Saturday, November 29, 2014

Crankshaft and CAM signal faking



I'm trying to fake the crankshaft and CAM signal with an arduino based system (www.gevcu.org) according to Volvo's specification. The output of my signal generator (https://github.com/neuweiler/EngineSignalSimulator) is looking a lot like the one in Volvo's spec. But somehow the ECU is not happy with it yet. I get DTCs about alorhythm problems. On the right side you see the spec sheet and the output on the oscilloscope. In the spec the crankshaft signal is supposed to be a sinewave with voltage range from 0.1V to 100V according to the motor speed (the rpm sensor it's a magnetic inductor). But on other pages, it's said to be a sinewave from 0.1 to 5V. So I'm not sure if my timing is out of spec or if it's just the voltage which does not conform to the speed of the signal or if the ECU is not happy because it's a square wave and not a sine wave.
I think I need a reference signal..

Tuesday, November 25, 2014

Construction Time, Part II

With the transmission fixed, my spirits rose again and I spent more time in the shop, finishing the battery bay. No rocket science but an important step. Also the brackets mentioned in the video, which will mount the rear part of the bay to the car, are already done and ready to be mounted. After checking the distances of the batteries again, I found out that it should work within the tunnel and on the left side. But the ones on the right side of the car, exactly where I'm working on in the video, are most likely a bit too close and I have to re-do them - sacrifying a bit of the safety space towards the rear strut. With some insulating rubber or silicone it should still be allright though.
To prevent the batteries from moving up in the construction, I first thought of velcro straps. But I think I'll be placing a plastic sheet on top of the batteries and then fill the space above with expanding construction foam. This should sufficiently hold the batteries in place vertically. The major forces will be picked up by the aluminum construction.





Transmission Repair

Finally, I've cut my video on the transmission repair. It came out quite lenghty and is totally not EV related. But it might be interesting for some folks dealing with transmission problems. Without spoiling the story, I think I got very lucky and am thankful that it didn't get as expensive as I first expected. One of the next videos will be a report about mounting the motor and transmission and probably we get some wheels spinning :) Mwha ha ha !



Saturday, October 4, 2014

Construction Time

The construction of the "underbelly battery support structure" is progressing well now. In the video below you'll get a look at how I attack the topic which might give you an idea for your conversion. In the meantime I also messed up with my transmission. When I got the adaptor plate, the transmission shaft and the motor back, of course I had to put everything together and spin up the thing. It was exhilarating to also see the output of the differential spin when selecting a gear. I knew from Volvo's service manual that I'd have to take care so the differential doesn't start to move the wrong direction. So I only selected a gear with 500rpm. All went well the first time. The second time during a demonstration to some friends, I was distracted and forgot to switch back to neutral. At 2500rpm: klong - krrrrrrrrrrrrrrrchchchch.. I immediately pushed the emergency stop button but due to inertia and no regen applied, it took about 10-15 very long seconds until everything came to a standstill. To say I was downhearted was an understatement. Looking inside the differential you could see loose gear wheels and a rod I didn't remember was there before. Comments from other car aficionados didn't lift my spirit either: "Looks like everything fell apart..", "This looks wasted..", etc.
So I went to the Volvo dealer/shop of my choice to ask if they could help with repairing the tranny. Apparently OEM's no longer fix transmissions. They just replace it with a new one. When the supportive mechanic looked up the price in the computer - keep in mind that Volvo's service part are not really cheap - his puzzled face and a "whoops!" didn't help my mood much either: 4200.- USD. !! :( That's not an option for my already strained budget. Then there's a transmission shop close to Lucerne: Automaten-Meyer. Their initial estimate: 1200.- USD for the work only, no parts included. That's better but still, there has to be another way. The used trannies from wrecked Volvo's were also in the range of 1500.- USD - with 200'000km on them, that's just to much. In Germany I found a dealer which sells new tranmissions for 1700.- Euro. This after-market transmission would be my fall-back. I found some details about the inner life of such a manual transmission on the net: here and and here. But what really gave me the right kick, was the detailed report of a diff replacement in an M56 transmission. Mine is a M66 but it was close enough, the M56 is its predecessor and told to be quite simmilar. Thanks to this post, I thought: Well, I have nothing to loose, let's open the darn thing and see what's wrong with it, maybe with a huge bunch of luck, I'll be able to fix it. Bear in mind that if anybody told me that I'd have to open and try to repair the transmission during this conversion, I'd have never started it at all. Opening transmissions was a big taboo for me - for someone with no machanical education, a software developer. Yet, I opened it.. found out what happened.. and ? The conclusion will follow in the next video.
Now that I've got all major parts, there's only one but very important thing missing: Nordlock washers. Based on experience and reports from evtv.me, they are certainly a part I have to include in my conversion to secure the bolts in the batteries. As I have 120 cells with 2 poles and some other connection terminals, I will require about 260 pieces for M8 bolts. I will not do it otherwise. To my impression, these Nordlock washers are the only thing which ensure tight and reliable connection to the terminals. Non-secure power connections are cause of heat, increasing resistance, more heat and potentially fire. So no trade-off there. I just haven't got around to get my hands on them.


Sunday, June 29, 2014

High Voltage

I'm mainly building a mounting rail below the car for the batteries at the moment. I must admit, it's the task which taxes me most in this conversion. I like designing, doing electronics and software - but metal construction is a bit foreign to me and I find it tedious. Yet, I'm a bit proud of my construction. In the pictures on the side you see the T-shape rails in the exhaust tunnel and the almost finished construction. The third picture shows how I used a hollow aluminum tube (square) as a spacer to attach a rail. Inside is a threaded rod which attaches the construction very firmly to the former gas tank mounting. The rails will be bolted together in the lower right side of the picture. In the fourth and fifth picture you see an example how the batteries will be placed in the construction. The L-shaped rails from left to right will be bolted down to the T-shaped rails on the side with gibs. The L-shaped rails are close to the poles of the batteries but I think it is still safe this way. The batteries will be strapped down with velcro bands. Klaus AG is also helping me to create a mounting structure for the motor to the original motor mount. More on these things later (incl. a video for evtv).

Aside from that, I'm prepping up my HV stuff. Buying some missing things like another contactor for the heating system (Gigavac GV200), a maintenance switch (Gigavac HBD41a), some high voltage - low current fuses from EV West, a Polycarbonate case for an HV distribution box (Hoffman Q403013PCICC). As I require 4 bus bars with 5 connections each, I decided to create them myself and save 400 CHF. In the metal shop of my choice, where I got all the rails, I also found an offcut copper bar which suited my needs: 2.5cm x 1cm, cut down into 4 pieces of 12cm length - all for 35.- CHF. I'm going to drill 8mm holes, insert screws with a copper washer and cast it into polyurethane resin - two blocks colored red and two colored black. They will be put into th HV distribution box together with 3 contactors, 1 pre-charge relay and some fuses (see overhauled wiring-diagram).

I'm itchy on getting my motor back with the adapter plate - I need to to get some wheels spinning to boost my motivation.. I'm getting closer, I know, but after one year, I must admit, I'm looking forward to put an end to crawling around under the car. I want to sit in it for once... :)




Friday, May 30, 2014

Shaft & Batteries




The shaft to connect the motor with the transmission is ready and looking very nice! I'm looking forward to get the setup spinning. But right now the motor and transmission are still at Klaus AG because the motor still needs to be mounted to the original mounting bracket. We have to find a good approach to replace the huge prototype made of wood with a construction made of aluminum. This is currently my biggest worry.
The battery construction is making great progress and I'm confident to get a working approach. I started to replace parts of the wooden construction with T-shaped aluminum struts and now I'm able to determine where and how many batteries can be placed. In the exhaust pipe tunnel 17 cells find their place, in the gas tank area at least 45, in the area of the spare tire 22-24. This leaves only about 36 cells which have to placed in the engine compartment. The huge trunk will remain untouched - and this with 120 cells of CALB CA 100Ah. Most of the cells are placed at the lowest possible point in the car - resulting in a low center of gravity. I will create a movie to illustrate how I did it.
Also the wiring diagram is 90% complete. My version of the GEVCU will support 1 relay to connect the pre-charge resistor with HV+, 1 contactor to connect HV-, one contactor for HV+ and only relay to switch the enable signal for the controller. The sequence is like listed before: first the pre-charge relay closes, no current is running yet and no sparks are created in the relay. Then the HV- contactor closes and the capacitors are pre-charged. The contactor is hermetically sealed and thus less prone to sparks. When the pre-charge cycle is completed, the HV+ contactor closes and then the pre-charge relay opens. The HV part is ready to operate and thus the "enable" relay is closed to signal the controller and DC-DC converter to start their operation.
Next steps: Finish battery construction and boxes and the motor mounting. Then I'll be able to connect the cables and get it moving. Timeline? Hopefully within the next 1-2 months.




Oh yeah, almost forgot: I made another small video about the day of electromobility here in Zurich and about the EFORCE - a conversion of a truck:

Sunday, March 30, 2014

Scotty, we need warp speed!

From now on we're proceeding according to the Star Trek "miracle worker" principle:

"Scotty, how long do you need to fix the wrap drive?"
"The damage is extensive, captain. It will take 6 hours."
"Scotty, I'll give you 4!"
"Ok captain, for you I'll do it in 2!"


According to the speed things were moving in the last 6 months, it would have taken me another 6 to get the car moving again. But with some luck one of the companies in the Swiss Automotive Group (my new employer) was able to help me out of my biggest conversion misery: Some very friendly and supportive guys in Klaus AG provided the kind of help that Scotty usually does. Within 14 days, a great and robust looking adapter plate was fabricated out of a 32kg aluminium block. While neither Volvo nor Ford could provide me the dimensions of the transmission, the guys in Klaus AG found a great approach to make a custom fit. They also came up with a better idea for the coupler. As the Brusa motor has an involuted spline shaft, I was chasing an idea with a shaft, a taper lock and some adapter ring to mount part of the clutch disk to it (see previous post). The guys at Klaus AG had some doubts about a possible imbalance in the shaft and that the resulting wobble might have damaged the bearings of the motor or transmission in the long run. So with the help of the company Okey, we decided to create a custom shaft. One piece that fits perfectly in the motor and can be inserted right over the transmission shaft. Should be delivered in week 18. A bit more expensive, but if it might save my motor, I'm ok with it. At least I'll be able to finally spin some wheels in week 18 ! :)


So with this problem almost solved, I only have two challenges keeping me from a first drive: the motor mount and the battery mount.
Inspired by the good news on the adapter, I had to move on with the battery mount. To keep the center of weight as low as possible and to save as much room as possible, I plan to use the tunnel of the exhaust pipe, the area of the gas tank and the spare wheel box before placing any cell in the engine compartment or the trunk. This will result not in one or two battery boxes but some special fabrication. I started with the cardboard box approach first as suggested by others. But I quickly found that it's just too soft and not fitting my requirements. As my brother-in-law is a carpenter and his shop is right next door, I started doing some template mountings out of unused wood. By using this wood, I found out how I can use stock mounting points in the car for a frame that holds about 40 cells under the car. In place of the wooden frame a metal construction out of L- or T-shaped bars will be used. They'll give the cells lateral and axial fixation. The cells will be lying on
the side and strapped down to the frame. E.g. in the tunnel, I'll be able to place 3 pieces of 100Ah CALB CA cells per location. Three sides of the "battery box" will be provided by the car, so I only need to fabricate a insulated cover to protect the cells from the elements. I could even run a pipe, to be able to heat the cells and charge them at outside temperatures below zero degrees Celsius. The cells will be accessible for maintenance by removing the bottom cover. There's a disadvantage for this approach though: Instead of connecting many cells with straps, I'll have to do a lot of cables running from one block of cells to the next.


When the adapter plate was ready, I had to stop my battery placement invention cycle. The parts were quickly assembled and the combination of transmission, motor and adapter plate was placed on a board with wheels (again, thanks to my brother-in-law) and placed under the engine compartment. By lowering the car lift carefully, the parts slid right into place. After 2 hours adjusting the angle and height of the transmission, it was mounted in its original place. By adjusting the height of the lift again, I aligned the other end - the motor - level to the car frame. Instead of mounting the motor directly to the car frame, I want to re-use the three stock engine mounts to keep things flexible and able to absorb shocks. The one on the left used to carry the major part of the weight and gave also some stability on torsional forces. The one at the bottom of the frame carried a mounting bracket for the long wheel shaft and was also likely the main actor in picking up torsional forces. Sitting there and looking at the various strange angles, I tried for hours to figure out a way to mount the parts. Then I decided to continue with wood. After 8 hours I created a make-shift construction which I think could carry the weight and the torsion forces (when made out of metal of course).

So if all goes well, I might be able to stick with my original target and get the car moving this spring (although I originally planned for the beginning of spring, not the end).

Monday, January 27, 2014

Brain training

I think I've found a possible approach to solve the issue with the coupler / adapter for the motor and transmission:
  1. I require a shaft for the motor. On one side it has to be splined (the half that goes inside the motor) and smooth on the other side with a key shaft (the part that sticks out).
  2. On the part of the shaft that is outside the motor, I'll attach a taper lock (aka quick detach bushing).
  3. A washer between the motor and the taper lock will keep the turning parts away from the Motor front.
  4. On the transmission side, I'll use the inner part of the clutch disk to attach a machined ring to it.
  5. The machined ring and the taper lock will be connected with bolts.
  6. A 20mm thick adapter plate will be machined for the motor and one 20mm thick plate will be machined to fit the bell housing of the transmission - both aluminum.
  7. The motor will be mounted to the transmission, both placed upright and then I'll spin the motor at different speeds and with minor adjustments, I'll try to get rid of misalignments / vibrations.
  8. Once the vibrations are (hopefully) gone, the two plates are clamped together, some holes will be drilled and the plates bolted together to lock them in place.
The taper lock on the motor shaft and the washer will make sure, the shaft doesn't move inside the motor. The motor shaft touching the transmission shaft ensures that it doesn't move out of the motor. So the coupler is fixed laterally although it's not tightly attached to the transmission shaft. There's some room for imperfection because the nut in the clutch disk isn't a perfect fit on the transmission shaft. So the whole connection is not a tight fit between the motor and the transmission. Hopefully this prevents damage to the motor's bearings and splined shaft.

So the only custom built parts should be the motor shaft and the adaptor ring. But as they're pretty simple, I think it won't be too expensive. Hopefully I can get the measurements of the bell house from Ford. This would help a lot to align the two parts precisely.

Saturday, January 18, 2014

It's cold in the garage

It's been awfully quiet here for a while - my apologies. I was changing jobs and going on a 6 week holiday in Australia with my family. This all needed some preparation and consumed some of my limited spare time.



So, in the meantime I was mainly doing dome rust treatment in the engine compartment. Wait a moment, rust in a 5 year old Volvo? Well, yes.. while working under the car it slowly dawned on me that I was working on a Ford and not on a Volvo. Almost every part of the frame is labelled "FoMoCo" - Ford Motor Company. And guess which were the only parts which had rust on them? Yes... kind of disappointing. But at least they stand up to it by labelling them.
I used a wire brush in a drill and removed the rust as good as possible. In accessible areas down to the bare and shiny metal. Then I applied 2-3
layers of Noverox (rust converter/primer) and 2-3 layers of black color.

Aside from that I was playing with the electronics of the car. It turned out that it was fairly easy to trick the ECU into not reporting any error on the dash. When I connected the throttle control unit (motor with two pots which report the throttle position), I only had to switch on the ignition and the throttle opened and closed completely when depressing the throttle pedal. Perfect, that's all I wanted. It looks all too good at the moment but my hopes raised considerably that I'll be able to use the CAN bus signals to control the power output of the electric motor - hence drive the car with precision throttle input and the ability to use stock cruise control. So I finalized the GEVCU module which requests the throttle position from the ECU and processes the CAN bus signal into another CAN message which controls the Brusa inverter. It all works fine - with the car on the lift.
Oh yes, and while it's hanging on there, more and more dust accumulates on the surface. It almost looks like one of those long forgotten vintage cars in an old shed. I have to move on a bit quicker. I want to get it rolling again until spring time.

One thing I discovered while playing and researching was that the generator of the ICE communicates via LIN with the ECU. As soon as the generator is ready to power the car, it signals the ECU and the ECU in turn notifies the climate control system that it's ok to enable consumers like seat heating, supplementary heating and rear window de-mist. As long as the signal is missing, none of these appliances will work. So either I have to find the CAN message spoof it or I have to immitate the LIN communication from the generator. Either way won't be that easy, I suspect. If anyone knows more about the LIN signals or how to connect a small microprocessor to the ECU, I'm all ears and would appreate some help.

Last but not least, my batteries arrived - yeeehaaa !! I ordered them from www.lithiumstorage.eu which is located in Zurich Switzerland, not more than 40km away from my garage. Good price and great service. I ordered 120 CALB CA100 cells and they arrived within a week at my doorstep. No import taxes and transport was almost free compared to international options. So one of those days I came home from work and there it was, the stack of 6 boxes with a net weight of 408kg. Holy moly! How am I going to fit this in the car ?!? :) Well, some have to go into the engine compartment, some in the exhaust tunnel under the car, some where the gas tank was and some where the spare tire would be.
If this is not enough, the rest will eat up some space in the trunk. There's going to be a lot of construction work to be done for the battery boxes.

Another thing that keeps my mind occupied is the adaptor for the motor. Originally I planned to attach the stock flywheel and clutch onto the motor - as it is most common for custom conversions. But I got serious doubts that it's a good idea to let the bearings in the Brusa motor to support the 20kg of the clutch. There'd be quite some shearing forces without the wheel supported outside the motor. Christian from www.turn-e.de told me that he did the conversion without a clutch and now uses the clutch pedal to control regen. What a great idea! I was always thinking about where to
put a lever for additional regen. Dumping the clutch also saves me from aligning it horizontally to get it working. My currently favoured approach are two-plates: one 2cm thick which perfectly fits the motor, the other als 2cm thick which fits onto the gearbox. The gearbox has two precision dowels to position the ICE. This second plate is made to fit the dowels and lock it into relative position to the gearbox. Then we will place the motor with its plate on top of the other plate (vertically) and let it run. With small adjustments it should be possible to eliminate all vibration caused by misalignment. Once the motor and the gearbox are aligned, the plates are clamped together and then locked down with a few very strong bolts. This way it should be possible to remove the motor and precisely put it back in place again without much effort.
The adapter between the involuted spline shaft of the motor and the shaft of the gearbox is going to be another challenge. I'm thinking about two shafts - one in the motor and one around the gearbox-shaft. They are connected via a flange. But honestly, I'm a bit at a loss here. Anyone got some experience with this or a hint?

Not much know-how to share for the moment.. I hope this changes in the next post.