More news on my car
Randy Pollock
After getting some errors from the controller for my motor, I've been
exchanging e-mail with Azure. They recommended checking the grounds
(all good) and making sure that the cable between the controller and
the motor did not pass too close to the speed sensor or serial
cables. I did all of that and took the car for a test drive this
morning. It seemed a little better, so I decided to make my first
drive on the freeway. I did about 3 miles - guessing that I was doing
65 MPH but I can't be too sure because the speedometer needs to be
calibrated. Since I am still running with a weak 12V battery and no
DC-DC convertor, I got off the freeway and started heading home on
surface streets.
A mile later, the car started hesitating going up a hill and I
smelled something burning. I figured the smell was the transmission
fluid I am leaking, but slowed down and was getting ready to find a
place to stop. The car got even more sluggish. Finally, the power-on
indicator light on the controller went out and I lost all motor
power. Turning off and back on the car wouldn't get the light to come
back on. After a couple of minutes of checking the 12V supply, I
decided to check the high voltage side (the Azure controller will not
turn on without both voltages present). I pretty quickly found the
problem - and a picture is worth a thousand words:
http://gallery.me.com/randypollock#100025
I fried my shunt. It looks like I am back in the garage for a
while. Damn.
Better luck to everyone else's projects. I'll let you know what I
learn.
- Randy
John RA Benson
due to a loose connection, creating arcing or high resistance which
generates heat and a self destructive reaction. I'm guessing based on
where it melted the shunt was bad to have melted that entire joint
(unless it's not rated for the current you pushed through it). Did
the ampmeter go wacky during the meltdown? How are the other
connections nearby?
Ouch -
jrab
On Oct 5, 2008, at 21:28, Randy Pollock <ra...@alumni.caltech.edu>
wrote:
OTTO
When you are readyto replace your shunt, you might want to consider
either locating it a little further away from conductive surfaces OR
adding some insulation between exposed DC parts and surrounding
components. Electric motor repair shops can sell you "fish paper"
which is strips or sheets of fiberglass which you can cut and silicone
(better than glue) in place as barriers in and around energized
components.
DC has a tremendous ability to arc especially when humidity is high
and accumulated dust is present, tracking can occur and this is usally
followed by more tracking and soon an arc.
Arcing didn't appear to be the cause of this problem. Overheating did
occur and it looked like the smaller wire either failed at the
connector as a result of the heating or perhaps it was loose or
overloaded.
Looking closely at the shrinkwrap on the power lead there doesn't
appear to be the heat damage that you would expect from an overload
condition. It appears that (as you surmised) that the shunt may have
been defective and as it began to fail the heat progressed from the
conductive fins outwards to the connection, though the base block
melting pattern would suggest the heat was higher at the connection
bolt.
One last (remote) possibility is that the bolt that connects the power
lead to the shunt block somehow was able to loosen or get hot which
caused the base block to melt. the high current condition drew more
curent than the shunt was capable of withstanding and it failed.
Again, the extreme melting under the connection bolt points to this.
Let us know what you found and photos are always welcome...
racunniff
> Randy,
> When you are readyto replace your shunt, you might want to consider
> either locating it a little further away from conductive surfaces OR
> adding some insulation between exposed DC parts and surrounding
> components. Electric motor repair shops can sell you "fish paper"
> which is strips or sheets of fiberglass which you can cut and silicone
> (better than glue) in place as barriers in and around energized
> components.
info
on the voltage ratings of the various popular shunts? I'm nearing the
end
of a 312V conversion, and am concerned about arcing with the typical
shunts out there. Should I just go for an extremely high-amp shunt
(1500A) and hope for the best? It's really hard to tell - none of the
spec sheets I've seen list voltage (and a few seem to imply that
the shunt was designed for 12V usage even though it does not
explicitly say that...)
I may just go with Hall-effect. Safer, but more expensive and fewer
options.
Mike S
Randy,
From your pictures, the fried shunt looks like it might be the 200-amp
model (Empro model HA-200-50). This is what came in the Electro-
Automotive kit I received but it is not the correct part. We all need
to be using is the 400-amp model (HA-400-50). The voltage across this
shunt at +200 amps will be +25 mV; just what the ±200 amp meter is
expecting to see at full current.
With the 200-amp shunt the EA meter will be pegged at ±100 amps and
for the same current, power dissipation in the 200-amp shunt will be
double that in the 400-amp version.
Let Mike Brown know about this -- he needs to send you the right
part.
(I wonder if that bolt on the left side might have been a little
looser than the other one...)
Mike Sasnett
Randy Pollock
> From your pictures, the fried shunt looks like it might be the 200-amp
> model (Empro model HA-200-50). This is what came in the Electro-
> Automotive kit I received but it is not the correct part. We all need
> to be using is the 400-amp model (HA-400-50). The voltage across this
> shunt at +200 amps will be +25 mV; just what the ±200 amp meter is
> expecting to see at full current.
Actually, I was one of the first to buy the kit and they sent me a +/- 50
amp gauge and shunt. I swapped out the gauge last weekend, but they
didn't include the correct shunt so I was still running on the old one.
The gauge was way out of calibration - charging at 12 amps (DC) showed
nearly 100 amps on the gauge.
> Let Mike Brown know about this -- he needs to send you the right
> part.
They sent a new shunt last night - I'll look at the part number carefully.
> (I wonder if that bolt on the left side might have been a little
> looser than the other one...)
I had the same thought.
Azure thinks the failing connection here might have been explaining some
of the weird data I saw coming out of the controller. We will see when I
replace the shunt...
- Randy