it. The Prisms and S-10s were converted by US Electricar, which is
(was) not in any way related to GM. (Confusingly, it was also not related
to US Electricar of Athol, MA, converter of the Lectric Leopard and a couple
other items in the early 1980s.)
As I understand it, USE bought Prisms from GM, gutted them, bolted in
Dolphin drives, and hung a battery tray underneath. It was a pretty clean
conversion, but it was not done by GM at the factory.
Or perhaps I'm reading what you wrote literally, and that's not what you
really meant?
David Roden - Akron, Ohio, USA
EV List Assistant Administrator
= = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
Want to unsubscribe, stop the EV list mail while you're on vacation,
or switch to digest mode? See how: http://www.evdl.org/help/
= = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
Note: mail sent to "evpost" or "etpost" addresses will not reach me.
To send a private message, please obtain my email address from
the webpage http://www.evdl.org/help/ .
= = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
--- End Message ---
--- Begin Message ---
does anyone know about this guys controller?
http://www.belktronix.com/dcmotorctrl.html
he wants $2k for it and a PFC charger +BMS...
I just like to get some feedback?
--- John Wayland <[EMAIL PROTECTED]> wrote:
> Hello to All,
>
> Roger Stockton wrote:
>
> >
> >What DC'ers need is a Curtis-killer, not something
> that tries to compete
> >with Otmar's already excellent offerings.
> >
> >
>
> Stop it Roger, that makes too much sense!
> Considering how lame the
> Curtis is, a 'Curtis Killer' isn't much of a
> challenge.
>
> >We need an updated Raptor 600. Forget the
> series-parallel shifting and
> >even the electric reverse; if someone is going
> direct drive or with dual
> >motors, they're probably going to want the extra
> oomph of at least a Z1K
> >anyway. Forget the 1000+ amps; if someone wants
> 1000A, they'll buy a
> >Z1K; if they want more, a Z2K.
> >
> >
>
> Stop it Roger, that makes too much sense!
>
> >96V to perhaps 192V is going to cover most anyone
> who
> >would have bought this controller instead of a
> Curtis or being pushed
> >into a Zilla due to choosing a pack voltage over
> 144V.
> >
> >
>
> Stop it Roger, that makes too much sense!
>
> >My essential spec/feature list for a Curtis-Killer
> is:
> >
> >- 600 adjustable battery amps
> >
> >- handles at least the range of 96-144V; 156V is
> almost essential, and
> >up to ~200V would be even better
> >
> >- precharge/contactor control
> >
> >- basic fault detection (open/short throttle, high
> pedal lockout)
> >
> >- rev limiter (and a tach drive output sure would
> be nice)
> >
> >- integrated heatsink (water-cooled is preferable
> >
> >- priced on par with 1231C + appropriate heatsink.
> >
> >- contactor failure detection
> >
> >- controller failure detection & auto contactor
> dropout
> >
> >
> >
> Roger, how on earth does one come up with such a
> wonderfully thought-out
> wish list, and leave out
> 'silent operation-no Curtis squeal!'???? Or, did you
> assume that no one
> in their right mind would ever again, design a
> road-going EV controller
> to make such an awful, irritating noise?
>
> From Mike Sandman:
>
> >I just ordered a kit from KTA using a 1231C-7701
> for my 120v system.
>
> I'm curious Mike, did they tell you up front that
> this controller will
> cause your EV's drive motor to make an irritating
> non-varying squeal
> when you take off or cruise with a light throttle?
> Did they tell you
> that this controller has no safety features? Did
> they tell you it has no
> automatic precharge, and that if you don't design
> your own, it could
> blow up when the line contactor engages and send
> your EV off at full
> throttle? Did they tell you it cannot be adjusted
> for a battery input
> current limit to protect your battery pack
> investment? My guess is no to
> all the above.
>
> If it's not too late, you could save a bundle by
> purchasing a used
> Curtis 1221B controller. It's a bit less powerful at
> 400 amps, but it's
> silent and you could get one for maybe $300 if
> anyone has one available.
> The money saved would allow you to build a precharge
> circuit and pocket
> at least $500 cash, too. Better still, a used Raptor
> 600 could be found
> for maybe $800, is silent, properly cooled, has dip
> switch programming,
> and 900 amps motor loop current.
>
> I understand having to watch one's EV budget, but
> with one as high as
> $15k the last area I would have cut costs on would
> be on the all
> important motor controller. You may have saved
> $500-$600 by stepping
> w-a-ay down to the Curtis, but you gave up a lot of
> features that would
> have made your conversion more fun, far more
> pleasant, much smoother,
> more versatile, twice as powerful, and far safer,
> and...you gained a lot
> of negatives as described above.
>
> From Steve Powers:
>
> >When you get to more than 600 batt amps, you are
> really killing the pack.
>
>
>
> Is this one of those comments again, where one has
> to have read
> everything that came before and know that you might
> be talking about wet
> cell type batteries (and just didn't specify this),
> or do you really
> think that currents this low will harm powerful AGMs
> like Optimas,
> Hawkers, and Orbitals? Those of us using AGMs of
> higher quality know
> that the opposite is true, in that the high
> performance type AGMs
> actually give their best service life routinely
> dishing out BIG
> currents. 600 amps from the better AGMs is nothing.
> The Optimas in Blue
> Meanie lasted 6.5 years, and pretty close to each
> time it was taken out
> for a run, 1000 amps were sucked from them multiple
> times. Sure, under
> all out racing where you might suck 1000 amps from
> small 24.5 lb.
> batteries for 12 continuous seconds, it is stressful
> and sometimes fatal
> for the batteries, but for normal fun street driving
> 1000 amps for 5-6
> seconds from larger 40+ lb. AGMs is kid's
> stuff....besides, you're
> already well above the speed limit in that amount of
> time and will be
> off throttle and back to sub 100 amp levels.
>
> One more item to be added to Roger's list...compact
> size. Oat's Z1K's
> small near cube-shaped super compact size is not
> only amazing
> considering the thing cranks out 1000 full amps,
> it's also a blessing in
> an already crowded under-hood compartment. Though my
> experiences with
> the DCP Raptor 1200 were nothing but positive, its
> large bulky size was
> not a plus.
>
> See Ya......John Wayland
>
>
>
>
>
>
>
>
__________________________________________________
Do You Yahoo!?
Tired of spam? Yahoo! Mail has the best spam protection around
http://mail.yahoo.com
--- End Message ---
--- Begin Message ---
Looks like the results of a well executed homebrew project,
though I would have problems with supplying 24V as he
suggests - from a tap in the pack. That guarantees to have
your batteries unbalanced after a while.
It sounds like he did not want to add a DC/DC converter
in his controller, so he stuck it to the person buying his
controller.
Just my opinion after looking at it for a few seconds,
others please jump in if you have any real experience
with thic controller.
Did you ask him how many he already sold and how busy
he is in repairing any?
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: +1-610-423-5743
Take your network further http://www.proxim.com
-----Original Message-----
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]
Behalf Of mike golub
Sent: Sunday, October 22, 2006 12:00 AM
To: [email protected]
Subject: Re: EV controllers? the 5th option...
does anyone know about this guys controller?
http://www.belktronix.com/dcmotorctrl.html
he wants $2k for it and a PFC charger +BMS...
I just like to get some feedback?
--- John Wayland <[EMAIL PROTECTED]> wrote:
> Hello to All,
>
> Roger Stockton wrote:
>
> >
> >What DC'ers need is a Curtis-killer, not something
> that tries to compete
> >with Otmar's already excellent offerings.
> >
> >
>
> Stop it Roger, that makes too much sense!
> Considering how lame the
> Curtis is, a 'Curtis Killer' isn't much of a
> challenge.
>
> >We need an updated Raptor 600. Forget the
> series-parallel shifting and
> >even the electric reverse; if someone is going
> direct drive or with dual
> >motors, they're probably going to want the extra
> oomph of at least a Z1K
> >anyway. Forget the 1000+ amps; if someone wants
> 1000A, they'll buy a
> >Z1K; if they want more, a Z2K.
> >
> >
>
> Stop it Roger, that makes too much sense!
>
> >96V to perhaps 192V is going to cover most anyone
> who
> >would have bought this controller instead of a
> Curtis or being pushed
> >into a Zilla due to choosing a pack voltage over
> 144V.
> >
> >
>
> Stop it Roger, that makes too much sense!
>
> >My essential spec/feature list for a Curtis-Killer
> is:
> >
> >- 600 adjustable battery amps
> >
> >- handles at least the range of 96-144V; 156V is
> almost essential, and
> >up to ~200V would be even better
> >
> >- precharge/contactor control
> >
> >- basic fault detection (open/short throttle, high
> pedal lockout)
> >
> >- rev limiter (and a tach drive output sure would
> be nice)
> >
> >- integrated heatsink (water-cooled is preferable
> >
> >- priced on par with 1231C + appropriate heatsink.
> >
> >- contactor failure detection
> >
> >- controller failure detection & auto contactor
> dropout
> >
> >
> >
> Roger, how on earth does one come up with such a
> wonderfully thought-out
> wish list, and leave out
> 'silent operation-no Curtis squeal!'???? Or, did you
> assume that no one
> in their right mind would ever again, design a
> road-going EV controller
> to make such an awful, irritating noise?
>
> From Mike Sandman:
>
> >I just ordered a kit from KTA using a 1231C-7701
> for my 120v system.
>
> I'm curious Mike, did they tell you up front that
> this controller will
> cause your EV's drive motor to make an irritating
> non-varying squeal
> when you take off or cruise with a light throttle?
> Did they tell you
> that this controller has no safety features? Did
> they tell you it has no
> automatic precharge, and that if you don't design
> your own, it could
> blow up when the line contactor engages and send
> your EV off at full
> throttle? Did they tell you it cannot be adjusted
> for a battery input
> current limit to protect your battery pack
> investment? My guess is no to
> all the above.
>
> If it's not too late, you could save a bundle by
> purchasing a used
> Curtis 1221B controller. It's a bit less powerful at
> 400 amps, but it's
> silent and you could get one for maybe $300 if
> anyone has one available.
> The money saved would allow you to build a precharge
> circuit and pocket
> at least $500 cash, too. Better still, a used Raptor
> 600 could be found
> for maybe $800, is silent, properly cooled, has dip
> switch programming,
> and 900 amps motor loop current.
>
> I understand having to watch one's EV budget, but
> with one as high as
> $15k the last area I would have cut costs on would
> be on the all
> important motor controller. You may have saved
> $500-$600 by stepping
> w-a-ay down to the Curtis, but you gave up a lot of
> features that would
> have made your conversion more fun, far more
> pleasant, much smoother,
> more versatile, twice as powerful, and far safer,
> and...you gained a lot
> of negatives as described above.
>
> From Steve Powers:
>
> >When you get to more than 600 batt amps, you are
> really killing the pack.
>
>
>
> Is this one of those comments again, where one has
> to have read
> everything that came before and know that you might
> be talking about wet
> cell type batteries (and just didn't specify this),
> or do you really
> think that currents this low will harm powerful AGMs
> like Optimas,
> Hawkers, and Orbitals? Those of us using AGMs of
> higher quality know
> that the opposite is true, in that the high
> performance type AGMs
> actually give their best service life routinely
> dishing out BIG
> currents. 600 amps from the better AGMs is nothing.
> The Optimas in Blue
> Meanie lasted 6.5 years, and pretty close to each
> time it was taken out
> for a run, 1000 amps were sucked from them multiple
> times. Sure, under
> all out racing where you might suck 1000 amps from
> small 24.5 lb.
> batteries for 12 continuous seconds, it is stressful
> and sometimes fatal
> for the batteries, but for normal fun street driving
> 1000 amps for 5-6
> seconds from larger 40+ lb. AGMs is kid's
> stuff....besides, you're
> already well above the speed limit in that amount of
> time and will be
> off throttle and back to sub 100 amp levels.
>
> One more item to be added to Roger's list...compact
> size. Oat's Z1K's
> small near cube-shaped super compact size is not
> only amazing
> considering the thing cranks out 1000 full amps,
> it's also a blessing in
> an already crowded under-hood compartment. Though my
> experiences with
> the DCP Raptor 1200 were nothing but positive, its
> large bulky size was
> not a plus.
>
> See Ya......John Wayland
>
>
>
>
>
>
>
>
__________________________________________________
Do You Yahoo!?
Tired of spam? Yahoo! Mail has the best spam protection around
http://mail.yahoo.com
--- End Message ---
--- Begin Message ---
it has no space for things.
Hmmm. As I remember the EV1 had a cavernous trunk. LR>..........
--- End Message ---
--- Begin Message ---
Chris said this about the Insight.
Actual driver's opinons apparently differ, but there
is indeed only the hatch area behind the two seats
and it has pretty high cargo floor level, so that is
one of the reasons for debate.
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: +1-610-423-5743
Take your network further http://www.proxim.com
-----Original Message-----
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]
Behalf Of Lawrence Rhodes
Sent: Sunday, October 22, 2006 12:48 AM
To: [email protected]
Subject: Re: Who did kill the electric car?
it has no space for things.
Hmmm. As I remember the EV1 had a cavernous trunk. LR>..........
--- End Message ---
--- Begin Message ---
Hi Cor,
Since your controller is a special model for V2G it may have extra
capacitors. Regardless, with my math below including ESR, the ripple is
calculated as 1.5V or 3Vpp. I'm surprised and previously incorrect. I
think my skepticism was due to looking at only film capacitors because I
mistakenly believed the electrolytics would not offer a low-enough
resistance. That said, a DC controller needs a lot of current at low
duty cycles: your same power draw with a DC motor at 10% duty would have
a voltage ripple of 12.4Vpp. Excessive? Probably not, but the power
dissipation likely exceeds the capacitor rating (1kW/8). The ripple also
increases as your minimum duty cycle drops. Feel free to go through my
math below.
Again, I'm surprised -- even the lowly Curtis 1221B seems to have
enough capacitance! Provided the caps are as low ESR as the example...
Thanks for including details from your controller for this little
exercise.
Cory Cross
One thing you did not compensate for is internal resistance (worse
numbers) and that capacitor current is [motor-battery] (better numbers
and something I forgot before). To get some numbers, I picked the most
expensive 1500uF, 400V electrolytic capacitor Digikey has:
ECE-P2GA152HA. Specs are:
50mm x 80mm
7.31Arms @ 105*C and 10kHz
.049 ohms ESR @ 20*C and 20kHz
This would give us a power dissipation capability of 7.31^2 * .049 = 2.6W ?
If your capacitors are 2.2x capacitance and same voltage, they likely
have an ESR of 22mOhms.
All eight of your capacitors would have a resistance of 11mOhms. This
contributes a voltage drop of [EMAIL PROTECTED] in each direction
(charging/discharging) -- without considering any inductance effects.
The capacitors, in this case, are either charging or discharging at
100A, wasting 110W of power in the capacitors. That's 13.75W each or 5
times the Digikey cap is for only 2.2x bigger, but that isn't excessive
and well within our error margin (which is, what, 500% by now? :) ) (I x
t) / C = V = .375 V, plus 1.1V for ESR: 1.5Vm or 3Vpp. Admittedly, much
lower than I thought.
Take into consideration, however, the DC motor controller which must
operate a low duty cycles and high currents. If we assume the same power
(312V*200A*50%duty) 31.2kW, a constant 20kHz frequency, and a duty cycle
of 10%, we still get an average battery current of 100A. Now our motor
current is 1000A. The first time period of 45us is where the capacitor
is receiving 100A and the second time period of 5us where it is
delivering 900A. In the first 45us, the capacitor has a voltage drop due
to ESR of 1.1V and a capacitance raise of .7V. During the discharge
cycle, it again loses .7V for capacitance, but now there is an ESR drop
of 10V! The power burned in the caps is 90% * 1.1V * 100A + 10% * 9.9V *
900A = 990W! They certainly cannot do that for long!
Cor van de Water wrote:
>Hi Cory,
>
>I apologize that I do not have exact numbers.
>The particular VFD is in the back of my garage, behind a
>stack of junk that will take me too long to pull out, also
>my controller did not open up far enough to be able to
>read the capacity of the bank of electrolytes, but let's
>throw some assumptions at it and see where we end up:
>
>I have a tiny VFD with a 1500uF 400V capacitor that is
>about 2" in diameter and 6" long.
>My controller is about 2' long and has 8 elco's in one row
>so that is about 3" diameter per capacitor and the length
>is another swag, but let's say it is not shorter than the
>1500uF one.
>Since 1.5 times the diameter packs 2.25 times the surface,
>the capacitors in my controller are likely to be at least
>3300uF if not more - I suspect them to be 4700 but lets stick
>with the minimum for now.
>The configuration of the 8 in two series banks of 4 parallel
>makes the total voltage and capacitance both double.
>So 800V and 6600uF minimum for the total.
>
>Since this is an AC drive with an inaudible (Sine) waveform,
>I guess that they switch at least at 20 kHz if not higher.
>This means that maximum AC current will likely be at 50%
>duty cycle, which is 25us on and 25us off. This gives us
>the time base. The max current my controller cranks out
>draws 200A from the batteries, I suspect that they limited
>the controller in SW.
>
>Now the formula to find the voltage ripple is:
>I x t = V x C and solving for V:
>(I x t) / C = V entering the values we found above:
>(200A x 25us) / 6600uF =
>200 x 25 x 10^-6 x 150 = 0.75V
>
>So my controller likely has less than 1V ripple at max
>switching power at its current 312V pack.
>I tell you that this capacitor bank is totally absorbing
>the switching effect and my batteries see a constant
>current being drawn by the controller with a very small
>ripple due to the size of the capacitors, while they supply
>the energy for the pulses to my motor.
>
>I trust Otmar that he has optimized the Zilla to include
>capacitors large enough to filter the ripple to a small
>enough factor that also there the batteries essentially
>see a constant current, although at 1000A the ripple
>may be larger than in my case, still I would expect it
>to be only a few percent of the DC current, so the
>batteries are not affected significantly.
>
>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: +1-610-423-5743
>Take your network further http://www.proxim.com
>
>
>-----Original Message-----
>From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]
>Behalf Of Cory Cross
>Sent: Saturday, October 21, 2006 10:51 PM
>To: [email protected]
>Subject: Re: Current multiplication, was: EV controllers? Practical
>parame ters
>
>
>Hi Cor,
> The first purpose of capacitors is to protect the switching
>elements. In order to feed the motor they must be capable of around
>250Arms (in your case), which I doubt, in addition to needing enough
>capacitance inversely to PWM frequency and minimum duty cycle. Do you
>have part numbers and/or specs on those capacitors?
>
> None of the golf cart controllers have enough capacitance. Judging
>by the size of the Zilla, it does not either. I bet the VFD capacitors
>are for power faction correction.
>
>Cory Cross
>
>Cor van de Water wrote:
>
>
>
>>Hi Cory,
>>
>>Good controllers have large capacitors at the battery input
>>for this very reason - to smooth out the short pulses into
>>a constant battery current with very small ripple.
>>It is indeed a measure of how good the controller is, with
>>most SCR controllers this smoothing is not present, so they
>>will peak the same batetry current as the motor sees, making
>>the Peukert effect take a larger toll, that is why Lee
>>suggested to add large capacitors at the input of an SCR
>>controller, to increase range by drawing a constant current
>>
>>
>>from the batteries instead of short high peaks.
>
>
>>So - it depends on the controller!
>>
>>BTW, in my 700V 250A AC controller more than half the volume
>>of the controller is taken up by 8 huge capacitors.
>>I think they are switched as 2 series groups of 4 parallel
>>if they are using standard 400V electrolytes, but I have
>>not cut that deep into my controller yet.
>>
>>I have a salvage VFD (Variable Frequency Drive) that is
>>specified for 600V 3 phase and it also has 2 banks of
>>capacitors switched in series to take the voltage.
>>
>>Note that the higher the switching frequency, the smaller
>>the capacitors can be, as long as their internal resistance
>>is low enough to take the AC current of the charging and
>>in particular the discharging peaks.
>>
>>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: +1-610-423-5743
>>Take your network further http://www.proxim.com
>>
>>
>>-----Original Message-----
>>From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]
>>Behalf Of Cory Cross
>>Sent: Saturday, October 21, 2006 5:18 PM
>>To: [email protected]
>>Subject: Re: Current multiplication, was: EV controllers? Practical
>>parameters
>>
>>
>>Cor van de Water wrote:
>>
>>
>>
>>
>>
>>>The modern switching controllers allow the *motor* current
>>>to me many times larger than the *battery* current, in
>>>particular at low motor RPM.
>>>
>>>
>>>
>>>
>>>
>>>
>>Nitpick -- they allow AVERAGE battery current to be lower. I recently
>>requested information (on the evtech list) on any experiments to see if
>>the high-frequency, low-duty-cycle draws consume at a Peukert's rate of
>>the average or motor currents, but there don't appear to be any studies.
>>
>>Cory Cross
>>
>>
--- End Message ---
--- Begin Message ---
Still learning & planning
Given a 96 Volt Flooded Lead Acid battery pack using 16 six volt batteries,
what is the formula used to determine wattage & amperage? I know...basic ohms
law...Please explain. Thanks.
I have two projects:
1. 1980 Jet ElectraVan
2. Bradley GTElectric
RE: ElectraVan...if I stay with original configuration of 16 six V
flooded....What charger recommended to replace existing Lester?
Also, what DC/DC converter recommended for this ElectraVan?
With all this talk about Curtis & Zilla controllers, trying to decide what
amperage controller I might use to replace my existing GE EV1 SCR controller
and advantages/disadvantages of doing this.
Lots of questions for one posting...Thanks for advice
Don Davidson
--- End Message ---
--- Begin Message ---
If I had unlimited $ today, what would be the cost and what would be involved
with modifying my original 96 V DC configuration Bradley GTE with GE EV1 SCR
controller/Lester Charger as Bradley provided to NiMH batteries?
Advantages/disadvantages of considering this approach?
Don Davidson
--- End Message ---
--- Begin Message ---
Check out http://www.expita.com/nomime.html for instructions for many
email clients.
Jude
David Roden wrote:
On 19 Oct 2006 at 23:10, Mike Sandman wrote:
jim, i just see the message below. must be an HTML format thing ???
This happens periodically. The problem is that the email sent contains both
plain text and html. The listserver deletes the html to save bandwidth and
to prevent malware from being transmitted, replacing it with the "deleted"
message.
The problem is that it leaves the mime information intact. Your email
reader thus sees that the message as a multipart one, with both plain text
and html. Unless nagged, most email readers assume that, given a choice,
you'd rather see the html version since most folks prefer eye candy when
it's available. So it obligingly displays the html - with no idea that it's
showing you nothing but the "deleted" message.
Fixing this properly requires action at the sending end. A little effort at
the receiving end will help too. So, please do this :
1. Configure your email program so that it sends ONLY plain text emails, NOT
html, NOT "rich text," NOT multipart emails with both plain text and html.
Send plain text only.
2. Then, configure your email program to display only plain text, when given
a choice between plain text and html.
I don't know how to make these changes with the commonly used MS program
Outlook Express; maybe someone else does. (Please post this and I'll save
it for next time.) Perhaps others will provide the information for the
programs they use for email.
With Pegasus Mail, the program I use :
Click Tools - Options.
Under the Outgoing Mail item, click on Message Formatting. Tick the box
marked "Disable all text styling options."
Under the Incoming Mail item, click on Message Reader. Clear the box marked
"Where a choice of format exists, display the fancy version."
The first fixes the problem. The second ensures that if somebody messes up
and sends multipart text / html anyway, the reader will still see the plain
text and not the truncated html.
Clear as mud, right?
David Roden - Akron, Ohio, USA
EV List Assistant Administrator
= = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
Want to unsubscribe, stop the EV list mail while you're on vacation,
or switch to digest mode? See how: http://www.evdl.org/help/
= = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
Note: mail sent to "evpost" or "etpost" addresses will not reach me.
To send a private message, please obtain my email address from
the webpage http://www.evdl.org/help/ .
= = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = =
--- End Message ---
--- Begin Message ---
Hello Don,
W=Q x E
Where W = Watts is the energy absorbed in a
electrical circuit in watts per second
Q = Coulombs is the quantity of electricity
which is transferred between two points
E = Voltage or Electro Motive Force
Dividing both sides of the equation by time it takes for the quantity Q to
flow in a circuits becomes:
W/T = (Q x E)/T or P = I x E
Where P Is the Power in amount of time.
I is the rate of flow in seconds or hour
E is the e.m.f
Simplify to:
W = I x E (watts or energy - no time)
Wh = I x E (watt hour - amount for 1 hour)
Example:
If you are charging a battery pack at 240 volts AC at 50 amps AC, then
240V x 50A = 12 KW or 12,000 watts (input into the charger)
If you charge for 1 hour, than this will become 12 KWH or if you charge for
1 second, than the cost of Power will be 12 Kw/3600 secs (3600 seconds in
a hour).
If you drive the EV for one hour at 50 battery amps and 100 volts, than you
will use about 50 ADC x 100 VDC = 5KW Per Hour or 5KWH.
The HP (horsepower) use, where 1 HP = 746 watts:
HP = Watts/ 746 = 5000 / 746 = 6.7 HP
If you look on some motor labels on a motor, or in the specifications of
that motor, You may see data as:
165V @ 175A 32HP 6000 RPM 83% Effiency.
To calculate this:
W = 165V x 175A = 28,875 watts (input energy)
28,875 x .83 = 23,966 watts (output energy)
HP = 23,966 / 746 = 32.1 HP
Roland
----- Original Message -----
From: "Don" <[EMAIL PROTECTED]>
To: <[email protected]>
Sent: Sunday, October 22, 2006 5:22 AM
Subject: Formula
> Still learning & planning
>
> Given a 96 Volt Flooded Lead Acid battery pack using 16 six volt
> batteries, what is the formula used to determine wattage & amperage? I
> know...basic ohms law...Please explain. Thanks.
>
> I have two projects:
>
> 1. 1980 Jet ElectraVan
> 2. Bradley GTElectric
>
> RE: ElectraVan...if I stay with original configuration of 16 six V
> flooded....What charger recommended to replace existing Lester?
> Also, what DC/DC converter recommended for this ElectraVan?
>
> With all this talk about Curtis & Zilla controllers, trying to decide what
> amperage controller I might use to replace my existing GE EV1 SCR
> controller and advantages/disadvantages of doing this.
>
> Lots of questions for one posting...Thanks for advice
> Don Davidson
>
>
--- End Message ---
--- Begin Message ---
Roland
Thanks. I will study this
Don
----- Original Message -----
From: "Roland Wiench" <[EMAIL PROTECTED]>
To: <[email protected]>
Sent: Sunday, October 22, 2006 9:19 AM
Subject: Re: Formula
> Hello Don,
>
> W=Q x E
>
> Where W = Watts is the energy absorbed in a
> electrical circuit in watts per second
>
> Q = Coulombs is the quantity of electricity
> which is transferred between two points
>
> E = Voltage or Electro Motive Force
>
>
> Dividing both sides of the equation by time it takes for the quantity Q to
> flow in a circuits becomes:
>
> W/T = (Q x E)/T or P = I x E
>
> Where P Is the Power in amount of time.
>
> I is the rate of flow in seconds or hour
>
> E is the e.m.f
>
>
> Simplify to:
>
> W = I x E (watts or energy - no time)
>
> Wh = I x E (watt hour - amount for 1 hour)
>
>
> Example:
>
> If you are charging a battery pack at 240 volts AC at 50 amps AC,
then
> 240V x 50A = 12 KW or 12,000 watts (input into the charger)
>
> If you charge for 1 hour, than this will become 12 KWH or if you charge
for
> 1 second, than the cost of Power will be 12 Kw/3600 secs (3600 seconds
in
> a hour).
>
> If you drive the EV for one hour at 50 battery amps and 100 volts, than
you
> will use about 50 ADC x 100 VDC = 5KW Per Hour or 5KWH.
>
> The HP (horsepower) use, where 1 HP = 746 watts:
>
> HP = Watts/ 746 = 5000 / 746 = 6.7 HP
>
>
> If you look on some motor labels on a motor, or in the specifications of
> that motor, You may see data as:
>
> 165V @ 175A 32HP 6000 RPM 83% Effiency.
>
> To calculate this:
>
> W = 165V x 175A = 28,875 watts (input energy)
>
> 28,875 x .83 = 23,966 watts (output energy)
>
> HP = 23,966 / 746 = 32.1 HP
>
>
> Roland
>
>
>
>
>
>
>
>
>
>
> ----- Original Message -----
> From: "Don" <[EMAIL PROTECTED]>
> To: <[email protected]>
> Sent: Sunday, October 22, 2006 5:22 AM
> Subject: Formula
>
>
> > Still learning & planning
> >
> > Given a 96 Volt Flooded Lead Acid battery pack using 16 six volt
> > batteries, what is the formula used to determine wattage & amperage? I
> > know...basic ohms law...Please explain. Thanks.
> >
> > I have two projects:
> >
> > 1. 1980 Jet ElectraVan
> > 2. Bradley GTElectric
> >
> > RE: ElectraVan...if I stay with original configuration of 16 six V
> > flooded....What charger recommended to replace existing Lester?
> > Also, what DC/DC converter recommended for this ElectraVan?
> >
> > With all this talk about Curtis & Zilla controllers, trying to decide
what
> > amperage controller I might use to replace my existing GE EV1 SCR
> > controller and advantages/disadvantages of doing this.
> >
> > Lots of questions for one posting...Thanks for advice
> > Don Davidson
> >
> >
>
>
--- End Message ---
