EV Digest 6099
Topics covered in this issue include:
1) Re: EV controllers? the 4th option...
by Jack Murray <[EMAIL PROTECTED]>
2) RE: Lee your BMS anyone else interested?
by Cor van de Water <[EMAIL PROTECTED]>
3) unsubscribe
by Darin Phillips <[EMAIL PROTECTED]>
4) Re: Lithium-ion batteries & Valence Group buy
by "Rush" <[EMAIL PROTECTED]>
5) Re: Pusher Trailer
by Jack Murray <[EMAIL PROTECTED]>
6) Re: Lithium-ion batteries & Valence Group buy
by "Philippe Borges" <[EMAIL PROTECTED]>
7) Heaters
by "Tom Gocze" <[EMAIL PROTECTED]>
8) Re: EV controllers? the 4th option...
by Lee Hart <[EMAIL PROTECTED]>
9) Re: A hundred tiny toy car motors vs one big one.
by Lee Hart <[EMAIL PROTECTED]>
10) RE: Heaters
by Cor van de Water <[EMAIL PROTECTED]>
11) Re: Input requested on heater assembly
by John <[EMAIL PROTECTED]>
12) Re: Vacuum system help
by John Nicholson <[EMAIL PROTECTED]>
13) RC controllers (was: RE: EV controllers? the 4th option...)
by "Myles Twete" <[EMAIL PROTECTED]>
14) Re: RC controllers (was: RE: EV controllers? the 4th option...)
by Jack Murray <[EMAIL PROTECTED]>
15) EV shows - crazy idea?
by "Paul G." <[EMAIL PROTECTED]>
16) Re: IOTA DC/DC VOLTAGE LIMITS?
by Rod Hower <[EMAIL PROTECTED]>
17) Re: IOTA DC/DC VOLTAGE LIMITS?
by MARK DUTKO <[EMAIL PROTECTED]>
18) Re:Current Eliminator news
by [EMAIL PROTECTED]
19) RE: Lee your BMS anyone else interested?
by Cor van de Water <[EMAIL PROTECTED]>
--- Begin Message ---
My R/C receivers sans the radio part can do PWM with input from a 5k
pot. This was used to test my sailboat electric outboard (PM DC) that
used a full bridge so would do forward, brake, reverse.
The receiver uCPU board is about $15 and about the size of your
thumbnail. It is limited to 16v though.
The power board with the mosfets was seperate.
It isn't the hard part for sure. 160v and 500amps is a whole different
animal. I've got a couple of the 600amp IGBT modules, I'll try it
today on an 8" ADC motor.
You can actually use an R/C system designed to control a 9v dc motor to
generate the PWM signal for a mosfet.
A low-side mosfet can switch with 10v, so that's looks possible.
Consider the radio link of the controller (with uses a 5k pot on the
throttle trigger) to be good isolation from the power side. :)
Jack
Danny Miller wrote:
Hey that's what I've been trying to say. The controller is the easy
part. There is no question in my mind that a microcontroller is a much
better choice than discretes for driving the signals. It's easy,
featured, safe, and reliable, relatively speaking. In the case of a
3-ph drive, it will compress months or years of board development
requiring a lot of experience into a few weeks of code development- or,
if you have experience, only a few days.
However, it's not going to make the transistors, transistor drivers,
capacitors, diodes, and kiloamp-capable interconnects any easier or
cheaper. Some of the conversation here seems to be leaning towards
"once we have a way to make a PWM signal, we basically have an open
source controller!" OK, enthusiasm is good, keep it up, but there is a
world of difference between having a controller and having the easy part
of it.
Danny
Lee Hart wrote:
A DC motor controller is a buck converter. It consists of a big
transistor, a big diode, a big capacitor, and a big inductor (usually
provided by the motor itself). The first challenge is to pick these
parts.
The "control" section is just a simple switching regulator IC chip --
any one of a hundred different chips will do. You can even do it
without ICs, a couple dozen discrete transistors, resistors, and
capacitors, if you like. So, the second challenge is to pick these parts.
The third challenge is to decide exactly how you wire up the
high-power section. You have to keep the leads as short and thick as
possible. It is extremely difficult to use PC boards for this; buss
bars or other techniques are better.
Finally, there is the thermal design; how to get the heat out of it.
Poor thermal design means you can't even get half the rated power out
of your expensive semiconductors. It usually involves big expensive
heatsinks and complex machining.
Let's concentrate on these area first. These are the hard parts, where
schematics don't provide much help, and most people have little or no
experience. Once you get the power section working, then you can
debate what microcontroller to use for your dash display, etc.
--
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 ---
Lee,
I was wondering about features of the balancer.
You mentioned that it took you a while to find out
that you had a bad battery, because the balancer
would continuously "prop it up" from the other 11
to keep it going. How did you find out one was bad?
It may have merit to indicate a (yellow) light when
the balancer is spending more than twice the average
time on a battery, to have an early indication that
one or more batteries are going south.
And does your BMS indicate (red) when a battery goes
below 10.5V to avoid damaging it, even though the BMS
is propping it up, it cannot supply the full drive
current. In extreme cases, a battery could reverse
and damage the BMS.
With the "idiot light" on the dash telling that a
battery is below 10.5V the driver can ease the foot
on the accelerator and limp to a safe place without
breaking anything.
In the case where one weak battery is the limit, the EV
would even be driveable after waiting a bit, because the
balancer will re-charge the weak battery and you can go
a little further again, until home or the 10.5V is reached
for all batteries, then you need to bring the plug to the
EV or tow it.
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 Lee Hart
Sent: Saturday, November 04, 2006 12:05 PM
To: ev@listproc.sjsu.edu
Subject: Re: Lee your BMS anyone else interested?
Don Blazer <[EMAIL PROTECTED]> wrote:
> I was looking at your BMS with NiMH batteries in mind now. You might
> not remember but a while back I was trying to get you to do a bulk
> order for use on AGM batteries. I would still be willing to buy in
> quantity for the best prices and sell at the quantity price break
> to all who is interested in making one.
Don, you may have meant to send this to me directly, but since it went
to the EV list, I'll answer it here.
Yes, I have been working on a new version of my Battery Balancer for
over a year now. Unfortunately, it is a home project, and so has a low
priority compared to "real work" and the "honey do" list.
At present, I have the new Relay board design done, and can order them
any time. I have the new Control board about 75% done. But there are
some loose ends that I can't help but keep tinkering with.
Specifically, the old control board used an external digital multimeter
with isolated RS-232 output to measure battery voltage and current. They
cost $50-$100. I'm experimenting with a cheaper on-board circuit to do
this (since everyone is obsessed with cost).
I am really at the point where I need to just "shoot the engineer, and
ship it!" I have several people who have been very patient, and just
need to go ahead and order the boards for them, saving any further
improvements for the next go-around.
--
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 ---
unsubscribe
--- End Message ---
--- Begin Message ---
David wrote -
> Cycle life : It could have changed since then, but the last number I saw for
> T-125 BCI cycle life was 650. Compared to that, a battery with a BCI rated
> cycle life of 2500 (however we must note that this is just an estimate for
> Valence) would be less than THREE times greater (or not quite 4 times as
> much, if you prefer the larger number). It's not "5 times better."
I got the cycle life from Trojanproductspecsguide.pdf, from their website. They
say 492 LEU's, Trojans way of measuring, but I just rounded it up to 500 for
ease. In the same way that Valence's website says 'thousands' of cycles, I
arbitraliary chose 2,500 for ease of numbers.
> 300% AS MANY cycles.
That is what I meant to say, thanks for clearing it up.
Rush
Tucson AZ
www.ironandwood.org
--- End Message ---
--- Begin Message ---
The idea of a car without a motor being pushed by a trailer however
could be problematic because the car itself would likely not be legal.
A trailer is probably defined as being pulled by a legal vehicle. A
pusher trailer is not being pulled, so might not actually be a legal
trailer. Now that is very legalistic, as many have said, unless some
person or company has a reason to complain about them, unlikely you'd
get into trouble, except perhaps if you got into an accident with it,
then the injuried party might use the lack of legal status against you.
Frankly, I think a pusher is an innovative idea, but wouldn't want on
myself. Building a car from scratch with an exchangeable drivetrain as
I mentioned before would be a better idea, but of course a LOT more work.
Jack
Peter VanDerWal wrote:
I believe the segway laws usually include a top speed. something like
10-12 mph, not vbery useful for a pusher trailer.
Totally irrelevent however, since trailers are already legal and included
in every states current laws. No state currently has a law AGAINST pusher
trailers, so why fight a battle that isn't there?
P.s. Self balancing is a pretty definitive description that means it
balances by itself. If it's attached to something, then it's not self
balancing.
Michael Perry wrote:
I inquired further and was told that no vehicle is legal, unless
specified
in the Oregon Statutes.
Here's a clever/stupid idea. Dean Kamen has paid to get laws on the
books in virtually every state to allow his Segway. The laws usually
state a "two wheel, self balancing, non tandem" vehicle is not regulated
by any existing vehicle laws (i.e. it's not a car, motorcycle, bicycle,
skateboard, scooter, etc.), and can be ridden "anywhere a pedestrian can
legally walk", without a helmet, driver's license, insurance,
registration, etc.
What if you build a Segway-like device as a pusher trailer? Two wheels,
side by side (non tandem), and self-balancing because it's attached to
the vehicle it is pushing.
Same for scooters: My BEST kids came up with at least half a dozen ways
to make a 2-wheel self-balancing, non-tandem vehicles without any
gyroscopes, computers, or complex gadgetry.
--
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 ---
Same analyze here as i don't need/want to pay premium for lithum advantage
(weight-volume saving)
i suppose reason it's not mass used for rEVolutionize EV world is that when
OEM ask for waranty...lithium manufacturer have to say no sorry...
but this is pur speculation from me...
I heard also than Milwaulkee tools give a 5 years waranty on their V28
product...so futur is open
cordialement,
Philippe
Et si le pot d'échappement sortait au centre du volant ?
quel carburant choisiriez-vous ?
http://vehiculeselectriques.free.fr
Forum de discussion sur les véhicules électriques
http://vehiculeselectriques.free.fr/Forum/index.php
----- Original Message -----
From: "Curtis Muhlestein" <[EMAIL PROTECTED]>
To: <ev@listproc.sjsu.edu>
Sent: Saturday, November 04, 2006 7:06 PM
Subject: RE: Lithium-ion batteries & Valence Group buy
> If they only have 5 years, then with average mileage of 15,000 a year, you
> will only get 75,000 miles out of the batteries and if they cost $25,000
for
> a set, that is $5,000 a year. I think I will buy Lead Acid.
>
>
> -----Original Message-----
> From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On
> Behalf Of Philippe Borges
> Sent: Saturday, November 04, 2006 10:31 AM
> To: ev@listproc.sjsu.edu
> Subject: Re: Lithium-ion batteries & Valence Group buy
>
> -TS:
> i don't know
>
> -Kokam:
> CEO answered me honestly: 5 years until important capacity loss (he
admitted
> 50% was possible)
>
>
>
> cordialement,
> Philippe
>
> Et si le pot d'échappement sortait au centre du volant ?
> quel carburant choisiriez-vous ?
> http://vehiculeselectriques.free.fr
> Forum de discussion sur les véhicules électriques
> http://vehiculeselectriques.free.fr/Forum/index.php
>
>
> ----- Original Message -----
> From: "Osmo S." <[EMAIL PROTECTED]>
> To: <ev@listproc.sjsu.edu>
> Sent: Saturday, November 04, 2006 6:01 PM
> Subject: Re: Lithium-ion batteries & Valence Group buy
>
>
> > Philippe Borges kirjoitti 4.11.2006 kello 15.19:
> >
> > > Am i alone wanting to know what Valence say about calendar life ?
> >
> > No, this is a good question you have raised. Maybe they don´t know it
> > at Valence??
> >
> > Do we know the calendar life of other lithium batteries? Kokam? TS?
> >
> > Osmo
> >
>
>
>
--- End Message ---
--- Begin Message ---
Now that it is the heating season in the fun states, there is a lot more
discussion about auxillary heaters.
It seems that a startup resistor with the rectactor controller solves this
problem without all the hassle of lots of high cost silicon.
Wouldn't it make sense for someone to package a three step (24/48/96V or
perhaps 36/72/144V?) system with a big starting resistor?
Years ago I did some ad hoc analysis as to how much time a resistor would be in
the circuit. An educated guess is 5%. That would be the time it provides heat.
The rest of the time it is out of the circuit and the efficiency is limited to
the losses in the diodes. That 5% is probably more than a heater affords us.
In Maine, we could use that heat about 8 months out of the year.
Wouldn't that give us all the tire squealing, flashy startup torque we want for
less money and not be a unserviceable hassle that we have all dealt with?
This will likely be part of my wintertime project, especially if I cook the
1231 in my S10.
--- End Message ---
--- Begin Message ---
Lee Hart wrote:
Dead ones are fairly common. I learned to build controllers
by buying dead ones and doing an autopsy.
Eric Poulsen wrote:
If the price were right (low), I might just buy a dead one.
Where would you find such a thing?
Let's start by asking on the list. Does anyone have a dead controller
that Eric could buy/borrow to trace out? It can be dead and unfixable.
The important part is the control board. The power board's schematic is
very simple and not a problem.
I have a couple dead Curtis controllers at the moment, but I'm fixing
them for people. One I bought for $500; it died from water damage, and
is mostly intact. The second Bob Rice sent me; it has a bunch of blown
MOSFETs but otherwise looks repairable.
The Curtis design isn't all that great... But it isn't bad, either.
When I look at that schematic, the control section seems really
old-fashioned.
It was designed about 15 years ago. Is that "old fashioned" to you? :-)
An advantage of this design is that all of the parts are simple,
inexpensive, and generic. A quad opamp only costs $0.25, and resistors
are only a penny. Every part was available 20 years ago, and will
*still* be available in another 20 years. It is simple to understand how
it works, or to fix or modify it.
The disadvantage is that there are a lot of parts to assemble by hand.
There is no micro, so no software, and no way to change limits except by
changing resistors. There are 3 pots to set the most common limits, but
it could use more. They designed a single board to be used in many
different controllers; so there are a lot of parts that are unnecessary
for any one specific version.
What's of particular interest is the power section and the gate drivers,
and the overcurrent sensing. Seems to me this is where the focus needs
to occur.
I agree. These are also the areas where a lot has been learned since the
Curtis was designed, so we can do better today.
Control is very important, but op amps and complicated linear circuitry
seems a bit much.
A lot of this circuitry is now available in off-the-shelf switching
regulator chips.
"The project" being the Curtis reverse engineering? Most people don't
have good things to say about them.
Most people don't have good things to say about *any* controller! They
are all too expensive, too hard to get, don't have the features they
want, etc. :-)
Basically, I think that if you want to build your own controller, and
don't have a lot of skill, then the best way to be successful is to
build a (simplified) version of the Curtis control board, and use it to
drive a power module with the transistors and diodes already in it.
I'm not sure a schematic that says "Beware ... made from chicken
scratchings" is all you say it is. It's a good reference, maybe.
But my entire point was that is _isn't_ by any stretch of the
imagination something you could build a working controller from
It is certainly good enough that anyone reasonably skilled in the art
*can* build a working controller from it.
Mr. Borges schematic is (literally) unreadable, scanned at too low
a resolution.
There must be some problem with different browsers. I can read 99% of
that schematic on my system. The parts I can't read, I can figure out
what they must be. Of course, you can also get the original book.
If I had two choices of 1) Buying a $2500 controller and using it,
or 2) Building a controller for a whole lot more and releasing the
design, I'd probably choose option #2.
That's good... altruism at its finest! I am certainly willing to do all
that I can to help in such efforts.
What rubs me the wrong way are the people that talk about what someone
*else* should do for them for free; but they aren't willing to
contribute anything themselves.
"you won't save money"...
The worst reason to get into EVs is "so I can save money on gasoline."
Once people get past that illusion, then they can make real progress.
There are many reasons to do things besides money!
Personally, I try hard to support the people that are building EV
products. If I want something like a Zilla controller, I'll *buy* a
Zilla. Otmar isn't getting rich; he works hard, and he deserves my money.
I only build my own if I *need* to for some reason. Maybe the product I
need doesn't exist, or doesn't fit, or is unavailable for some reason.
Or I have a new idea I want to try out. Or, for the educational value.
Or, just for fun. I almost never do it to save money.
I might like to point out that neither books nor magazines are free,
and don't tell me the library has this sort of stuff, because they
don't. I've checked.
Then you are looking the wrong places. Go to any university engineering
library. They have more technical books than you could possibly read in
a lifetime, all free. Even your local public library is probably part of
an inter-library loan program; they can get books from other libraries
for you for free.
Subscribe to any of the dozens of technical magazines. They are almost
all free, provided you "look like" a working engineer to them.
Request copies of application notes from all the big semiconductor
manufacturers. They are all free, and have huge amounts of data.
And, there are vast online resources.
But again, don't cripple yourself by thinking that "everything has to be
free". Why should I help deadbeats and freeloaders that want everything
for nothing?
--
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 ---
GWMobile wrote:
Does the efficiency of a motor go up or down with size?
Generally, the bigger the motor, the more efficient it is. So, it is
usually better to have one big motor than many smaller ones.
--
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 ---
Trouble is, you can't control when you need heat.
Doing constant cuise on the freeway won't allow you
to use the heater, or your speed drops...
You may be able to use the waste heat from a
startup resistor to _add_ to the heat, but
you better not rely on it in winter months.
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 Tom Gocze
Sent: Saturday, November 04, 2006 2:44 PM
To: ev@listproc.sjsu.edu
Subject: Heaters
Now that it is the heating season in the fun states, there is a lot more
discussion about auxillary heaters.
It seems that a startup resistor with the rectactor controller solves this
problem without all the hassle of lots of high cost silicon.
Wouldn't it make sense for someone to package a three step (24/48/96V or
perhaps 36/72/144V?) system with a big starting resistor?
Years ago I did some ad hoc analysis as to how much time a resistor would be
in the circuit. An educated guess is 5%. That would be the time it provides
heat. The rest of the time it is out of the circuit and the efficiency is
limited to the losses in the diodes. That 5% is probably more than a heater
affords us. In Maine, we could use that heat about 8 months out of the year.
Wouldn't that give us all the tire squealing, flashy startup torque we want
for less money and not be a unserviceable hassle that we have all dealt
with?
This will likely be part of my wintertime project, especially if I cook the
1231 in my S10.
--- End Message ---
--- Begin Message ---
On Friday, November 3, 2006, at 08:54 PM, Lee Hart wrote:
John wrote: The thermal switch is in series with the 12v that supply
to the HV relay package (from KTA) . Also in this circuit is the
blower relay that is stock on the truck and a rocker switch that will
be mounted on the dash.
The part shown is a thermal fuse, not a switch. It opens if it ever
gets too hot, and remains open forever.
That is correct. I used the wrong term. If the temp in the heat
exchange area rises above 77 degrees Celsius in theory the relay
feeding traction pack voltage to the heater element opens and I have to
get in there and replace the fuse to get heat back on.
There should also have been a thermal switch, which is a bimetallic
switch that opens if it gets too hot, and closes when it cools back
down.
Both of these parts from the stock heater are AC only, and won't
reliably work over about 30 volts DC.
Now I fear the the thermal fuse I show in my picture may not be
appropriate. It is a replacement part for home appliances running on
120v AC.
http://www.goodmans.net/get_item_th-tf77c_thermal-fuse-77-degrees-
celsius.htm
However, it is in a 12v DC circuit, and I could not begin to estimate
how many amps would be involved so please follow along and let me know
if I am creating a fire hazard.
I can't correctly draw circuits on paper never mind ASCII so I will try
and describe with words.
1) 12V + to stock fuse panel in the truck (currently a 10 amp fuse)
2) Fuse panel to stock blower relay (relay closes when fan control is
anywhere other than off. For reasons I don't understand the truck
computer wanted to know when the fan was on)
3) Blower relay to manual rocker switch (rated for 20amps 12V DC
purchased from Pep Boys) to be installed on the dash somewhere
4) rocker switch to thermal fuse
5) thermal fuse to 12 v + on KTA heater relay package with snubber
network
6) KTA relay to ground.
So when the blower and rocker switch are both on, and the thermal fuse
is still closed, the KTA relay closes and sends traction voltage to the
heater element.
Should this work and be safe?
John
--- End Message ---
--- Begin Message ---
On Sat, 2006-11-04 at 10:08 -0500, John wrote:
> Unfortunately I still have a small leak somewhere in the system but can
> hold a vacuum for 5 min before the pump kick in again.
You might try pressurizing the system and using a soap solution to check
for bubbling - you may locate the leak that way.
John
--- End Message ---
--- Begin Message ---
Jack-
I think clarification is in order here.
You make it sound as if the rear end of the recevier does PWM.
It does not. If there is PWM, it's the servo motor electronics which does
the PWM.
For decades the standard for R/C servo signals from receiver to servo has
been PPM---i.e. Pulse Position MOdulation.
The multichannel demod signal in the receiver gets demultiplexed before
leaving the R/C receiver, then the demux'd PPM signals for each channel get
sent out each servo channel connector. That signal isn't PWM, it's a
position command to the servo, with the servo motor circuitry determining
position based on the width between the pulses. IIRC, the null, or midrange
position command PW is roughly 1.3ms, with +/- 0.6ms or so as the dynamic
range from null. The servo motor circuitry uses the pulse width to
determine where to position the servo motor.
Now, it probably is the case that the output of servo drive electronics is
also a pulsed output drive to the motor. And I suppose it effectively is a
PWM signal.
But the command from the receiver is not PWM.
So, back to what I think your idea for the basic PWM is:
Dissect a R/C servo and remove the 5 or 10k pot which provides position
feedback on the output. In its place, wire in a separate external pot,
which now would instead act as a position command signal for an open loop
drive instead of a feedback signal for closed loop servo control. Sure,
that would work as long as the particular servo electronics is okay with a
PPM signal coming in with infinite inter-pulse period. Which may well be
commonly the case.
-MT
> -----Original Message-----
> From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]
> Behalf Of Jack Murray
> Sent: Saturday, November 04, 2006 1:25 PM
> To: ev@listproc.sjsu.edu
> Subject: Re: EV controllers? the 4th option...
>
>
> My R/C receivers sans the radio part can do PWM with input from a 5k
> pot. This was used to test my sailboat electric outboard (PM DC) that
> used a full bridge so would do forward, brake, reverse.
> The receiver uCPU board is about $15 and about the size of your
> thumbnail. It is limited to 16v though.
> The power board with the mosfets was seperate.
> It isn't the hard part for sure. 160v and 500amps is a whole different
> animal. I've got a couple of the 600amp IGBT modules, I'll try it
> today on an 8" ADC motor.
>
> You can actually use an R/C system designed to control a 9v dc motor to
> generate the PWM signal for a mosfet.
> A low-side mosfet can switch with 10v, so that's looks possible.
> Consider the radio link of the controller (with uses a 5k pot on the
> throttle trigger) to be good isolation from the power side. :)
>
> Jack
>
> Danny Miller wrote:
> > Hey that's what I've been trying to say. The controller is the easy
> > part. There is no question in my mind that a microcontroller is a much
> > better choice than discretes for driving the signals. It's easy,
> > featured, safe, and reliable, relatively speaking. In the case of a
> > 3-ph drive, it will compress months or years of board development
> > requiring a lot of experience into a few weeks of code development- or,
> > if you have experience, only a few days.
> >
> > However, it's not going to make the transistors, transistor drivers,
> > capacitors, diodes, and kiloamp-capable interconnects any easier or
> > cheaper. Some of the conversation here seems to be leaning towards
> > "once we have a way to make a PWM signal, we basically have an open
> > source controller!" OK, enthusiasm is good, keep it up, but there is a
> > world of difference between having a controller and having the
> easy part
> > of it.
> >
> > Danny
> >
> > Lee Hart wrote:
> >
> >> A DC motor controller is a buck converter. It consists of a big
> >> transistor, a big diode, a big capacitor, and a big inductor (usually
> >> provided by the motor itself). The first challenge is to pick these
> >> parts.
> >>
> >> The "control" section is just a simple switching regulator IC chip --
> >> any one of a hundred different chips will do. You can even do it
> >> without ICs, a couple dozen discrete transistors, resistors, and
> >> capacitors, if you like. So, the second challenge is to pick
> these parts.
> >>
> >> The third challenge is to decide exactly how you wire up the
> >> high-power section. You have to keep the leads as short and thick as
> >> possible. It is extremely difficult to use PC boards for this; buss
> >> bars or other techniques are better.
> >>
> >> Finally, there is the thermal design; how to get the heat out of it.
> >> Poor thermal design means you can't even get half the rated power out
> >> of your expensive semiconductors. It usually involves big expensive
> >> heatsinks and complex machining.
> >>
> >> Let's concentrate on these area first. These are the hard parts, where
> >> schematics don't provide much help, and most people have little or no
> >> experience. Once you get the power section working, then you can
> >> debate what microcontroller to use for your dash display, etc.
> >> --
> >> 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 ---
Correct, you need more than just the receiver, you also need the motor
controller. You are using the positive output from the R/C motor
controller that normally would go to the drive motor, to drive the gate
of a power Mosfet in your 120v 600amp power loop.
Would I recommend you actually do this for a real car? Probably not,
but it should work, and you can buy a $39 r/c car at radio shack, the
point being the PWM signal is not an issue.
My NiMH battery chargers will also output PWM at 12volts with 16 speeds
using a 5k pot which is used for r/c motor brush breakin to limit their
voltage after changing the brushes. I'm using that board to test the
IGBT modules, they need 15v for the gate drive. Of course I changed the
firmware so it isn't a fixed 16 speeds.
Jack
Myles Twete wrote:
Jack-
I think clarification is in order here.
You make it sound as if the rear end of the recevier does PWM.
It does not. If there is PWM, it's the servo motor electronics which does
the PWM.
For decades the standard for R/C servo signals from receiver to servo has
been PPM---i.e. Pulse Position MOdulation.
The multichannel demod signal in the receiver gets demultiplexed before
leaving the R/C receiver, then the demux'd PPM signals for each channel get
sent out each servo channel connector. That signal isn't PWM, it's a
position command to the servo, with the servo motor circuitry determining
position based on the width between the pulses. IIRC, the null, or midrange
position command PW is roughly 1.3ms, with +/- 0.6ms or so as the dynamic
range from null. The servo motor circuitry uses the pulse width to
determine where to position the servo motor.
Now, it probably is the case that the output of servo drive electronics is
also a pulsed output drive to the motor. And I suppose it effectively is a
PWM signal.
But the command from the receiver is not PWM.
So, back to what I think your idea for the basic PWM is:
Dissect a R/C servo and remove the 5 or 10k pot which provides position
feedback on the output. In its place, wire in a separate external pot,
which now would instead act as a position command signal for an open loop
drive instead of a feedback signal for closed loop servo control. Sure,
that would work as long as the particular servo electronics is okay with a
PPM signal coming in with infinite inter-pulse period. Which may well be
commonly the case.
-MT
-----Original Message-----
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]
Behalf Of Jack Murray
Sent: Saturday, November 04, 2006 1:25 PM
To: ev@listproc.sjsu.edu
Subject: Re: EV controllers? the 4th option...
My R/C receivers sans the radio part can do PWM with input from a 5k
pot. This was used to test my sailboat electric outboard (PM DC) that
used a full bridge so would do forward, brake, reverse.
The receiver uCPU board is about $15 and about the size of your
thumbnail. It is limited to 16v though.
The power board with the mosfets was seperate.
It isn't the hard part for sure. 160v and 500amps is a whole different
animal. I've got a couple of the 600amp IGBT modules, I'll try it
today on an 8" ADC motor.
You can actually use an R/C system designed to control a 9v dc motor to
generate the PWM signal for a mosfet.
A low-side mosfet can switch with 10v, so that's looks possible.
Consider the radio link of the controller (with uses a 5k pot on the
throttle trigger) to be good isolation from the power side. :)
Jack
Danny Miller wrote:
Hey that's what I've been trying to say. The controller is the easy
part. There is no question in my mind that a microcontroller is a much
better choice than discretes for driving the signals. It's easy,
featured, safe, and reliable, relatively speaking. In the case of a
3-ph drive, it will compress months or years of board development
requiring a lot of experience into a few weeks of code development- or,
if you have experience, only a few days.
However, it's not going to make the transistors, transistor drivers,
capacitors, diodes, and kiloamp-capable interconnects any easier or
cheaper. Some of the conversation here seems to be leaning towards
"once we have a way to make a PWM signal, we basically have an open
source controller!" OK, enthusiasm is good, keep it up, but there is a
world of difference between having a controller and having the
easy part
of it.
Danny
Lee Hart wrote:
A DC motor controller is a buck converter. It consists of a big
transistor, a big diode, a big capacitor, and a big inductor (usually
provided by the motor itself). The first challenge is to pick these
parts.
The "control" section is just a simple switching regulator IC chip --
any one of a hundred different chips will do. You can even do it
without ICs, a couple dozen discrete transistors, resistors, and
capacitors, if you like. So, the second challenge is to pick
these parts.
The third challenge is to decide exactly how you wire up the
high-power section. You have to keep the leads as short and thick as
possible. It is extremely difficult to use PC boards for this; buss
bars or other techniques are better.
Finally, there is the thermal design; how to get the heat out of it.
Poor thermal design means you can't even get half the rated power out
of your expensive semiconductors. It usually involves big expensive
heatsinks and complex machining.
Let's concentrate on these area first. These are the hard parts, where
schematics don't provide much help, and most people have little or no
experience. Once you get the power section working, then you can
debate what microcontroller to use for your dash display, etc.
--
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 ---
Currently, it seems, photos from EV shows and races are scattered all
over the web. Past shows can be hard to find and I wonder if I found
all the good photos. I am considering creating a web site, (along the
lines of the EValbum) but dedicated to EV shows, races and other EV
gatherings. I own the domain EVfun and I thought it would be a good web
home for EVents.
It would be easiest to host the photos on this web site, that would in
no way restrict the photographer or club from having the photos on
their web site too. Alternatively, it is possible to link to other
sites by show but I fear moved or changed link issues could become an
issue. My son Phillip is working on his web design degree and has been
hand coding html for several years already. We have been talking about
a scripted site that would make my work reasonably easy, but not the
canned site approach (custom scripting by Phillip.)
Good idea? Crazy talk? I welcome the lists input!
Paul "neon" G.
--- End Message ---
--- Begin Message ---
Jeff,
The 240Vac input IOTA rectifies and makes that input
340Vdc nominal. It then bucks that voltage down to
13.8Vdc. Most if not all 240Vac controls including 3
phase inverter controls (which I designed at Baldor
Motors and Drives) have a +/- 10% minimum range for
the input voltage. Therefore a 268Vac input is 379Vdc
max and my inverter controls shut down on overvoltage
at 380Vdc. I think this is a safe range for the IOTA.
The bus caps are probably rated for 400Vdc and 450Vdc
surge. The IGBT's are rated for 600Vdc. I don't
think 354Vdc is anything to worry about.
Rod
Designing 28Vdc inverters these days.
FYI. I'm currently working on a 24Vdc 3 phase inverter
for BLDC that is 0.7" in diameter. Talk about putting
10 pounds of *hit in a 1 lb. bag!!
--- Jeff Shanab <[EMAIL PROTECTED]> wrote:
> Sorry, I have a 288V nominal pack but here is one
> data point for the record.
>
> My end of charge voltage is 354 Volts. I have been
> worried that that
> may take the iota out.
> The IOTA is connected to the pack with only a fuse,
> It stays charged and
> it goes to sleep nicely.
> I have readlighted* a few times and it doesn't
> hickup.
>
>
>
>
> *redlighted in this context is when I suck the 1/2
> dead AGM pack below
> 10V per module and light up the red LED on all the
> regs at once(The
> reflection of 17 red LED's out of the corner of my
> eye, a thought it was
> police lights, LOL). Put that is still in the 240V
> range.
>
>
--- End Message ---
--- Begin Message ---
Rod- what would you see as the min DCV for the 220 V 55A Iota?
M
On Nov 4, 2006, at 5:13 PM, Rod Hower wrote:
Jeff,
The 240Vac input IOTA rectifies and makes that input
340Vdc nominal. It then bucks that voltage down to
13.8Vdc. Most if not all 240Vac controls including 3
phase inverter controls (which I designed at Baldor
Motors and Drives) have a +/- 10% minimum range for
the input voltage. Therefore a 268Vac input is 379Vdc
max and my inverter controls shut down on overvoltage
at 380Vdc. I think this is a safe range for the IOTA.
The bus caps are probably rated for 400Vdc and 450Vdc
surge. The IGBT's are rated for 600Vdc. I don't
think 354Vdc is anything to worry about.
Rod
Designing 28Vdc inverters these days.
FYI. I'm currently working on a 24Vdc 3 phase inverter
for BLDC that is 0.7" in diameter. Talk about putting
10 pounds of *hit in a 1 lb. bag!!
--- Jeff Shanab <[EMAIL PROTECTED]> wrote:
Sorry, I have a 288V nominal pack but here is one
data point for the record.
My end of charge voltage is 354 Volts. I have been
worried that that
may take the iota out.
The IOTA is connected to the pack with only a fuse,
It stays charged and
it goes to sleep nicely.
I have readlighted* a few times and it doesn't
hickup.
*redlighted in this context is when I suck the 1/2
dead AGM pack below
10V per module and light up the red LED on all the
regs at once(The
reflection of 17 red LED's out of the corner of my
eye, a thought it was
police lights, LOL). Put that is still in the 240V
range.
--- End Message ---
--- Begin Message ---
The CE dragster went to the qt.finals today taking out the track and summit
champ,Mike Z.This was a buyback 1st round race and he bought back into the 2nd
round only to take me out in the 4th round.Still the $400. will come in handy.
Dennis Berube Next sat.A Wally trophy up for grabs.
--- End Message ---
--- Begin Message ---
Lee,
What would be needed to use your balancer on a pack of 26 batteries?
I tried to figure it out from the schematics and I can see that
there is a limit of 15 currently.
I also read that Al Godfrey had used a "bank-switching" technique
to use it on 26 YT. Do you know what he needed to change in the SW
to make the bank-switch compare all 26 (27) batteries?
(Do you have the sample source code, I could not find it on the
Balancerland website)
I like the idea of using the DC/DC also to charge the aux battery,
avoiding to have 2 DC/DC converters.
Only drawback is that during driving the voltage is lower than
normal in a car with alternator cranking the voltage up to 14V
or somewhere close.
Also, since the truck has so many aux loads (vacuum pump, power
steering pump and then the normal fans and pumps and electronics)
I think about installing a high current DC/DC, which means that
the relays would need to be higher current rating and cannot be
PCB mounted.
Maybe I should stick with a Vicor 300V module and have a separate
DC/DC for the aux battery.
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 Lee Hart
Sent: Saturday, November 04, 2006 12:05 PM
To: ev@listproc.sjsu.edu
Subject: Re: Lee your BMS anyone else interested?
Don Blazer <[EMAIL PROTECTED]> wrote:
> I was looking at your BMS with NiMH batteries in mind now. You might
> not remember but a while back I was trying to get you to do a bulk
> order for use on AGM batteries. I would still be willing to buy in
> quantity for the best prices and sell at the quantity price break
> to all who is interested in making one.
Don, you may have meant to send this to me directly, but since it went
to the EV list, I'll answer it here.
Yes, I have been working on a new version of my Battery Balancer for
over a year now. Unfortunately, it is a home project, and so has a low
priority compared to "real work" and the "honey do" list.
At present, I have the new Relay board design done, and can order them
any time. I have the new Control board about 75% done. But there are
some loose ends that I can't help but keep tinkering with.
Specifically, the old control board used an external digital multimeter
with isolated RS-232 output to measure battery voltage and current. They
cost $50-$100. I'm experimenting with a cheaper on-board circuit to do
this (since everyone is obsessed with cost).
I am really at the point where I need to just "shoot the engineer, and
ship it!" I have several people who have been very patient, and just
need to go ahead and order the boards for them, saving any further
improvements for the next go-around.
--
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 ---
