EV Digest 5072
Topics covered in this issue include:
1) Re: Motor Shafts, Pressing 101
by Jim Husted <[EMAIL PROTECTED]>
2) RE: Budget EV - motor mount and coupler
by "jerryd" <[EMAIL PROTECTED]>
3) Geo Metro EV on a budget
by "jmygann" <[EMAIL PROTECTED]>
4) Re: Budget EV - contactor controllers, low cost EV's
by "jerryd" <[EMAIL PROTECTED]>
5) Re: Fw: Geo Metro EV on a budget
by "jmygann" <[EMAIL PROTECTED]>
6) Re: Fw: Geo Metro EV on a budget
by Bill & Nancy <[EMAIL PROTECTED]>
7) Motor Shaft to Flywheel Coupling
by Jeff Shanab <[EMAIL PROTECTED]>
8) Re: Geo Metro EV on a budget
by Stefan Peters <[EMAIL PROTECTED]>
9) Re: Budget EV - contactor controllers, low cost EV's
by Tim <[EMAIL PROTECTED]>
10) EMeter quirks
by "John G. Lussmyer" <[EMAIL PROTECTED]>
11) RE: I want to build a PWM DC motor controller
by "Roger Stockton" <[EMAIL PROTECTED]>
12) Power Supply Isolation
by "Bill Dennis" <[EMAIL PROTECTED]>
13) RE: I want to build a PWM DC motor controller
by "Roger Stockton" <[EMAIL PROTECTED]>
14) Re: Fw: Geo Metro EV on a budget
by "Lawrence Rhodes" <[EMAIL PROTECTED]>
15) RE: Power Supply Isolation
by "Roger Stockton" <[EMAIL PROTECTED]>
16) Re: I want to build a PWM DC motor controller
by Lee Hart <[EMAIL PROTECTED]>
17) Re: Fw: Geo Metro EV on a budget
by Lee Hart <[EMAIL PROTECTED]>
18) honda hub adapters - any vendor done these?
by "Monty McGraw" <[EMAIL PROTECTED]>
19) Re: EMeter quirks
by Nick Viera <[EMAIL PROTECTED]>
20) Re: Power Supply Isolation
by "Paul Compton" <[EMAIL PROTECTED]>
--- Begin Message ---
Hey Rod
I'm actually using a cheap import 20 ton press that's about 15 years old and
shows the scars and stretchmarks of spitting out it's fair share of shafts.
I've been known (not that I encourage such things of course) to throw a
"little" cheater bar for extra persuading in coaxing a shaft out from time to
time to ; )
Your ten ton should be able to get a 6" golf cart motor sized armature shaft
out, but you may have to use the above noted persuading techniques to succeed
(not that I would recommend such things). If you can get the shaft to "pop"
loose the battle is won in the removal end of things.
On these armatures there is not a spacer between the comm. and the armature
lamination's (older motors did have them). Being there is no spacer to support
the comm from crushing into the lamination's, you can NOT use the comm as the
base to press a new shaft into the armature. If you try to do this you "will"
crush your armature.
I machine out old motor housings and pipe to create pressing cups. They are
machined out to grab just that little outer lip of the lamination's area that's
left above the winding wires. This does two things. First, it allows you to
press the shaft back in without pressing against the comm. The really almost
needed thing though is that if you try to use a small pipe or press tube on the
inside of the armature windings it tends to cause the lamination's to pinch the
shaft, so you need the cup to start anyway. If you grab the outside of the
body it allows the lamination's to slide outward, like how an oil seal works.
Machining out a pressing cup will be a must if you want to proceed.
Okay you've got your cup done you say(man that was quick, LOL). Well lets
pop a shaft out. First thing you need to do is heat that armature up to
350F'ish degrees in a oven (you need to use welding gloves to hold, and well
your hands are getting hot holding it). A household oven should be just fine
and there should be little to no fumes produced by this. I don't want to get
letters of hate from those who rested the insulation windings against the
elements or the BBQ pit idea didn't workout, etc. so use common sense.
Heating the armature causes it to swell and is what allows one to press a
Bake-O-Lite comm'ed armature without crushing the comm on the down stroke.
Once you have heated the armature rest it on the pressing cup (comm. up) and
apply some WD-40 or equivalent between the shaft and the comm. bore hole. Let
it sit for a minute before starting the pressing.
So here is where it comes down to the hairy area and the pressure starts,
hehehe. Bottom line is you are looking for just enough pressure to create the
"pop". The result you are looking for is a slightly lowered shaft. A slightly
lowered shaft and comm. combo is bad. I still pop comms, but it is rare. I do
know when to stop and to bore the shafts out of the comms so they can't be
effected any longer.
Lets say everything is going just fine and you have removed the shaft. Turn
the armature over and install the shaft of your choice. Before you install
that new shaft make sure you can insert the shaft into the comm area. For me
it's just a matter of sanding the knurl area that the comm sits on. Use a dab
of loc-tite on that area when repressing shaft. Failure to do this can result
in pushing the comm out as the shaft is re-installed. There is no reason to
worry about the comm not being pressed on, trust me here.
As to finally answer your question, these shafts sit on four knurl-like
grooves that are actually punched into the shaft. I will make a point of
grabbing some pics for you all to see. I'm quiet sure it will strip most of
your gears as to what actually keeps these shafts locked down. The Siamese 8
shaft made by Keith at Dutchman was in fact splined as you describe and is
working just beautifully.
If in fact anyone would like to attempt this call me up for you begin so that
you fully understand everything correctly and well for morale support :P
As for Rod, if you work on motors you have a really good shot at success.
Talking about this has brought some fond memories of 50 ton shaft pop's that
would give me heart flutters. Picture Rudman charging in the rain kinda
flutters, LMAO
Hope this helped
Best of luck
Jim Husted
Hi-Torque Electric
PS: FYI if you pop the comm you can bore the shaft out and space the brush
ring if it's a baby "pop" : o
Cya all!
Rod Hower <[EMAIL PROTECTED]> wrote:
Jim,
Just curious about the press you use for re-shafting.
I use a 10 ton triple compound press to do 1" diameter
shafts
on my 64 frame BLDC motors (these are used in busses
to blow air
on evaporators and condensors, usually 2 Hp continuous
rating).
I was thinking of pressing a new shaft in a golf cart
motor I have sitting around.
Do you use any locktite on the shaft?
What type of splines do you put on the shaft?
I use 8 total equally spaced with maximum upset height
of 0.006", although I usually file these down slightly
to make it press
a little easier.
Thanks,
Rod
--- Jim Husted wrote:
> Hey Steve
>
> I wanted to stop a little myth here as far as
> shafts are concerned. If ones shaft happens to be
> of a type that can not be used (or used easily), or
> in fact is broken or incorrectly machined by a local
> guy the motor is not ruined. I can press new (and
> sometimes salvage) shafts into the armature so that
> the motor is not only repaired but in fact becomes a
> little more EV usable.
>
> I would highly advise people to not weld anything
> onto the shaft as this will make doing a bearing
> change extreemly difficult. As far as using the
> opposite shaft end with no keyway, you could easily
> find soemone to cut one as a low cost option. This
> option would use the CE (I assume?) as the drive
> end. This could cause even more heat to be put on
> the brush end, and you could end up with a CE
> bearing burner. I do see lifttruck drive motor
> using the CE as the drive end but for EV's it may
> not be the best choice.
> Rather than cut the shaft off maybe you could use
> it for an RPM sensor even if it needed to be trimmed
> down, as an RPM gauge would be nice to have for all
> users.
>
> I did a re-shaft for Howard down in Cali (couple
> months back) after a local shop botched a shaft
> extention attempt. I actually used one of Waylands
> old ADC8 shafts (a spoil of war) and mod'ed it to go
> into one of those 15" Prestolite motor without
> issues. So hopefully Howard is out there cruising
> in his Rabbit with a little Wayland legend under the
> hood, lmao.
>
> I'm actually moding an ADC9 shaft to go into the
> 12 brushed MKB motor I have pics of on my site some
> of you may have seen. The down side to the re-shaft
> option is having to ship it to me or someone who
> specializes in these DC motors. Any WE should be
> able to press shafts. I can tell you that by the
> time you buy adapters and machine shops mods you may
> be getting close or possibly exceeded the costs of
> just re-shafting your armature. Now, not all motors
> are created equal so I'd need a motor number (or
> pic) to see what options you have for this motor.
>
> Thought you could use this info before making up
> your mind as to how you want to proceed.
> Best of luck
> Jim Husted
> Hi-Torque Electric
>
>
>
>
>
> [EMAIL PROTECTED] wrote:
> It seems like this is where most of the beginners
> falter. I have seen some
> pretty bad setups over the years, and I have also
> seen some good ones. Most
> conversion books say stick with the professionals.
> Have them make you a nice
> taper lock and go with that. I've seen the old set
> screw idea, on my first car.
> I've also heard of people doing that. I'm sure
> people have tried press fits
> and welding. Welding is risky business if you get it
> out of round. Your
> motor will pretty much be scrap if you mess it up.
>
> Stick with the professionals ... or at least seek
> the guidance of an
> experienced Mechanical Engineer and Machinist. Well,
> I did that. I consulted with
> "myself" and a few of my friends. But, I am still
> not sure if my set-up will
> work. I have a unique situation of a nonstandard
> "motor shaft." Actually, my
> motor shaft has several options:
>
> 1. The motor shaft itself has a 7/8" diameter (max
> OD) x 1-1/2" length, 13
> tooth straight splined shaft. That leaves a few
> options:
> A. Use a 7/8" diameter taper lock over that shaft.
> But, I am not
> sure it will have the surface area to transmit the
> torque. Also, there is no
> keyway, so I am concerned about that. Available from
> McMaster.com (5926K22
> Steel Quick-Grip Keyless Bushing for 7/8" Shaft
> Size, for 1-3/4" Component Bore In
> stock at $53.15 Each)
> B. As suggested by others on the EVDL, get the
> mating part for that
> 7/8", 13 tooth splined shaft. I do have it. But, it
> is also an odd shape.
> There is an area on is about 1.61" diameter by 1"
> length. But, where can I get
> a taper lock to fit over a 1.61" diameter. It is not
> even close to any of
> the sizes I have seen. Maybe the closest is 19 mm.
> But, I am not sure how much
> variation those taper lock connections can
> accommodate. I looked at making a
> custom adapter for that thing. Grind in 2 D flats
> 180 degrees apart and
> slide over a custom mating part. The problem is that
> the mating part will have
> four inside corners with 0 radius that can't be made
> with conventional milling.
> The only way I see to make this is wire EDM which
> makes the adapter cost
> about 5x my motor cost. Also grinding those flats
> into the hardened steel gear is
> a total nightmare. This would work, but very
> expensive.
> C. Custom design a coupler with the mating inside
> spline that bolts
> up to the flywheel and slides over the motor shaft.
> No problem to design
> this. I can model it up in CAD and get it machined,
> but again, it has many 0
> radius inside corners and can only be made with wire
> EDM.
>
> 2. Flip the motor the other way and use the 1" non
> keyed tail shaft. I
> didn't want to do this. I was planning to cut that
> shaft off. Also, it has no
> keyway. The motor face mounting holes on this side
> are smaller bolt size and
> there are only 3 of them. It would be much better to
> bolt up to the actual motor
> face.
>
> Any suggestions?
>
> Thanks,
>
> Steve
>
>
>
>
>
>
>
>
>
>
> ---------------------------------
> Yahoo! Photos
> Got holiday prints? See all the ways to get quality
> prints in your hands ASAP.
>
>
---------------------------------
Yahoo! DSL Something to write home about. Just $16.99/mo. or less
--- End Message ---
--- Begin Message ---
Hi Glen, Mike and All,
----- Original Message Follows -----
From: Mike & Paula Willmon <[EMAIL PROTECTED]>
To: [email protected]
Subject: RE: Budget EV - motor mount and coupler
Date: Sat, 07 Jan 2006 23:00:03 -0900
>pics 6 and 7 in this lineup give the basic idea of a clutch
>plate center adaptor. Remember though you lose the use of
>the clutch.
>
>http://www.geocities.com/CapeCanaveral/Lab/4429/cl3.htm
Nice method of doing it !!
>
>-----Original Message-----
>From: [EMAIL PROTECTED]
>[mailto:[EMAIL PROTECTED] Behalf Of Glenn
>Meader Sent: Saturday, January 07, 2006 2:50 PM
>To: [email protected]
>Subject: Budget EV - motor mount and coupler
>
>
>Was Re: Geo Metro EV on a budget
>
>> jerryd says: Use 2 box sections to mount the motor and a
>> make a coupler with the clutch plate center brings this
>to about $60.
>
>Could you give more details on this?
>
>What's a "box section"?
Just 2 square or rectangle alum extrusions or one could
use flat plate or large angles. I like the box as it give
spacing to fit the coupler though to be honest, I've never
done an EV this way but many other motor, trans, alts, ect
connections. I prefer dirct drive for my built from scratch
EV's but hate to see people spend so much on an adaptor
plate, couplings.
I'd measure the distance needed for the coupler to find
the box thickness needed and mount the boxes on the motor,
align it up then drill the boxes to transmission bell
housing attach bolt holes.
>
>How do you make a coupler with a clutch plate center for
>$60? Any writeups on this?
>
>
For the motor get most any coupler half of the right
type, shaft size, or whole one and retain the piece that
fits the transmission pilot shaft, usually the clutch
spline. Now cut the spline, ect, from the rest of the plate,
ect and weld, bolt, ect it to the coupler half at the
correct length to mate the motor shaft to the transmission
shaft. You could possibly need a bearing on the trans side
depending on which unit you are doing..
If one needs to machine something in this sometimes
you can put it on the motor and run it on 12-24 vdc and use
it as a lathe.
HTH's,
Jerry Dycus
>
--- End Message ---
--- Begin Message ---
Newbie here
Now I am really confused.
On my metro I currently have one contactor and a 5 speed trans. 48 volt
My thought is to go to a Alltrax 7245 controller.
What are the alternatives ??
--- In [EMAIL PROTECTED], Stefan Peters <[EMAIL PROTECTED]> wrote:
>
> jerryd wrote:
> > Let's be
> > pratical and stop pricing EV'ers out of the market because
> > some of us are hung up on EC's..
> > HTH's,
> > Jerry Dycus
> >
>
> Hear, hear!
>
> Though hopefully you can get some of the automation benefits of a
> monolithic EC at the price of a DIY 5 speed contactor controller
someday ;)
>
--- End Message ---
--- Begin Message ---
Hi Glen and All,
----- Original Message Follows -----
From: "Glenn Meader" <[EMAIL PROTECTED]>
To: <[email protected]>
Subject: Budget EV - contactor controllers
Date: Sat, 7 Jan 2006 15:52:45 -0800
>Was: Geo Metro EV on a budget
>jerryd wrote:
>> With a 3-5 speed contactor controller you won't notice
>> the smoothness difference between the CC and the EC as
>after starting up...
>
>Where do you get, or how do you make a 3-5 speed contactor
>controller?
As Lee said, it's just some big switches, contactors
as we call them that series/parallel the batteries to 2
voltages, say 36/72vdc in this case to give you 2 speeds, a
starting resistor to keep starting amps down and for parking
lot speeds. For more speeds you can series/parallel them
into 4 packs, use more resistor speeds or weaken the field
winding to make the motor turn faster by shorting the fields
out with a resistor or 2 partically. I do this by using a
12-18" piece of 10-12 gge solid copper wire and a contactor.
This is what some golf carts and many forklifts use
reliably. Forklifts sources can be a good place to find used
ones cheap, free from abandoned, junked, surplus ones as
well as motors, chargers. Also surplus suppliers. GC part's
are only good for under
1,000 lb EV's.
Golf cart chargers are a good, cheap way to charge
flooded batteries cheap and ferro chargers that just have a
cap and a transfrmer can be easily modifiedc by using a
bridge rect instead of the stock full wave rectifier to get
72vdc charging. You can parallel then to get higher charge
levels. Fork lift ones sometimes can be used the same way.
There is no reason you have to go into debt to have
an EV. If you study, good at scrounging, you can do it for
under $1,000 !!
I built my E woody from scratch for under $800 so
if one wants an EV, they can do it at a reasonable cost if
you don't get gadgetitis and go with proven,industrial
methods.
HTH's,
Jerry Dycus
>
>Any pointers to info on this?
>
--- End Message ---
--- Begin Message ---
Us guys in the Ev bike group do a 10 mile "round trip" comparrison...
Maybe a 10 and 20 mile round trip "elapsed time" with "total
weight" of vehicle would be interesting ...
I will try the 10 mile first..
--- In [EMAIL PROTECTED], Tim <[EMAIL PROTECTED]> wrote:
>
> It looks like the 48 volt system is more efficient @43mph. (48v x
210a =
> 10,080 watts vs. 120v x 90a = 10,800 watts)
> Efficiency is the key, and comparing total input power @ a given
speed
> or watt hours per mile gets us to compare apples to apples.
>
> I believe the two above cars could still reduce their consumptions
by an
> additional 10 - 20%. Have all their fluids been replaced with
> synthetics? (i don't care which brands) Low rolling resistance
tires
> installed? Alignment done? Verify no brake drag?
>
> Idea...is there a public spreadsheet or equivalent that would
allow
> everyone to input there pack volts, pack amps, at a preset speed,
or
> watt hours per mile at a preset speed, so we all can compare
> efficiencies between cars? Kind of like what
http://www.dragtimes.com
> does with times but we all would be comparing efficiency.
>
> Tim
--- End Message ---
--- Begin Message ---
I'm running 72 volts, 40mph in 2nd and 45 mph in 3rd @100 amps.
Bill
jmygann wrote:
How fast can you go @ 100 amps ? How many volts ?
--- In [EMAIL PROTECTED], Bill & Nancy <[EMAIL PROTECTED]> wrote:
1 hr rating @ 200 amps. I hadn't even considered the controller
specs.
It is best for pack longevity to keep running amps below 100? Just
learning as much as I can about driving the vw and the most efficiant
operating range.
--- End Message ---
--- Begin Message ---
Good ideas, here are a few more
Make your own taperlock, I did. In your case reporoduce the outside
dimensions of a taper lock and send it out and have the splines broached
on the inside, then split it with a cut off wheel.
http://cvevs.jfs-tech.com/300zxdrivetrain.jpg
imageine 6 1/4-20 SHCS into the red part
Toolpost OD grind, or lathe, the adapter you have down to a size that
is mateable.
I suppose If you have room make a short dummy shaft that rides on
bearings and allow the motor to slide up and connect and the clutch to
mount to the reproduced crank pattern. Think old water pump setup. This
coul reach over to frame rails and allow you to eaisily remove motor.
Wild idea: If that motor shaft spline will fit a clutch disk, turn the
clutch around? or go clutchless bolting the disk to a tranny adapter or
find a spline to spline adapter.
No wire EDM, Broach. There are places you can send a part to and they
will push or pull a broach thru it/into it.
--- End Message ---
--- Begin Message ---
jmygann wrote:
Newbie here
Now I am really confused.
On my metro I currently have one contactor and a 5 speed trans. 48 volt
My thought is to go to a Alltrax 7245 controller.
What are the alternatives ??
--- In [EMAIL PROTECTED], Stefan Peters <[EMAIL PROTECTED]> wrote:
jerryd wrote:
Let's be
pratical and stop pricing EV'ers out of the market because
some of us are hung up on EC's..
HTH's,
Jerry Dycus
Hear, hear!
Though hopefully you can get some of the automation benefits of a
monolithic EC at the price of a DIY 5 speed contactor controller
someday ;)
Oh, that's just me yacking... notice the "someday" part.
There is the Navitas controller at cloud electric, 72-96V, 600A, lots of
features:
http://www.cloudelectric.com/item.jhtml?UCIDs=866086%7C1321277&PRID=1495620
They also carry a 48V, 400A one for much less.
Or you can wire up a 3 speed contactor controller.
--- End Message ---
--- Begin Message ---
It's possible to save even a little more money. If using the original manual transmission and clutch, this will give yourself additional speeds and you can then eliminate the starting resistor...just select 36 or 72 volts via the 2 speed electric contactor controller and let the clutch out slowly in first gear like you do on a gasser then once you cannot accelerate any further then shift to second gear, then third, etc till you obtain your desired mph. With a 5 speed tranny you can have a possible 10 speeds.
The above scenario would yield the approx theoretical following speeds if 72v
and 5th gear equaled 60mph.
72v 5th gear = 60.0mph
72v 4th gear = 43.9mph
72v 3nd gear = 32.4mph
72v 2nd gear = 21.4mph
72v 1st gear = 13.9mph
36v 5th gear = 30.0mph
36v 4th gear = 22.0mph
36v 3rd gear = 16.2mph
36v 2nd gear = 10.7mph
36v 1st gear = 7.0mph
Maybe use 36v for city driving and 72v for hwy?
Note - The motor should be sized as not to exceed 58ft/lbs @36vdc (keep from
destroying tranny).
Tim
jerryd wrote:
series/parallel the batteries to 2 voltages, say 36/72vdc in this case to give
you 2 speeds, a
starting resistor to keep starting amps down and for parking lot speeds.
Hi Glen and All,
----- Original Message Follows -----
From: "Glenn Meader" <[EMAIL PROTECTED]>
To: <[email protected]>
Subject: Budget EV - contactor controllers
Date: Sat, 7 Jan 2006 15:52:45 -0800
Was: Geo Metro EV on a budget
jerryd wrote:
With a 3-5 speed contactor controller you won't notice
the smoothness difference between the CC and the EC as
after starting up...
Where do you get, or how do you make a 3-5 speed contactor
controller?
As Lee said, it's just some big switches, contactors
as we call them that series/parallel the batteries to 2
voltages, say 36/72vdc in this case to give you 2 speeds, a
starting resistor to keep starting amps down and for parking
lot speeds. For more speeds you can series/parallel them
into 4 packs, use more resistor speeds or weaken the field
winding to make the motor turn faster by shorting the fields
out with a resistor or 2 partically. I do this by using a
12-18" piece of 10-12 gge solid copper wire and a contactor.
This is what some golf carts and many forklifts use
reliably. Forklifts sources can be a good place to find used
ones cheap, free from abandoned, junked, surplus ones as
well as motors, chargers. Also surplus suppliers. GC part's
are only good for under
1,000 lb EV's.
Golf cart chargers are a good, cheap way to charge
flooded batteries cheap and ferro chargers that just have a
cap and a transfrmer can be easily modifiedc by using a
bridge rect instead of the stock full wave rectifier to get
72vdc charging. You can parallel then to get higher charge
levels. Fork lift ones sometimes can be used the same way.
There is no reason you have to go into debt to have
an EV. If you study, good at scrounging, you can do it for
under $1,000 !!
I built my E woody from scratch for under $800 so
if one wants an EV, they can do it at a reasonable cost if
you don't get gadgetitis and go with proven,industrial
methods.
HTH's,
Jerry Dycus
Any pointers to info on this?
--- End Message ---
--- Begin Message ---
I've just finished re-connecting everything and am working on
charging my NiCd pack in my Sparrow.
One thing I've noticed is that the EMeter current reading is about
15-20% higher than my Analog ammeter I have on the charger
shows. (Analog meter was showing 21A, EMeter was 24.4A). Checking
mv across the Ammeter shunt with my Fluke meter, showed the Ammeter
was quite correct.
Checking mv across the 500A shunt for the Emeter gave some odd readings.
Across the 2 big bolts for the power lines, it was 2.0mv (which is correct).
Across the 2 small sense line bolts I was getting 2.4mv.
AARRGGHH!!!
--
John G. Lussmyer mailto:[EMAIL PROTECTED]
Dragons soar and Tigers prowl while I dream....
http://www.CasaDelGato.com
--- End Message ---
--- Begin Message ---
Stefan Peters [mailto:[EMAIL PROTECTED] wrote:
> I thought the "buy it one part at a time from retailers"
> parts cost was more then the retail cost of most electronic
> gadgets? Is that different in this industry?
Dunno. What I do know is that there is less than $200 worth of parts in
a Curtis, probably less than $100 when bought in quantity. A Curtis is
fully analog, no software to write and no micro to program, so building
a Curtis clone seems to be a much more attractive proposition for the
average hobbiest on a budget even if the parts cost totalled $500 at
Digi-Crime, etc.
> But you have to include the cost of a BMS in that as
> well to get an equivalent exchange.
Not really. Even if we assume that you will get a really good BMS
functionality incorporated into your controller it is unlikely that many
people will spend *more* on a "controller" as a result. Unfortunately,
even though your controller/BMS may represent extremely good long term
value the fact is that most EVers are too cheap to buy a decent charger
never mind spending extra on a BMS.
I think you have to consider the BMS functionality as a side
benefit/feature and assume that those who consider your system will do
so because they can build your *controller* cheaper than they can
build/buy a conventional controller. After all, their EV is useless
without a controller, but is perfectly usable without a BMS.
Unfortunately, I think that looking just at cost it would be cheaper for
a hobbiest to build a basic PWM controller (and easier too).
> > Could you give a paragraph or two summary/overview of how
> > your system is expected to work?
> >
>
> I'm still figuring that out with these posts ;)
> Think of LEGOs
Problem is, I am quite familiar with how LEGOs work, but despite all you
have written about your system I have only a foggy idea of how it might
work, and am not at all sure that my ideas accurately reflect what you
have in mind.
The names you have assigned your blocks ("Power Modules", "Subnet
switch", etc.) sound impressive, but really don't tell me anything about
what they do or how my EV traction wiring would look with them vs a
conventional PWM controller or a conventional contactor controller, etc.
My main concern is that I *think* you envision being able to switch each
battery individually into or out of the traction pack, and doing so in
such a way that each battery may be placed in series or in parallel with
the other batteries in the string. This ability is required since you
must be able to place the batteries in series to achieve the higher
voltages required for higher speed operation, and you must be able to
place the batteries in parallel if you are to achieve the lower-per
battery current draw that a PWM controller provides at low motor
voltage/high motor current operation.
This suggests to me a nightmare of high current wiring and fairly
complicated switching capability at each battery, with all the losses
that the switches and extra connections entail. And, don't forget that
the 2/0 cable and terminals typically used to interconnect our traction
batteries are neither particularly light or cheap.
It also suggests to me that each battery must be charged individually;
i.e. the "rest" state of the system is with all batteries disconnected
from one another. Certainly, if one took advantage of your system to
mix and match batteries of different types or size they could not charge
them in a series string even if the controller was willing to connect
them in such a way for charging.
Finally, the use of switches at each battery implies to me much higher
losses than a conventional system. Consider the specs for a Curtis
1231C (a common PWM controller, but hardly state-of-the-art): 0.30V drop
per 100A, 550A maximum (72-120V model). This would be 1.65V drop @
550A. Let's assume the 120V pack sags to 117V at this load, so 1.65V
corresponds to about 1.4% loss.
What sort of loss would your system have at 550A if the pack were
composed of 10 12V batteries? How about if it were composed of 20 6V
batteries (the battery of choice for cost-conscious EVers)?
> It takes more electronic skill (and parts cost) to build a
> basic power supply then to flash a micro.
Of course; it is not an "electronic" task to flash a micro. It is not a
matter of what is harder to do, but rather what one is comfortable with.
I think a lot of EV hobbiests are far more comfortable soldering and
bolting things together according to provided instructions than using PC
to flash a micro, even if flashing the micro is actually a simple task.
> I've seen Curtis schematics. Looks like greek spaghetti to me.
I'd really like to see schematics for your system (don't bother with the
details of the micros or exact FET configuration even, just show a
little box for each of your modules and include standard switch symbols
with appropriate connections to the traction wiring in place of each
FET/contactor, etc.). I have this feeling that it will make the Curtis
schematic look like child's play, but perhaps you will set me straight.
> > A (good) contactor controller re-arranges the battery pack into
> > different voltages, while ensuring that all batteries are
> > used at each voltage level (so that all are discharged equally).
>
> That would always require a "matched" set of batteries, even
> then, you will be throwing away perfectly good lead sometime.
Unless you start by populating your traction pack with random mismatched
batteries scaveged from here or there, you always start off with a
pretty matched set of batteries. If the controller discharges them
equally and they are recharged equally then they tend to remain fairly
well matched over most of their usable life.
> > I think the rectactor style contactor controller may have the most
> > promise for "modernisation"
> That's a very good idea, but you can't do the PWM with
> contactors.
Of course you can; a contactor is a switch just like any other, you just
can't switch it as fast as a solid state switch and it will wear out
sooner if you switch it more often. PWM does not have to be performed
at the hundreds or thousands of Hz; the sort of manual PWM that a
contactor controller is called upon to perform is quite slow, perhaps
selecting the higher speed step for a few seconds, then the lower speed
step for 10s or more, etc.
The improvement here is that the micro can look after switching back and
forth between the appropriate steps, at a rate appropriate for the types
of switches in use, without the user having to vary the throttle
position at all.
> I'm a firm believer of "use your resources where they work
> best". Since I'm a network & software engineer with tons of
> experience on high scalability projects, I figured that this
> would be the most appropriate application of what I know,
> versus what I don't.
Sounds good to me. I don't mean to sound like I'm trying to discourage
you; I just want to understand what it is you are building and to try to
understand how it compares to the alternatives. Only then can I
appreciate what sort of advantages it might offer.
Cheers,
Roger.
--- End Message ---
--- Begin Message ---
There was a discussion here a couple of months ago about how some EVers on
the list use power supplies as DC-DC converters. If the power supply is no
longer receiving AC to isolate through a transformer, how does it maintain
isolation between the traction pack and the 12V system when receiving just
DC from the traction pack?
Thanks.
Bill Dennis
--- End Message ---
--- Begin Message ---
Danny Miller [mailto:[EMAIL PROTECTED] wrote:
> I get 3.779 times the DC resistance for 2/0, or a 278% increase, for
> 10KHz.
I'm wondering if you are noting this as a potential concern for the
networked controller proposal, or for typical PWM controllers, or
something else?
Basically, with a typical PWM controller it is of little concern since
the ripple on the battery cabling is a small fraction of the average or
DC current (like 10% or less). There will be more ripple in the motor
loop, expecially with a low inductance motor, however, it is fairly
common to use doubled runs of 2/0 in the (much shorter) motor loop vs a
single run in the battery loop.
It might be of some interest for the networked controller scheme if it
PWMs the batteries at such low frequencies that it becomes impractical
to use capacitors to smooth the ripple, however, at lower frequencies
the skin depth increases such that the effect becomes even less
significant.
Cheers,
Roger.
--- End Message ---
--- Begin Message ---
1221B. Lawrence..........
----- Original Message -----
From: "James Massey" <[EMAIL PROTECTED]>
To: <[email protected]>
Sent: Sunday, January 08, 2006 1:10 AM
Subject: Re: Fw: Geo Metro EV on a budget
At 01:06 AM 8/01/06 -0800, Lawrence Rhodes wrote:
That being said. A controller can fail at any time. I was crusing at 100
amps at 45mph and all hell broke lose. You have to have contactors &
breakers for safety. When my 144vdc contoller went it was full speed &
everything failed to stop the car. Letting off the throttle. Turning the
key off. The fuse didn't blow & the contactor fused. The breaker saved
me & the motor. .
Hi Lawrence
What controller?
Regards
James
--- End Message ---
--- Begin Message ---
Bill Dennis [mailto:[EMAIL PROTECTED] wrote:
> There was a discussion here a couple of months ago about how
> some EVers on the list use power supplies as DC-DC
> converters. If the power supply is no longer receiving AC to
> isolate through a transformer, how does it maintain isolation
> between the traction pack and the 12V system when receiving
> just DC from the traction pack?
The first thing a switching supply does is to rectify the incoming AC to
DC. It then takes this DC and chops it at a high frequency so that it
can be passed through a small transformer that converts it up or down
(but usually down) to the desired voltage. The output of this
transformer is isolated from the input, and is then rectified and
filtered.
So, it doesn't matter whether you feed the supply AC or give it DC
directly.
Cheers,
Roger.
--- End Message ---
--- Begin Message ---
Cory Cross wrote:
>> When the controller turns on, the current ramps up (because current
>> cannot change instantly in an inductor -- the wiring). When it turns
>> off, the current keeps coming, boosting up the voltage on the
>> capacitors, which is then drawn down next time the controller turns on.
You have the right general idea, but the details are wrong.
The motor has most of the inductance. During the controller "on" time, motor
current ramps up. During controller "off" time, motor current ramps down. The
difference between the peak and valley motor current is the motor "ripple
current". Suppose that during the "on" time motor current ramps up from 90 to
110 amps; during the "off" time, it ramps back down form 110 to 90 amps. Then
motor current is 100 amps average, with 20% peak-to-peak ripple.
The motor ripple current doesn't really bother it. It causes minor torque
pulsations and vibration; but electric motors are so smooth compared to ICEs
that no one notices. Ripple causes the motor to act as a loudspeaker, so you
can hear the controller frequency; this bothers some people but not others.
It's the BATTERY ripple current that we are more worried about. In a PWM buck
converter with no input capacitors, the battery current iz ZERO during the
controller "off" time. During the controller "on" time, it follows the motor
current; 90 to 110 amps for the above example. So at a 50% duty cycle, the
battery current is 0,0,90,110 amps; the average is 50 amps battery current.
The controller is converting full pack voltage at 50 amps average into half
pack voltage at 100 amps average for the motor.
But the battery isn't delivering 50 amps; it's delivering 100 amps. So, you
have to use 100 amps in Peukert's equation to estimate the amphour capacity;
not 50 amps. If you have some huge 1000 amphour forklift battery, the
capacity difference between 50 and 100 amps is insignificant. But if you have
little 50 amphour EV batteries, that's the difference between the 1-hour and
the 0.5hour rates -- there's a *big* loss of capacity.
So, the purpose of the input filter capacitors is to smooth the huge battery
ripple current to reduce or eliminate this Peukert effect. It's not the
capacitance per se that helps; it's the ESR of the battery vs the capacitors
that matters. If they have the same ESR, then half the ripple comes from the
capacitors, half from the battery, and you've cut your range loss in half.
Roger Stockton wrote:
> the inductance of the battery wiring is (usually) *very* small, and so
> the ripple current they see is quite small compared to the average value.
This is true, but the inductance of the batteries and their wiring causes
another problem.
When the controller switches "off", the battery current goes from max to zero
*very* quickly. A modern IGBT- or MOSFET-based controller turns off in less
than 1 microsecond. The battery wiring inductance fights this rapid change in
current, which causes a voltage spike across the controller. If there are no
input capacitors, then the transistors or diodes will be destroyed by this
voltage spike.
Because it is so brief, the capacitors to handle this problem can have less
capacitance, but need an exceptionally low ESR. Special film capacitors work
the best, but you can also get by with lots of cheaper electrolytics in
parallel.
>> I suppose you could use enough capacitance to solve the problem on the
>> battery end, but you'll need to add more inductance or resistance to the
>> motor. Being that more resistance is bad, you'll have to add quite an
>> inductor that can take full motor current -- probably will be larger
>> than the motor itself!
These are separate problems, as described above. If you want to add motor
inductance, the easiest way is to use a motor with more iron in its outer
casing; that's what fork lift motors do. The motor is heavier, but not as
heavy as adding a separate inductor.
> This is exactly the approach I am taking with my EV1 SCR controller. It
> is simply impractical to add enough bus caps to absorb the required
> ripple current at its low (<300Hz) switching frequency, so I am also
> adding a largish inductor to the motor loop.
Adding inductance to the motor loop may not reduce battery ripple current
enough to matter, unless the motor ripple is huge. A typical forklift motor
has about 300uH of inductance. Let's say your EV1 applies 48v at 300 Hz and
50% duty cycle (1.66 msec on-time). Then peak-peak motor ripple is I=VT/L =
48v x 0.0016ms / 0.003h = 266 amps. That's not bad.
But if you used an Advanced DC motor that has more like 100uH of inductance,
peak-peak ripple would be tripled, to 800 amps. That's pretty bad!
However, the EV1 doesn't run at 300hz except when just getting started from a
dead stop. It quickly brings the frequency up to 1 KHz or so, bringing the
ripple back down under 250 amps. So, I think you still may be better off
adding input capacitors instead of output inductors.
--
Lee A. Hart 814 8th Ave N Sartell MN 56377 [EMAIL PROTECTED]
--- End Message ---
--- Begin Message ---
Bill & Nancy wrote:
> Your recommendation is to run 90 amps or less on a 120 volt system, so
> running 150 amps or less on a 72 volt system would be equal?
Correct. My ComutaVan, which is similar in size and weight to a Geo Metro (but
a lot more wind resistance) drew about 200a at 55 mph with a 72v pack.
Actually, the pack of twelve golf cart batteries delivered about 60v at 200a
to the motor doing this, which is about 12 horsepower to the wheels.
Pack voltage doesn't really matter, if the rest of the system is designed
appropriately. The Lectric Leopard with a 48v system drew about 250 amps to
do the same thing. It had two strings of eight 6v batteries, so each
delivered 125a and sagged to about 40v. 40v at 250a is 10 horsepower, which
is about right given the reduced wind resistance.
My rebuilt Lectric Leopard has a 132v system, with ten 12v AGM batteries. It
draws 80 amps at 55 mph, with the batteries sagging to 120v. 120v at 80a is
9.6 horsepower; a bit less due to my pack weight being 300 lbs less.
--
Lee A. Hart 814 8th Ave N Sartell MN 56377 [EMAIL PROTECTED]
--- End Message ---
--- Begin Message ---
I've got a Honda B16 SC4 5-speed manual transmission. I'd like to get an
adapter made for an Advanced DC Motor, I think the 8" motor is the best fit.
Is there any vendor that has done Honda transmission adapters?
Thanks,
Monty McGraw
Spring, TX
--- End Message ---
--- Begin Message ---
Hi,
John G. Lussmyer wrote:
I've noticed is that the EMeter current reading is about
15-20% higher than my Analog ammeter I have on the charger
shows. (Analog meter was showing 21A, EMeter was 24.4A). Checking
mv across the Ammeter shunt with my Fluke meter, showed the Ammeter
was quite correct.
Checking mv across the 500A shunt for the Emeter gave some odd readings.
Across the 2 big bolts for the power lines, it was 2.0mv (which is correct).
Across the 2 small sense line bolts I was getting 2.4mv.
Hmmm... that is weird. Though rest assured that yours isn't the only
E-meter that has these additional "features" ;-)
The E-meter in my Cherokee has developed an annoying habit in that it
consistently shows a current of +00.1 to +00.2 Amps when there is no
current flowing across the shunt. Thus, according to my E-meter, my EV
is magically recharging itself anytime it isn't being used. After a day
it will show that the batteries have received almost 1kWh of energy!
The only thing I could come up with was that maybe the little DC/DC
converter I am using to power the E-meter is producing some ripple in
it's output that is bothering the E-meter; or something similar that the
E-meter is too lame to deal with...
--
-Nick
1988 Jeep Cherokee 4x4 EV
http://go.DriveEV.com/
http://www.ACEAA.org/
--------------------------
--- End Message ---
--- Begin Message ---
Switching power supply (simplified)=;
input noise filter
input rectifer
input capacitor
switching element
high frequency transfomer (20Khz up to 2Mhz in some cases) provides the
isolation
output rectifier
output inductor
output capacitor
The isolation is done after the AC is converted to DC, so you can just feed
it DC.
Paul Compton
www.sciroccoev.co.uk
www.morini-mania.co.uk
www.compton.vispa.com/the_named
----- Original Message -----
From: "Bill Dennis" <[EMAIL PROTECTED]>
To: <[email protected]>
Sent: Sunday, January 08, 2006 8:31 PM
Subject: Power Supply Isolation
There was a discussion here a couple of months ago about how some EVers on
the list use power supplies as DC-DC converters. If the power supply is
no
longer receiving AC to isolate through a transformer, how does it maintain
isolation between the traction pack and the 12V system when receiving just
DC from the traction pack?
Thanks.
Bill Dennis
--- End Message ---
