EV Digest 6063
Topics covered in this issue include:
1) RE: dc-dc working
by Cor van de Water <[EMAIL PROTECTED]>
2) RE: I'm finally getting messages again
by Cor van de Water <[EMAIL PROTECTED]>
3) Re: Dual motors with "hopped up Curtis" - seires or parallel?
by Cory Cross <[EMAIL PROTECTED]>
4) Re: EV controllers? the 4th option...
by "Rich Rudman" <[EMAIL PROTECTED]>
5) Re: Dual motors with "hopped up Curtis" - series or parallel?
by Steve Condie <[EMAIL PROTECTED]>
6) Re: Dual motors with "hopped up Curtis" - series or parallel?
by Cory Cross <[EMAIL PROTECTED]>
7) Re: Dual motors with "hopped up Curtis" - seires or parallel?
by mike golub <[EMAIL PROTECTED]>
8) Precharge Circuit Design
by Chris Jones <[EMAIL PROTECTED]>
9) Re: Splined Adapters, Was: Inexpensive DC Motors with Keyed Shafts
by Eric Poulsen <[EMAIL PROTECTED]>
10) Re: EV controllers? the 4th option...
by "Rich Rudman" <[EMAIL PROTECTED]>
11) Re: EV controllers? the 4th option...
by "Rich Rudman" <[EMAIL PROTECTED]>
12) Re: EV controllers? the 4th option...
by "Rich Rudman" <[EMAIL PROTECTED]>
13) Re: EV controllers? the 4th option...
by "Rich Rudman" <[EMAIL PROTECTED]>
--- Begin Message ---
David,
I know.
The reason is that if you draw any significant current, the
pre-charge does not happen, the resistor will burn up and
the controller will fault (or, if not protected, the main
contactor will close with a BANG of shorting 3000 Amps onto
the controller's caps, see the other discussion post.
Therefor I asked what the current draw is of an *unloaded*
Iota. If that is much smaller than your pre-charge current
then it could be done to give it a pre-charge.
I think Lee suggested the same.
It is _required_ to put a 12V relay in the DC/DC output to
make sure that the precharge happens without load on the
DC/DC converter. Only after the main contactor has closed
and the current can be drawn directly from the pack can you
close the 12V relay on the DC/DC output and start charging
your aux battery and supply current to your car's 12V system.
So if the Iota draws less than 100 mA from the pack when idle
and your pre-charge uses a 10 Ohm resistor, which will start
the pre-charge at 15A from a 144V pack and slowly taper off
when the capacitors will become charged, the effect of the
Iota's 100 mA on the 10 Ohm pre-charge resistor will be that
the finishing voltage will be 1V lower than pack, but long
before the finishing voltage is reached will the main
contactor close, so the difference will even be smaller
than 1V.
(100 mA at 144V is almost 15 Watts, this should be sufficient
to run even a not-so-efficient DC/DC when idling.)
Hope this clarifies,
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 David Brandt
Sent: Wednesday, October 25, 2006 4:07 AM
To: [email protected]
Subject: Re: dc-dc working
The zilla manual specifically requires nothing to be placed after that
circuit.
David Brandt
----- Original Message ----
From: Cor van de Water <[EMAIL PROTECTED]>
To: [email protected]
Sent: Monday, October 23, 2006 4:52:11 PM
Subject: RE: dc-dc working
If your problem is inrush then a parallel resistor
does not help - a series resistor maybe.
Altenatively - what is the unloaded draw of the Iota?
You may be able to direct connect it to the battery input
of the controller, so you are after the precharge circuit,
this should take care of slowly ramping voltage, avoiding
the inrush, but you will need a high-current 12V relay
in the output to avoid any significant current draw while
the precharge circuit is trying to get the capacitors
charged up.
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 David Brandt
Sent: Monday, October 23, 2006 8:47 AM
To: EVDLposts
Subject: dc-dc working
Well, I discovered something. A KLKD fuse can blow without any visible
indications. And the inrush current for an Iota 240V DC-DC will blow a 5A
KLKD fuse instantly. I had to short the fuseholder with a test clip
(serving as a fusible link) before inserting a fuse. Then removed the clip
and all was well.
Any recommendations with avoiding this problem in the future? Maybe a
resistor could be added across the fuse?
David Brandt
--- End Message ---
--- Begin Message ---
Jim,
You're so welcome.
Actually this whole situation was very fresh in my
memory as my caompany had a glitch in their email
system last week, which caused the SJSU list server
to think that the messages could not be delivered
(even while I got several of them anyway) but I could
tell that there was a period of time that I had not
received any external email, so I knew that our
server was not working correctly.
What followed was exactly what David Roden described:
- bounce several emails of the SJSU list server and
it will stop sending you emails for a while, in
my case I have exactly 1 day (24 hours) hole in
received messages.
- if the messages are accepted after a day, you will
see the responses on emails that you have not yet
received because they are still buffered until:
- after 2 days (48 hours) the bounced and all other
queued messages will be delivered, so you will see
the original emails to which you have already seen
the replies. You can verify this by:
- look in the email for the "sent" date. This date
will be 2 days earlier than the "received" date
as indicated by your email program.
This is also how I figured this behavior out myself,
as I could see an original email come in 2 days late
and a reply that was sent within minutes of the
original email, so I knew that the 2 days delay was
not for everyone, but just for my private queue at
the SJSU server, which was faithfully buffering all
my EVDL messages until my email would be up and
running again.
Amazing that this server can maintain individual
buffer queues for each of the hundreds? thousands?
of subscribers.
(I know, it's really just pointers, it is not
exploding all email messages themselves to everyone
but still it is quite a piece of administration.)
Note: this also explains the occasional late email
that was sent but only 2 days later it suddenly shows up
as the glitch was too short to cause no messages to
be sent for a day, but the single unsent email is
still buffered for two days, to show up well after
the expected moment of delivery.
Maybe the server can be configured to start sending out
all email as soon as it finds that email is accepted again.
It is a bit confusing to have the replies a day before
the original post is received, because the buffer has to
wait another day when the stop-send is lifted.
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 Jim Husted
Sent: Wednesday, October 25, 2006 6:07 AM
To: EVDL
Subject: I'm finally getting messages again
Hey all
Looks like the logjam is breaking up and I got some
messages rolling in here again. Thanks to all that
offered help but it looks like it was some kind of
glitch. Cor wrote me and said wait 2 days and they'll
come through and well he was right (thanks Cor)
They are not yet caught up but I'm getting them in at
least. Good thing being I was starting to detox from
lack of EVDL injections, LMAO!
Cya
Jim Husted
Hi-Torque Electric
__________________________________________________
Do You Yahoo!?
Tired of spam? Yahoo! Mail has the best spam protection around
http://mail.yahoo.com
--- End Message ---
--- Begin Message ---
Hi Steve,
If you put them in series and a wheel spins, you'll get no torque out of
either. Why not one controller/motor and buddy-pair the batteries to
72V? As a bonus, your system will be rather immune to failure of any
single component.
Cory Cross
Steve Condie wrote:
>I asked this before, and didn't get any takers, but snce it fits into the "4th
>option" thread discussing "Curtis killer" controllers I'll ask again. The
>situation: I've got a Tropica which should weight about 2000 - 2400 pounds
>when finished, with dual 6.7 inch ADCs - one per rear wheel, belt driven, and
>designed to run at 72 volts. GThe question is this: With a 144 volt
>Logisystems "hopped up Curtis" 1209, should I run the motors in series or
>parallel most of the time?
>
>Roger Stockton said:
>
>I would stay far, far away from the hopped up Curtii until they have
>been proven in *on-road* applications. The Curtis has a particularly
>poor thermal design, so I would expect these hopped up controllers to
>offer little if any benefit in the real world since the controller will
>go into thermal cutback to protect itself when it gets hot. There also
>isn't a huge amount of space in the Curtis case to add much more bus
>cap, which really leaves only the option of replacing the stock caps
>with better quality units (lower ESR, higher temp, etc.) A hopped up
>Curtis may do just fine in a lightweight vehicle like a golf cart,
>where it is unlikely to see peak amps for more than a second or two at a
>time, but life in a 3000-4000lb+ on-road EV is much more demanding.
>
> (By the way, Jim at Logisystems discourages the use of their "hopped up
> Curtii" in heavier vehicles as well.)
>
>Steve Clunn said: Keep the thing cool , alot of time my old curtis would get
>so hot you couldn't put your hand on it . Don't drive in you hi a gear , steep
>hills use 1st but still keep you rpm's up by driving somewhat fast . With a
>120v system you can have your foot to the floor and be crusing in 2nd without
>blowing up the motor , the controller is full on and is not working hard , if
>your cursing in 4th the controller is working harder as the motor voltage will
>be lower and the motor amparge will be much higher ( to get some power ) this
>means more amps flowing through the controller and so more heat, wear and
>tear, motor is also working harder.
>
>I don't have a transmission, but I have the option of wiring the motors in
>series or parallel (or switching between the two.) figure with the motors in
>series, all else being equal, the controller will be dishing out half the amps
>and twice the volts as it will at the same speed with the motors in parallel.
>That should make the controller cooler and happier, no? So what's the
>unintended consequence I'm overlooking?
>
>
>---------------------------------
>How low will we go? Check out Yahoo! Messenger’s low PC-to-Phone call rates.
>
>
>
--- End Message ---
--- Begin Message ---
All good ideas
But..
Simple assembly, wide operating windows,
NO plumbing..
Are very serious features.
Serious enough to over shadow the draw backs.
And.. I succeded in DElurking Otmar...
Mission accomplished.
----- Original Message -----
From: "Nick Viera" <[EMAIL PROTECTED]>
To: <[email protected]>
Sent: Friday, October 20, 2006 8:14 PM
Subject: Re: EV controllers? the 4th option...
> Hi Rich and all,
>
> Rich Rudman wrote:
> > The specs list something like this: 96 to 450 volts. DC 550 to 1100
> > amps IGBT based.
>
> I'd be more interested in seeing controllers built as several
> different models for several distinct voltage ranges. For example, build
> a different model for 60-160V nominal, 160-260V nominal, and 260-360V
> nominal. Note that the min/max voltage ranges would overlap so any pack
> should be fully compatible with at least one model. This would (at least
> theoretically) allow the controllers to be less expensive and more
> efficient because they can be better tuned for the voltage and current
> ranges involved.
>
> > Precahrge and contactor control. High speed Real time Rev limter.
> > Digi pot or Dip switch selectable RPMs. Battery current and Motor
> > current adjustable...With a 25 turn pot.. No
> > programming head or computer needed.
>
> Of course... and I'd argue that these features should be mandatory
> for any EV motor controller, period. However, using potentiometers for
> these voltage/current/RPM limiter adjustments is something I'm concerned
> about. I think these settings need to be resistant to adjustment error
> that could be caused by wear, vibration, temperature, curious fingers,
> etc... and software control is certainly far less prone to such problems.
>
> Yes, getting a computer set up to modify settings can be a pain, but
> in this case I think the benefits of software control really out-weight
> any such hassles. Also, the settings should be as accurate as possible,
> at least with respect to the battery-side settings.
>
> > AIR COOLED! The heat sink from a PFC50 comes to mind, It's about %40
> > better than the old Raptor Wakefield heat sink. Of course water
> > cooling is doable...for a price. But lets not go there right yet.
>
> To put it nicely, I'm really not impressed with air-cooled EV gear.
> Maybe its because I've driven my EV through the south Texas summers, but
> I just don't feel that air-cooled electronics are a good solution for
> EVs. Reliability in hot climates being one important issue, size being
> another. Along the same lines, I believe EV electronics need to be
> properly protected from the automotive environment... meaning they
> should be at least splash proof from all sides. Many air cooled products
> lack such protection, and IMO this isn't a great design habit for this
> application.
>
> > PB6 throttle pot needed. No fancy throttle boxes or hand wound
> > sensors needed. Fault protected High and low pot.
>
> This of course makes good sense as it allows for easy replacement of
> the throttle pot in the future (no need to build a custom replacement).
>
> My 0.333 kWh,
>
> --
> -Nick
> 1988 Jeep Cherokee 4x4 EV
> http://Go.DriveEV.com/
> http://www.ACEAA.org/
> --------------------------
>
--- End Message ---
--- Begin Message ---
How often do you spin your wheels? My thinking is to be able to switch to
parallel for spin-inducing conditions (Ice? Stuck in mud? Dude - I live in a
California suburb!), or to allow me to exceed the designed top speed of 60 mph
at 72 volts. (See California comment above. 70+mph is a fact of life here.)
But I'm still concerned about possible adverse impact on the maxed out
controller circuitry.
Hi Steve,
If you put them in series and a wheel spins, you'll get no torque out of
either. Why not one controller/motor and buddy-pair the batteries to
72V? As a bonus, your system will be rather immune to failure of any
single component.
Cory Cross
Steve Condie wrote:
>I asked this before, and didn't get any takers, but snce it fits into the "4th
option" thread discussing "Curtis killer" controllers I'll ask again. The
situation: I've got a Tropica which should weight about 2000 - 2400 pounds when
finished, with dual 6.7 inch ADCs - one per rear wheel, belt driven, and
designed to run at 72 volts. GThe question is this: With a 144 volt
Logisystems "hopped up Curtis" 1209, should I run the motors in series or
parallel most of the time?
---------------------------------
Stay in the know. Pulse on the new Yahoo.com. Check it out.
--- End Message ---
--- Begin Message ---
Steve Condie wrote:
>How often do you spin your wheels?
>
Hmm, I have RWD in the snow -- so often.
> My thinking is to be able to switch to parallel for spin-inducing conditions
> (Ice? Stuck in mud? Dude - I live in a California suburb!), or to allow me
> to exceed the designed top speed of 60 mph at 72 volts. (See California
> comment above. 70+mph is a fact of life here.) But I'm still concerned
> about possible adverse impact on the maxed out controller circuitry.
>
>
>
Why not add in some field weakening? You only need to increase your
speed 15% or so.
Cory Cross
>Hi Steve,
> If you put them in series and a wheel spins, you'll get no torque out of
> either. Why not one controller/motor and buddy-pair the batteries to
> 72V? As a bonus, your system will be rather immune to failure of any
> single component.
>
> Cory Cross
>
> Steve Condie wrote:
>
> >I asked this before, and didn't get any takers, but snce it fits into the
> >"4th
> option" thread discussing "Curtis killer" controllers I'll ask again. The
> situation: I've got a Tropica which should weight about 2000 - 2400 pounds
> when
> finished, with dual 6.7 inch ADCs - one per rear wheel, belt driven, and
> designed to run at 72 volts. GThe question is this: With a 144 volt
> Logisystems "hopped up Curtis" 1209, should I run the motors in series or
> parallel most of the time?
>
>---------------------------------
>Stay in the know. Pulse on the new Yahoo.com. Check it out.
>
>
>
--- End Message ---
--- Begin Message ---
I thought Logisystems has a controller that can
control two motors, and I'm surprised you didn't try
that method?
--- Steve Condie <[EMAIL PROTECTED]> wrote:
> I asked this before, and didn't get any takers, but
> snce it fits into the "4th option" thread discussing
> "Curtis killer" controllers I'll ask again. The
> situation: I've got a Tropica which should weight
> about 2000 - 2400 pounds when finished, with dual
> 6.7 inch ADCs - one per rear wheel, belt driven, and
> designed to run at 72 volts. GThe question is this:
> With a 144 volt Logisystems "hopped up Curtis"
> 1209, should I run the motors in series or parallel
> most of the time?
>
> Roger Stockton said:
>
> I would stay far, far away from the hopped up Curtii
> until they have
> been proven in *on-road* applications. The Curtis
> has a particularly
> poor thermal design, so I would expect these hopped
> up controllers to
> offer little if any benefit in the real world since
> the controller will
> go into thermal cutback to protect itself when it
> gets hot. There also
> isn't a huge amount of space in the Curtis case to
> add much more bus
> cap, which really leaves only the option of
> replacing the stock caps
> with better quality units (lower ESR, higher temp,
> etc.) A hopped up
> Curtis may do just fine in a lightweight vehicle
> like a golf cart,
> where it is unlikely to see peak amps for more than
> a second or two at a
> time, but life in a 3000-4000lb+ on-road EV is much
> more demanding.
>
> (By the way, Jim at Logisystems discourages the use
> of their "hopped up Curtii" in heavier vehicles as
> well.)
>
> Steve Clunn said: Keep the thing cool , alot of
> time my old curtis would get so hot you couldn't put
> your hand on it . Don't drive in you hi a gear ,
> steep hills use 1st but still keep you rpm's up by
> driving somewhat fast . With a 120v system you can
> have your foot to the floor and be crusing in 2nd
> without blowing up the motor , the controller is
> full on and is not working hard , if your cursing in
> 4th the controller is working harder as the motor
> voltage will be lower and the motor amparge will be
> much higher ( to get some power ) this means more
> amps flowing through the controller and so more
> heat, wear and tear, motor is also working harder.
>
> I don't have a transmission, but I have the option
> of wiring the motors in series or parallel (or
> switching between the two.) figure with the motors
> in series, all else being equal, the controller will
> be dishing out half the amps and twice the volts as
> it will at the same speed with the motors in
> parallel. That should make the controller cooler
> and happier, no? So what's the unintended
> consequence I'm overlooking?
>
>
> ---------------------------------
> How low will we go? Check out Yahoo! MessengerÂ’s low
> PC-to-Phone call rates.
>
>
__________________________________________________
Do You Yahoo!?
Tired of spam? Yahoo! Mail has the best spam protection around
http://mail.yahoo.com
--- End Message ---
--- Begin Message ---
Thank you everyone for the precharge circuit
discussion. I want to add a precharge circuit to my
conversion to help minimize contactor welding and
controller filter capacitor failure. But I have a few
design questions:
Regarding the resistor choice:
I have a Curtis 1231C-8601. The manual says 750 ohms
at only 25W, but fresh off a charge I too am at it's
peak input of 175V, so that's 233 mA and 41W. Since
25W is the continuous rating and this is a very narrow
and occasional occasional pulse, 25W would probably
do. But the Reliability Analysis Center (RAC)
Electrical Derating for Optimum Performance handbook
suggests 50% power derating (and 70% for voltage) for
power wire wound resistors. So something like the
Ohmite WHF90L750J 90W, 750 ohm resistor for $21 from
Allied Electronics seems like a better match.
Regarding the relay choice:
As suggested for efficiency, I would like to power a
12V coil relay in series with this resistor from the
ignition path, downstream of my AC interlock relay but
upstream of my potbox microswitch which powers the
main contactor and KSI relay so it goes on with the
key when the charger isn't powered up. Of course
selecting an automotive relay with contacts rated at a
high enough voltage and DC current is important.
Considering the inductance of these high power
resistors, I wonder what voltage this relay would see?
If this is an issue, I have seen non-inductive power
resistors that might go up this far wattagewise but I
haven't looked in to them yet. The relays I have
bought from Kragen don't come with specs -- any
suggestions on a relay to use? And similar question
for the KSI relay -- I just put in an automotive
relay, but I don't know it's contact rating, so at
175V I might be overstressing it voltagewise. I think
the KSI circuit has less inductance than a wire wound
precharge resistor will have.
And what about a fuse?
Shouldn't we insert a fuse in series with the
precharge resistor and relay? I'd hate to see that
resistor burn if it developed a short, even though the
RAC estimates only a 4% chance that shorting would be
its failure mode. Current rating for my case would be
~.5A, and I think it won't see the voltage spike so
they typical 250V automotive type would suffice.
Thank you,
Chris Jones
--- End Message ---
--- Begin Message ---
Bruce wrote:
Thanks Eric! The splined adapters from auto parts stores sound like a good
way to go. Can you say what they are used for in ICE cars? How would
someone find out what sizes and spline numbers are available?
The one that Mr. Stockton found is a G&G Manufacturing "Weldahub." Near
as I can tell, it's for PTO attachments on tractors and other large
equipment. The idea of the weldahub is that you get the right size hub
for your splined shaft, and then you weld an appropriate sprocket onto
it, and use it with roller chain.
Thing is, there's no reason you can't use the weldahub as an adapter to
other things. You can turn it down on a lathe to ensure its
concentricity, then use it in your motor-to-transmission adapter. You
could even have a keyway cut into it.
http://www.ggmfg.com/product_subcategories.asp?cat=weldahubs
Which Surplus Center motor did you get? Was it their part # 10-1422? How
did it work out?
No, I went with the other option, the 10-1422. It's a 7" diameter
series-wound motor. 48V, 5.6HP. The shaft is a 1.25" 14T splined job.
It's dirty, and covered with thick gray paint, but it seems to be in
pretty good shape from what I can judge, and what Mr. Husted tells me.
It's heavy, > 80 lbs, probably closer to 100. The end housings are cast
iron, which accounts for quite a bit of weight. The ADC motors have Al
ends.
Downside is that this motor is _definitely_ used, despite the website
claiming it to be new. I originally paid $400, which included
shipping. It arrived poorly packed and with a smashed armature terminal
and bent hanging bracket. I contacted Surplus Center about it's packing
and state, and suggested a $200 refund. So, in the end I got the motor
for $200. There is very light rust pretty much everywhere but the
shaft. I'm presently cleaning it up, and it's looking better all the
time, now that I can see the shiny metal under all that paint. Biggest
drawback is the large 11" diameter faceplate. This can be cut down on a
lathe, but then you'd need to drill NEMA mounting holes, and have the
front centering disk reduced. Even though it's diameter is 7", and the
ADC 6.7" has a 2" diameter centering disk, this is probably too small
considering the 1.25" shaft. Heck, the front roller bearing is > 2"
diameter.
http://www.surpluscenter.com/item.asp?UID=2006102515051215&catname=electric&qty=1&item=10-1422
----- Original Message -----
From: "Eric Poulsen"
... Spline adapters / shafts are available...
I picked up a 14T splined motor from Surplus Center. One of the other
list members (Roger Stockton, in fact. Thanks again!) here found a
splined adapter at a local (to him) auto parts store, and sent it to me.
--- End Message ---
--- Begin Message ---
Line for line folks...
----- Original Message -----
From: "Roger Stockton" <[EMAIL PROTECTED]>
To: <[email protected]>
Sent: Saturday, October 21, 2006 12:55 AM
Subject: RE: EV controllers? the 4th option...
> Hey Rich,
>
> > What do we need in the way of Controllers and what
> > would you folks pay for a new design?
>
> What DC'ers need is a Curtis-killer, not something that tries to compete
> with Otmar's already excellent offerings.
This is exactly what I am proposing... High voltage Curtii Killer, with some
goodies, Not many.
>
> 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.
As the Guy who thought up the Raptors , Exactly what I had in mind. Stay
under 1000 amps, or real close anyways.
Most of US need 500 amps , Want 2000. Need pays the bills Want.. Well Even
I had to TRADE to get a Z2K
>
> The Curtis maxes out at 144V and 500A; the Raptor went to 156V; it would
> be nice to perhaps push this upwards a bit nearer the 200V mark, but not
> if it causes a serious price hit. The Raptor had battery current
> adjustable to 600A and a non-adjustable motor loop current of 900A max.
> Making the motor current adjustable as well would be nice, but probably
> not essential. Forget 72V or less since AllTrax covers that market
> nicely already. 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.
>
Keep in mind Altrax is The other side of What DCP was.. I am the other
half. Damon has the less than 72 volts MOSFET world pretty much sown up.
I would be a fool to go there.
Nope Not gonna do that!
I am proposing a IGBT module based machine. Ot says it can't be done. I say
the reduction in labor Takes his cake rather handily.
I just told all you wanna be hardware hackers exactly what to do, and close
to how to do it.
The voltage limits I think are a waste of time. Build for the high volts,
and Well at the lower voltages the losses could be less, But, That's not the
point.
The point is a easy to buld easy to service, simple power stage simple
analog controller that just gets the job done.
You want software? Fine You write the code and support it.
By the time most folks get the code solid enough for shipping, I will be
shipping about the 250the unit.
> My essential spec/feature list for a Curtis-Killer is:
>
> - 600 adjustable battery amps and better current multiplication than the
> Curtis.
The Curtii just use diodes, No freewheel. Kinda basic.
I am talking 550 amps per module.. and a VSat and Average mode combined
control loop.
The Curtii and the Raptors used JUST a Vsat controller. VERY simple, But
can be fooled. It worked for a LOT of folks and still is working for most of
us.
The average current controller is a simple Power supply Buck controller..
Just might be the same chipset that runs about 315 PFC chargers to date.
>
> - handles at least the range of 96-144V; 156V is almost essential, and
> up to ~200V would be even better, but stay under the next quantum leap
> in cost that occurs somewhere north of 144V.
In IGBTs there is no Quantum leap. The leap is from MOSFET to IGBTs. That
will be my entry ticket.
I don't do Fets anymore.
The market sector that is Symbiotic to High voltage chargers is ABOVE 144.
There are reasons for my Madness.....
>
> - 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, but the essential
> thing is that the controller comes with the required heatsink (and fans,
> if required); you can't have one without the other)
>
Yes campers.. the heatsink will be in the same place as on a Raptor/Rex.
Same idea, Better heatsink. Maybe a Copper thermal spreader.. and Really
nasty fans. The nastier fas the, less need for
Water cooled. And the better hot weather performance. Yea I know about
that. Still The Raptors did pretty well in Hot. Better than Curtii by about
10x. Funny what real finned heatsinks can do.
Don't sweat that, Thier will be a thermal performance curve. As all units
do. Even Otmar can't run at 2000 amps with water cooling indefinatley. Not
as his published water flow numbers. At 10x, Yea maybe some cold water Say
40 Deg F. Then he might scrub full power losses for 60 seconds.
I am NOT making an Zilla Killer, I am making a Dumb as a Box of Rocks high
voltage controller to fill the Gaps from Curtii to Z1K.
> - packaged to survive the automotive underhood environment. The
> electronics need to be sealed decently in a package affixed to one side
> of the heatsink; none of this uncoated/barely coated PCB in a ventilated
> box with wet and dirty air being blown through for cooling. And, try to
> avoid RTVing stuff up to the point of making it esentially
> non-servicable (Raptor power stage) unless you're convinced it is so
> bulletproof that you're willing to just replace failed units as a
> complete assembly should the need ever arise.
>
Umm request noted.
I still have a running 8 year old unprotected Raptor....
The Raptor power stage was designed to toss away. It was a real effort just
to make one. And that I have done a few times.
Cute But... non serviceable. And a process nightmare to assemble. 40 lbs
of 550 Deg F Copper plate, tooks guts and skill.
All Ot's tuff if Conformally coated, But he still does not make under hood
spec, What that's IPC56 or something like that.
No I am not making a submersable product.
> - priced on par with 1231C + appropriate heatsink. At $2k a Z1K LV is
> only about $500 more than the 1231; if you can't deliver something for
> at least a couple hundred less than a Z1k LV, the only time you'll sell
> any is when Otmar gets backlogged.
Your pricing is not correct. If Ot can't do Z1Ks,, then the market is Wide
Open Right?
First one with product gets to set the value.
I am not going try to meet any spec price number.
Not for a High voltage capable controller in a VACUME tight market.
Shouldn't I ask what I want and let the world Scream and yell then pay the
bill.
That's How MicroSoft does it..right??
>
> A couple nice-to-haves include:
>
> - contactor failure detection
>
> - controller failure detection & auto contactor dropout
>
> - ability to set a low voltage limit and have the controller reduce
> power to keep the pack from remaining below this limit for more than
> some small (or adjustable) number of seconds. Or have the controller
> accept an undervoltage signal from an external BMS, such that it will
> reduce the current limit as required to keep the undervoltage signal
> from remaining 'on' for more than the set amount of time.
IF I build a charger, COUNT on the REGbuss BMS and MK2C and MK3 having a
comm path to the charger.
That right there should put the Fear into Otmar's product line.
I don't have to do any voltage det if the Regs are there to keep the
batteries safe.
OEMers will or should KILL for this concept. Integrated BMS support is now
the entry fee for system safe controllers of all kinds.
>
> Your spec list reads like you might want to avoid a micro in this
> product entirely; is that the case, or do you simply not want to support
> comms between the controller and a PC/Palm/etc.?
>
The Cute and quick design specfically does not use a Micro. BUT.. thier
might a LOT of spare room on the
Back edge of the PCB board.. Lots of room for stuff that make be needed.
> If you absolutely don't want the headache/dev effort of supporting
> comms, at least have the decency to provide thumbwheels or indexed
> rotary switches, etc. for positive setting of current limit(s) and RPM
> limits. Neither has to resolve down to 10A or 10RPM, etc.; something
> like 8 50A steps of adjustment for current (allows 200-600A in 50A
> steps), and 8 250 (4500-6500) or 500RPM (2500-6500) steps for the rev
> limit is fine and there is then immediate visible means for the user to
> confirm the desired setting.
Clearly NOT a chance. All the stuff is a bugger to keep clean and water
resistant.
That's Zilla territory, You get 25 turn pots that are behind lexan or
Aluminum plates. Just to keep water out, Simple More Amps/less amps.
You are clearly thinking about digital controls. Curtii have pots behind a
screw in plug. THAT what I am talking about.
mayabe RPM is a dip switch... But... I have taken hits for having those
exposed on the chargers.
>
> >From a cost point of view, it is probably cheaper to stick a $2 LCD on
> the unit and one or two buttons to allow navigation for setting anything
> you want to configure, assuming you have a micro inside the controller
> anyway. This would also allow you to seal up the electronics more
> effectively.
Yer Joking Right???
I said simple controller... NO micro.
>
> Not sure why you are opposed to going water-cooled. If the
> cost/challenge of boring passages in a block of aluminum is an issue, it
> ought to be feasible to simply mount the power devices to one side of an
> aluminum plate perhaps 3/8" thick and then bolt a plastic (or aluminum)
> plate/block to the other side that has the cooling channels milled into
> its surface. A bit of Permatex or a paper gasket to seal the two and
> you're done. It can't be cheap to source a sufficiently massive chunk
> of air-cooled heatsink and then add the cost of fans of sufficient
> quality to last a reasonable time. It wouldn't be cheap, but if you
> also provided water cooling of the bus caps you'd dramatically increase
> their longevity and you might just be able to push them enough harder to
> use fewer (or cheaper) caps and come out close to neutral on cost. (I've
> got an old Curtis controller on the shelf with heatsunk caps...)
>
Oh God.. here we go... HEY PUMPS AND FANS AND RADIATORS are a real pain to
make support and install.
You have not done so resentley have you??? I have.. All the cooling mumbo
Jumbo has to be done one way or the other...
Doing it once and right at the point of use is By far the cheapest and
cleanest way to solve the problem. Water cooling just moves that issue
away from the
actual power stage.
Again...I said cheap and simple.
Water cooling is easy.. for a while... But the point is to NOT use it unless
you have to.
Ease of install in the primary reason for aircooled. Everybody said we
couldn't do it on the Raptors. We did. It worked.
For all but the really hot folks.
> I think you're missing the boat with encouraging people to use a PB6.
> The pot is junk, and at $75 a pop there should be a good revenue
> opportunity here by offering your own competitively-priced version
> based on a quality pot and intended to survive the automotive underhood
> environment.
Yup but I don't care where you get a 10k pot.
A PB6 is cheap and they are every where. Multi sourced, and well.
What else matters?
Oh you mean I should hand make somthing out of PVC pipe and old Phone wire??
Been there done that.
Or do you mean a $500 throttle pot from a quality ICE doner rig?
that might meet spec..
Hey the EV world uses PB6 pot boxes... I am taking the easy way our here.
Heck Otmar does.
>
> >From your comments on cost, etc., it sounds like maybe you ought to
> stick with the chargers instead of spreading yourself too thin with yet
> another product to support. I don't think there is a market for another
> $2k+ controller unless it fills some need that the Zillas don't. The
> only one I can think of is the battery range from 360-450VDC, and that's
> not going to be a controller with significant demand, nor will it be
> cheap enough for anyone to care if it covers pack voltages below that
> range. Another possible 'hole' in the Zilla lineup is something over
> 1000A but below 2000A; not sure there'd be any real demand for a
> "Z1.5K", nor if it would really be possible to offer such a beast at a
> sufficiently lower price to steal customers away from the Z2K.
>
Clearly you don't see the demand that I do for the reasons I do. But that's
OK Roger...
The need here is: A way to get the job done, and not take 2.5 man years to
do it.
And yes I have all the work I need, I can't get chargers out of here, let
alone a complete controller product line build up.
The market hole is below the Z1K.
If I do this right I don't take any of Ot's market.
He just gets to know I am there.
By the way listers this Thread DElurked Otmar in about 45 minutes, We have
had a long talk about many things.
So.. my confidence level in Otmar is a LOT better than it was when I first
started this thread.
> Cheers,
>
> Roger.
>
Off to make Green boxes and Plug in Hybrids.
Got one in the new shop. It runs!!!
White Bird is in Kingston right now.
Madman
--- End Message ---
--- Begin Message ---
I do remember what I paid for it.
And your comments are worth aLOT more than that.
As I surmised.. there is a need for simplicity in controllers.
Not more features just solid ones.
And if there is a Open sourced thread for doing DC controllers..
I have just outlined a heck of a lot of hard won information.
Madman
----- Original Message -----
From: "Paul G." <[EMAIL PROTECTED]>
To: <[email protected]>
Sent: Saturday, October 21, 2006 8:50 AM
Subject: Re: EV controllers? the 4th option...
> Wow, this thread sure has gotten a lot of attention!
>
> On Oct 20, 2006, at 9:15 AM, Rich Rudman wrote:
>
> > The specs list something like this:
> >
> > 96 to 450 volts. DC
> > 550 to 1100 amps IGBT based.
>
> How many models are you talking about? Reading ahead it looks like a
> low current and a high current model, but are you talking about one
> model handling the full voltage range?
> >
> > Precahrge and contactor control.
>
> Good ideas.
> >
> > Battery current and Motor current adjustable...With a 25 turn pot.. No
> > programming head or computer needed.
>
> I like this. Would there be any motor voltage limit control? The upper
> end of the voltages you are proposing are a bit high for most of the
> motors in use.
> >
> > AIR COOLED! The heat sink from a PFC50 comes to mind, It's about %40
> > better
> > than the old Raptor Wakefield heat sink.
> > Of course water cooling is doable...for a price. But lets not go there
> > right
> > yet.
>
> I like air cooled!
> >
> > High speed Real time Rev limter. Digi pot or Dip switch selectable
> > RPMs.
>
> I'm not sure of the need, as after market units are available that work
> fine with the throttle pot you plan to use (reading ahead again :-)
> Just make sure that 0 ohms and/or infinite ohms result in no output. Of
> course, if the option costs you very little to include...
> >
> > PB6 throttle pot needed. No fancy throttle boxes or hand wound sensors
> > needed. Fault protected High and low pot.
> >
> > Case size the 500 in a PFC30 case and Bigger one.. in a maybe a PFC50
> > case
> > But that's way too big.
> > Fine I will design a case that works the best.
>
> Just a bit of what runs through my head...
>
> It seems that high voltage controllers tend to have quite a price
> premium, I'm not sure of all the why's, but I am aware that the 144v or
> 156v limits of most low voltage controllers have been putting lids on
> people. Perhaps something with 250 volt parts instead of 200 volt parts
> so 192 volts would be the upper limit (remember Auburn.) You made a
> proposal though, so I bet you have something in mind to deal with the
> cost.
>
> The amp ranges you tossed out seem to be close copies of Curtis at the
> low end and Z1k at the upper end. When Otmar gets on top of demand it
> may be better to have something that has its own niche. A controller
> for up to 192 volt packs and with 700 amps on tap would have its own
> market space.
>
> I snipped the price part, it left me scratching my head. Are you
> talking about high voltage and 1100 amps for around $2k? That would be
> a steal! Are you talking about a 550 amp controller for $2k? That is
> going to tend to draw Curtis comparisons in many applications. I'm not
> sure if that is a good thing or a bad thing but I notice that even the
> low voltage Z1k tends to get Curtis comparisons.
>
> Again, I was just tossing out some of what ran through my head. If you
> find any of it helpful that's great. If not, well, remember what you
> paid for it <g>.
>
> Paul "neon" G.
>
--- End Message ---
--- Begin Message ---
----- Original Message -----
From: "David Roden" <[EMAIL PROTECTED]>
To: <[email protected]>
wants the moon for the price of Peoria. ;-)
>
> Sure glad I'm not the one trying to build components for EV hobbyists.
I'd
> lose my composure, if not my shirt. ;-)
>
>
> David Roden - Akron, Ohio, USA
> EV List Assistant Administrator
>
That's how I feel, Cheap prices for world class components.. nice way to
stay poor.
One reason I am not booking orders for the proposed controller yet.
I am not into bottom fishing for product market segment.
I have all the market segment I can stand with the Chargers.
You can't have all the features and not jack to cost right out of our
market.
Keep in mind ... a PFC50 From Xantrex or Sorenson is a Buck a watt. That
would make a PFC50 cost $12,500.
At a buck a watt I don't have a market.
With a CAN Buss and Isolation and UL an CE and FCC certs... my stuff would
be a Buck a watt. or More.
The question is not what I CAN make but what you all would buy. Since we are
a real small pond, pricing ones self out of a market is easy to do.
Keep this in mind when Thrashing me about costs of Controllers.
I starved to get DCP of the ground, I Got Debt getting the PFC chargers off
the ground... I don't need more hunger or Debt.
Rich Rudman
Manzanita Micro
--- End Message ---
--- Begin Message ---
Quickly
No it a all in one charger controller.
What a pain.
ACP has it's issues with thier system.
$250 is a very small amount for the time and effort needed to pull this off.
Once it's done it's %100 value Before the product ships.
This is the way it has to be.
I have not decieded to build the controllers yet...
This is all market research. I get to know what you guys think, and play
hard ball with the conecpts.
It's what made the Raptors... Same threads different Millenium.
Madman
----- Original Message -----
From: "James Massey" <[EMAIL PROTECTED]>
To: <[email protected]>
Sent: Saturday, October 21, 2006 4:23 PM
Subject: RE: EV controllers? the 4th option...
> At 12:55 AM 21/10/06 -0700, you wrote:
> >Hey Rich,
> >
> > > What do we need in the way of Controllers and what
> > > would you folks pay for a new design?
> >
> >What DC'ers need is a Curtis-killer, not something that tries to compete
> >with Otmar's already excellent offerings.
>
> G'day Rich, and All
>
> I agree with the comments of Raptor 600 capacity controller, and also the
> "all in one box" charger and controller.
>
> Maybe there is another option.
>
> I know that this is a 'non-commercial' approach, and leaves the designer
at
> considerable risk of "loosing their shirt" by designing something that
does
> not have to be built by them. The Raptor and Zilla design strategy
involves
> specialised construction methods that almost prevent other people from
> working on them. Being several thousand miles from the designer/builder I
> am relying on the Zilla that I bought to be reliable, but buying the next
> controller I'd look very closely at a controller that was designed to be
> repaired by competent techs, anywhere, and also something that I could buy
> in a partial assembled form to add the bits that I can get here, plus some
> of my time to complete the controller:
>
> A controller that is designed as a kit, using silicon that comes in a much
> more expensive package than would be used in a commercial construction.
> What I'm thinking of is isotop modules or something else that is industry
> standard and would allow for a 600A controller to be built using maybe 2
or
> 4 modules. This gets around the 'can full of dead Fets' when designing a
> controller. Yes, this will make it more expensive per unit in comparison
to
> a 1-design system, but it will allow for the builder to get E-bay or other
> surplus components, and save money, or go high-voltage or high-spec and
> build a megawatt controller if they have the cash.
>
> What this will involve is to design a control board and power boards in a
> manner that allows the assembly to go onto a flat surface, so the builder
> can choose air or water cooling. Supply the controller as anywhere from a
> pre assembled control board with bare power stages and no input DC/DC
> converter up to a fully assembled and tested controller. This would allow
> for the builder to put it together as any voltage he/she chooses, or to
put
> it together initially as a 450A controller and upgrade the power stage
> later to whatever power level is desired. It may be possible for such a
> system to be built as a sep-ex as some have requested with a single power
> device in the field but measuring (and limiting) the curent from the
armature.
>
> Peripheral things like precharge control and reverse/valet mode etc would
> need to be addressed on merit. Pre-charge control, current and RPM limit
> I'd consider essential. Reverse and series/parallel may be able to be
> designed as a 'tap-in' point in the controller for shift blanking and
valet
> mode power reduction (or reverse reduction) in the manner of open source
> projects able to be added on as people need them.
>
> For someone like Rich to take on a project like this, I feel that it would
> take something like a subscription project. Everyone who thinks they will
> want one (or more) puts up (for example) $250 to Rich. They get the value
> of what they put up as a discount on their kit once they are ready and
Rich
> gets the cash to do the project.
>
> What do people (and especially Rich, or anyone else who may be in a
> position to design such a system) think?
>
> Regards
>
> [Technik] James
>
--- End Message ---
