Sunday, October 21, 2007

Hair Dryers Installed

I took a few hours today to get the small hairdryers installed.

In order to have a convenient switch to activate the hairdryers, I hijacked the defroster knob switch. I was lucky enough to have this option on my car. The wires from the defroster knob sit in their own plastic tube, so it was easy to find the yellow/red wire going towards the center wiring column of the car under the clutch pedal. This wire provides 12V when activated, just the thing I need to fire up the 12V DC relay.

Here's the travel hair-dryer attached to the heater intake on the passenger side of the car. The coupling is just short enough to let the hairdryer not interfere with the hood spring. The pipe clamps make the whole thing very snug.

Here's the DC relay mounted on the firewall just behind the middle battery box above the foot pedals. I used two 10-32 rivnuts to bolt this thing in. The two hair-dryers are tied into the terminals on the left. The negative battery cable is on the lower right (black wires) and the upper right comes from the positive battery post in the fuel compartment with an inline 10 amp blade fuse.

Remember the yellow/red wire from under the dash in the top picture? It's tied into the top coil terminal on the relay here. Since the chassis of the car is ground, I simply tied the bottom coil terminal to the mounting bolt (small blue wire).

For testing, I started with one hair dryer on the "low" setting with 5-amp fuse. It ran well (albeit loudly) and provided a slight warming into the cockpit. If I put the hairdryer on "high," the 5-amp fuse blows immediately. The hairdryer is very loud on "high" so I put both hair dryers on their "low" settings and used a 10-amp fuse. I suppose in desperate move to heat the cockpit, I could put in a 20-amp fuse and run one hair dryer on "high" since the high setting pulls about 15 amps. I'll try the "low" settings for now and see how it goes.

Next up: experimenting with an optocoupler to capture voltage for the "fuel gauge"

Friday, October 19, 2007

Serial Port and Hood Pin

Today was a bit slow. I received some AMP pins in the mail that fit the special connectors on the DMOC445 controller. To make the serial port more robust, I crimped the pins onto the ends of the 16-gauge wires and inserted them into the 8-position connector that came with the DMOC.

Here's the finished connector. It's far more reliable than the simple sockets I had soldered to the end of the cable. See this post for the prior implementation.

I also took a little time to install the hood pin that I got from Schucks yesterday. Drilling a rectangular hold through the engine compartment lid was very tricky. I covered up most of the rough edges with the plate that comes with the hood pin. At least it's a lot nicer than the block of wood I was using before.

Just for grins, I fitted one of the hair dryers to the passenger side heating duct and plugged into the wall outlet to see how well it would heat the car. It's rather loud, but produces a light stream of warm air. I wouldn't use this to fully heat the car, but it'll probably make a decent defroster.

Tuesday, October 16, 2007

Dedicated Charging Outlet and Hairdryer Coupling

After sleeping for 14 hours last night, I think I'm somewhat recovered from that nasty cold. I dropped by Parkrose hardware after work and picked up some stuff to install in the EV.

Since I have a gas range, I replaced the 240V breaker for the non-existant electric range with a 20-amp 120V breaker dedicated to charging the 914 EV.

I purchased two small travel hair dryers over the weekend and found some plumbing couplings that seem to fit quite well on both the hair-dryer and the heating system input. I forgot to bring the relay to the hardware store, so I'll probably get the electrical connectors tomorrow and install the system over the weekend.

I also ordered some of the custom sockets that go inside the 8-pin DMOC connector so that the serial port will be much more of a permanent installation instead of just sort of dangling out of the controller.

Big thanks to Ross Cunniff for pointing out that hood pins are available at the local auto-parts store. I searched high and low for these at the regular hardware store but didn't think to look in the "bling" (ahem, I mean performance) section of the auto store.

Monday, October 15, 2007

Running the AC24 at High RPM - maybe not

I'm still big-time sick today, but recovering quickly. In my efforts to understand the torque of the AC24 motor a bit more, I looked at the datasheet online at Azure Dynamics:

Hmm, it looks like the optimal operating point is really around 4500 RPM and torque drops off quite a bit at higher revs than that. In my frenzy to go on the freeway in first gear, I didn't really look at the operating characteristics of the motor. Also spinning the input shaft of the transmission above 4500 RPM continuously might wear it out sooner. Oh well, back to bed to recover...

Saturday, October 13, 2007

Initial Fuel Gauge Driver Schematic

Since I'm stuck indoors today to recover from a sore throat, I put together the following schematic to drive the fuel gauge based on the pack voltage.

(Click to enlarge the image)

With the exception of the 4N25 opto-isolator, which I salvaged from a dead UPS, all parts are available at Radio Shack. I hope to implement this soon. Comments and feedback are welcome.

Friday, October 12, 2007

Finishing the Serial Cable and Testing the Fuel Gauge

I had a bit more time today to work on some other small projects.

Here's the 914 patiently waiting for its rebuilt transmission from Rennsport Systems. It'll be awhile since Rennsport won't start on the rebuild until November.

Up until this point, the serial cable from the DMOC445 AC controller to the passenger seat was routed along the outside of the car. I cleaned this up today by taking out the two seats, pulling forward the upholstery in front of the firewall and drilling a hole for the serial cable. For some reason, I had an extra cap-plug left over from the kit, so I used it to protect the cable from the sharp metal edges. The cable now runs neatly behind the seats and into the central armrest.

I got really upset when putting the seats back in because I dinged the paint near the door handles with the sharp edges from the seat rails. I'll patch this up with touch-up paint, but I'm still frustrated with my clumsiness. Working in this small single-car garage is a challenge.

I also started researching how to drive the fuel gauge with the voltage on the batteries. The fuel gauge reads the resistance of the fuel sensor which varies from zero ohms (full) to 75 ohms (empty). I verified this by shorting out the sensor circuit and putting 75 ohms in the circuit. The system pulls 100 milliamps at zero ohms and 50 milliamps at 75 ohms.

For you electrical engineers out there, I'll probably implement a circuit with a 4N25 optoisolator to measure the voltage on the main battery pack and drive the fuel gauge with an op-amp controlling current through a transistor in emitter-follower configuration. I'll post the schematic here after I get it up and running. There's also a fuel-empty light that I can drive if the battery voltage gets too low or the voltage droops because I've accelerated too long and drawn down the main pack voltage. Any suggestions on how to make this circuit work are welcome.

I started getting a sore throat today, so I'm going to bed early and will probably take it easy tomorrow by working on the fuel gauge circuit. I'd really like to get the hair dryers installed soon so I can drive in the colder weather.

Cheers and good night...

Finishing the Rain Gutter for the Engine Compartment

The major project of the day was finishing the rain gutter under the engine compartment lid. The original gutter was too deep and interfered with the rear battery box. As described in a previous post, I purchased a piece of ABS plastic from TAP plastics and bent the edges to make a rain catch.

Here's the plastic that I bent a few weeks ago. ABS plastic has the nasty property of deteriorating quickly in sunlight, so I spray painted it with three coats of Krylon black satin paint to protect the plastic. I also drilled holes where the lid mounting bolts were and cut out the bottoms of the drain wells on either ends.

Here's the plastic installed on the underside of the engine compartment lid. There are three bolts on the top edge that hold it on. The bottom edge is held on by sliding it under the horizontal support beam. I used silicone caulk to seal the three bolts to prevent leaks.

Since the new rain gutter sits much higher than the old one, I added some cheap 99 cent funnels from the hardware store and cut them so that the top was level with the ground. The larger funnel input area also allows the downspouts on the gutter plastic to be slightly off center in case I goofed up a bit.

Here's the engine compartment lid re-installed. I also cemented two small rectangles of plastic under the mesh to either side (see small red C-clamps) to help drain rain away from the circuitry in the engine compartment.

After the cement on the small side pieces was dry, I was able to close the hood as shown above. ElectroAuto still hasn't delivered the hood pins and I suspect that I'll never see them at this point, so I'm still using a small square of plywood to hold the lid shut.

Wednesday, October 10, 2007

Giving up the Tranny to a Higher Mechanic

After receiving admonishment from two mechanics about driving around with a resonant frequency, I dropped the transmission out of the 914 and gave it to the experts at Rennsport Systems this morning for a precision overhaul. This is the fourth time dropping the motor/tranny, so I was able to do the whole process in less than an hour. Giving up the tranny was an ego battle for me because I had spent so much time rebuilding the transmission in the first place.

Rennsport deals with racing Porsches, so I suspect I insulted their intelligence when I said the electric motor only puts out 72 ft/lbs of torque and rattled at a "low" RPM of 5400. Anyway, my time is getting more valuable these days and I just want this thing working. They said they'd get to it at the start of November. I've got hairdryers and a fuel gauge to work on in the meantime, possibly even an MP3 player...

Monday, October 8, 2007

On the Road Again

I started commuting to work again today. I think the transmission is making slightly more noise than before but otherwise the car is fine. The RPMs are limited to 5300 RPM, so I won't have any resonant activity. I'm really happy that the tachometer is hooked up. It really helps me keep tabs on what happens to the motor/transmission at different speeds under different driving loads. It's also much easier to watch than the speedometer hidden behind the right edge of the steering wheel.

The car also seems to be unaffected (so far) by driving in the rain. We'll see how long that lasts. I do have issues if I get in the car all wet from running through the rain, since my body heat evaporates the water and fogs up the windshield. Turning up the vent fan on full helps quite a bit, but having the hair dryers will definitely help.

I called Rennsport Systems to ask for a transmission rebuilt quote and they suggested budgeting between $2000 and $3000 for parts and labor (Ouch!). Alan at A &P quoted about $425 for labor plus parts (better). I could also get a transmission off e-bay for about $375 plus shipping. With the spare parts I can salvage from the second transmission, I'm thinking that having Alan do a rebuild would be the best thing.

My plan is to drive the car for a week, show it off at the EV meeting on Thursday and take it in to Alan on Friday (if he has time) to upgrade the brake cylinder and align the front end. We can chat about the best direction for the transmission from there.

Perhaps I'll work on the fuel gauge in the meantime.

Saturday, October 6, 2007

Thoughts at the End of the Day

Well, the car is all back together again. I really can't say it's any better than before; however, I've learned a tremendous amount in the process. After balancing the flywheel twice, adding rubber grommets, replacing the transmission mounts and putting in a new pilot bearing in, I still can't get the vibrations to go away or diminish much.

Given that the second transmission didn't vibrate much at all, even over 7000 RPM, I'm really suspecting the existing transmission has a bent shaft in it or something. There's definitely 10 mils of runout on the input shaft, but I didn't think that would be a big deal. If I look back over my transmission rebuild at the link here, I remember that the bearings in the transmission were shattered. I never really understood why and just replaced the bearings. Perhaps one or both of the shafts were bent already and caused the bearings to break. I'm also guessing that running the transmission through its natural resonant frequency several times over the past week wasn't good for it either. I might be able to salvage a straighter shaft from the second transmission, but I don't know if they are different between the side-shifter and the tail shifter version. The next step is to drive around a bit more and see what happens. Perhaps the transmission will fail, but who knows.

Since the RPM gauge and indicator lights are working, I'm thinking about driving the fuel gauge with an opto-isolated circuit from the 144V pack. It needs opto-isolation so that there are no connections between the 144V system and the 12V system. I'll post that circuit if and when I get it working. Best wishes to all the other EV folks working hard out there to get their cars running.

Replacing the Pilot Bearing

I purchased a new pilot bearing for the flywheel to see if replacing it would help with the rattles.

Flywheel - it's what's for dinner! I've found during my transmission rebuild that heating up components in the oven (200 F) sometimes helps with removing the bearings, so I heated up the flywheel prior to replacing the pilot bearing. I also put the new pilot bearing in the freezer to make it smaller.

Here's the heated flywheel, clutch side up. I put a 9/16" socket on the pilot bearing and placed a steel block over that to even out the pushing forces. After a few good whacks, the bearing popped out.

Here's the re-installation. Again, I put the new bearing in the freezer to help it go into the hot flywheel easier. I first tried to line the new bearing up and used the steel block to keep it straight while I tapped it in. I used the 9/16" socket again to finish the job.

I searched online to see which way the bearing went in. The one comment I found online said that it didn't matter; however, if I look at the bearing, one side is clearly wider to present the needle rollers to the incoming shaft. I chose to install the bearing with the embossed text on the rear (non-clutch) side to see if that would help.

Getting the Tachometer and Lights Working

After receiving an updated .css file (ccShell config file) from Azure Dynamics for the DMOC445, I was able to access the EE1SpeedoDiv variable. After setting this to a non-zero value, I attached a scope to pin 25 of the main DMOC connector and (yay!) got a pulse train corresponding to the RPM of the motor.

Here's the pulse train on the oscilloscope. It has a 50% duty cycle and swings the full 12V range. This makes it ideal to drive the 914 tachometer.

Since pin 25 wasn't one of the original wires brought out of the wiring harness by Electro Automotive, I had to modify the DMOC445 connector. This proved to be easier than I thought. Since I had modified the regenerative braking system to not drive the brake lights, I had three extra wires left over. I'll use these to tap into DMOC monitoring functions to drive the tachometer and the red/green indicator lights on the dash.

Here's the main DMOC connector removed from the DMOC. To open up this connector, I used an icepick to lightly bend aside the two small black holding tabs on either end. It was then relatively easy to pop off the red cover.

Here's the same connector with the red cover popped off. You can see the pins that were previously installed by Electro Auto. If you use an ice pick to pry apart the two holding tabs under any pin, that pin will release and slide out the back of the connector.

With my prior modifications, I didn't need the blue, yellow, black/yellow or brown/blue wires, so I pulled them out of the connector and put them back in to new places. Note that your colors may be different from mine. I kept the black/yellow line at pin 19 to act as a ground. I put the blue wire to pin 11 (the READY led) and the yellow wire to pin 34 (the FAULT led). The brown/blue wire went to pin 25 (SPEEDO_BUF) to drive the tachometer.

Now, I was concerned that the tachometer would pull way too much current and possibly short out the DMOC driving signal, so I created a simple 555 timer circuit to drive the tach input signal (black/purple wire in the main wiring harness). The 555 timer doesn't drive that hard and I didn't see any degradation in its output signal. Given that, I simply tied the SPEED_BUF signal from the DMOC directly into the tachometer input wire and (voila!) the tach started working!

Here's the tachometer at around 2600 RPM. I had to adjust the EE1SpeedoDiv parameter inside the DMOC so that the tachometer matched the ISR2Hertz RPM value shown on the laptop. My calibration value for EE1SpeedDiv was 70, but that might vary with different gauges.

Next up was tying the READY and FAULT output lines into the green oil light indicator and red generator light indicator. The original kit doesn't use the red light and the green light is simply grounded. The READY/FAULT output signals on the DMOC are unbuffered 3.3V digital logic outputs. Since the indicators on the dash already have one side tied to the 12V supply, I simply used a 2N2222 transistor (buy a pack of them at Radio Shack) to drive the two indicators. The 3.3V DMOC signals drive the base of the transistor through a 1K ohm resistor. The collector is tied directly to the indicator light and the emitter is tied to ground. When the signal goes high, the transistor turns on, pulling current down through the indicator bulb.

Here's my simple test setup. I just used simple alligator clips to attach everything. By trying different resistor values, I found a reasonable base drive resistor empirically that doesn't load down the DMOC signal but still provides enough current to turn on the transistor.

After making the test circuit, I took a breadboard from Radio Shack and soldered all the components together with some quick-connect solder connectors. The circuit worked fine and the two dashboard indicators now show all the codes that were listed in a previous post.

Here's a simple schematic to show how the transistors and tachometer are hooked up. Click to enlarge the image.

Next up: replacing the pilot bearing in the flywheel and putting the whole thing back together.

Friday, October 5, 2007

Trying a Different Transmission

Last week Craig at Camp914 offered to lend me another transmission to see if I had vibrational issues.

I drove out to Camp914 this morning. I wish to express much gratitude to Craig for letting me borrow this spare tranny. Since it's an older tail-shifter and many of the external pieces are missing, he even gave it to me.

Here's my test setup again. The working transmission is tied to the motor at left and resonates at 5400 RPM.

Here's the alternate transmission. Again, I don't have the clutch friction disk installed, so the only thing that the flywheel is touching is the input shaft inside the pilot bearing.

The moment of truth: After spinning up the system past 7000 RPM, I got no vibration whatsoever. Rats, there must be something up with my own transmission. I suspect that there's something up with looseness in the transmission shafts or perhaps rattling in the pilot bearing. I got a new pilot bearing yesterday so I'll see if that helps. It looks like this is an isolated case and not everyone will have this resonance. I wish all the other 914 AC kit owners luck.

Just for yucks, I pulled out the lower transmission fluid plug on the second transmission. The lower plug has a built-in magnet to pull out all the metal shavings floating around in the fluid. This plug was completely packed with metal shavings. One of the common hacks you can do if you have a disposable second transmission is to put the bottom plug from the second transmission in the top hole of the first. That way you have two magnets pulling metal shavings out of the fluid for you. This plug also had a hex head on it (instead of a hex depression) which makes it much easier to unscrew when you need to replace the transmission fluid. I think I'll keep it...

Next up: Getting the tachometer and warning lights on the dash tied into the DMOC445 controller.

Note: Ross Cunniff is burning the midnight oil over at
to try and get his 914 AC done this weekend. I wish Ross the best in getting the second 914 AC kit on the road!

Monday, October 1, 2007

Getting Expert Advice on Resonance

After spending most of yesterday researching machine resonance, I chatted with one of the mechanical engineers at work today about the resonance problem. After drawing a picture of the motor/transmission assembly, it became rather obvious to both of us that we have the equivalent of a vibrating violin string. The motor/transmission is attached to the car somewhat rigidly at both ends. The main excitation mass in the system is the flywheel, which sits very nicely right in the middle of the structure. At the natural resonant frequency of the system, the tiny vibrations from the flywheel shake the center of the assembly up and down, just like passing a violin bow over a string.

Since the natural frequency of the system is sqrt(stiffness/mass) and mass is rather hard to change, he suggested adding stiff support bars to the bolts on the adapter plate and bolting the other ends of the bars to strong parts of the car chassis. This would increase the stiffness factor and raise the resonant frequency. This is similar to placing your finger in the middle of a vibrating violin string and having the frequency jump up an octave. The string still vibrates, just at a much higher frequency.

He did lend me an accelerometer which I tied to my analog oscilloscope to measure the three degrees of motion on the motor housing. I didn't get very accurate measurements because I didn't want to hold the system in resonance for too long, but it was good to see that the system underwent 5-6 Gs (5-6 times the force of gravity) during its oscillation at resonance.

On a slightly different note, I also probed the encoder signals inside the position feedback box attached to the DMOC445. In case I can't get a speedometer signal out of the DMOC, I just might risk tapping into the position encoder. It seems like a much easier idea than adding an opto sensor on the flywheel. The pulses coming out of the position box are close to a 50% duty cycle and we get 64 pulses per revolution of the flywheel. This is not surprising since the EE1EncoderPulses variable in the DMOC445 is 64.

At this point, I'm thinking of giving up and just limiting the RPM to 5400 and looking at ways to add support bars to the motor adapter plate later. I just want to get this thing going again.