Winter Projects

[Guest mjg86]

Hi All

I haven't posted on the forum before but have been working hard on some interesting winter projects so thought I would share the experience and lessons learned! I have 2B running a stock 2l pinto which I built with my Dad (who also has one) which was IVA'd in Dec 2013 and hit the roads last summer. Had a great couple of months showing it off but after a grand total of 700 miles and with the cold weather approaching I decided it was time to return it to the workshop in search of a bit more power!

I live near London and whilst I have a garage just about bit enough to squeeze the 2B into, doing any proper work isn't really feasible and so the car went back to my Dad's workshop which unfortunately is a 2hr drive away. Because of the distance it limited me to working on the car just on weekends and not every weekend and so a big project needing a lot of man hours i.e. an engine change wasn't really going to be possible. To get the biggest performance gains with limited time and money I decided to go down the route of upgrading the camshaft and moving to bike carbs. For the camshaft the Kent FR32 seemed like the best option, giving a decent power increase for normal road driving. Fitting the cam was the most time consuming part of the upgrades as although the engine stayed in the car it did involve a lot of stripping down to get access to the head which also came off for a bit of a clean up. Swapping the cam and setting the timing correctly was a bit tricky although luckily Dad had done it before, if not I would probably still be working it out! We fitted a vernier cam which was pretty expensive although allowed us to achieve the best timing accuracy possible and so get the most out of the new cam.

Swapping over to the bike carbs was pretty simple in comparison as I bought complete kit from DanST which included ZXR1200 40mm carbs, custom manifold, trumpets, air filter & baseplate and all the fixtures / fittings. Fitting it was pretty straight forward although I didn't realise when I started this part of the work that the pinto fuel pump wouldn't be suitable as the bike carbs need a low pressure supply. This was rectified by fitting a ZX9R fuel pump source from eBay. Getting the carbs up and running was fairly easy although the float valves had ceased due to them being sat dry for a long period of time so the carb did spray fuel all over the garage when it was first primed! The fix was pretty easy though, the carbs came off, float bowls taken apart and a bit of pressure put on the float got them unstuck and moving again. Once the carbs were re-fitted the engine started first time so was really pleased and the noise it makes put a big smile on my face

When driving the car over the summer I found the throttle response not to be very sharp or consistent, it seemed very flat at certain rev ranges and better at others and I attribute this to the ignition timing and so as part of these upgrades I also wanted to move to 3D ignition mapping. I hoped the 3D mapping would give a better response and improve the feel of the car. I had a look at the options available which seemed to come down to the Nodiz and the MegaJolt. The Nodiz seemed pretty expensive at ~£300 for what it did and the MegaJolt was not a huge amount cheaper, also the build your own option for the megaJolt seems to have disappeared? The cost of the units plus the costs of the trigger wheel, crank position sensor, TPS, coil packs and EDIS module (needed for megaJolt, ~£60 for a refurbished one on eBay) made neither of them seem like a very good option, which got me thinking how hard could it be to make my own!?!

Fundamentally an ignition controller only ECU is not a very complicated thing, it calculates the engines rpm and keeps track of its absolute angle from a signal from the crank position sensor and measures load on the engine either from a throttle position sensor or a MAP sensor. It then uses these to calculate a current advance value from a lookup table which you provide and then generates a spark at the correct position. I am an electronics engineer so had a few bits a pieces lying around and after a couple of hours messing around one evening I convinced myself that it would be possible and not overly complicated to make one myself so I got busy designing.

The first stage was the circuit and PCB design which was done in a couple of evenings. This bit was the most worrying as due to the number of surface mount components I couldn't really prototype anything easily before hand so I had to take the plunge and get one made hoping that everything would work as expected. The PCB arrived in a couple of weeks and it didn't take long to solder everything in place and to my surprise it powered up with no smoke or bangs which was very good news. I set up some other kit I had available at work to act as a sort of engine simulator where I could generate a simulated output from the crank position sensor and TPS and so got on with writing the code which would calculate the engine position from the crank sensor as this was the most complicated bit. After a couple of attempts I ended up with an algorithm that seemed pretty robust and would very accurately keep track of the engine angle at rpms between 5 (low rpm function required for starting the engine) and 10000. After getting this working I then put in the function the read a position and lookup an advance from a 3D table and wrote the software which would allow me to configure and map the vehicle once it was fitted to the car.

To get the ignition controller working on the car I did need to fit a 36-1 trigger wheel again sourced from DanST, a crank position sensor, a coil pack and made some modifications to the wiring loom to integrate all of these. These parts were all easy to source from eBay. So this weekend was the first real test of the ECU as it was connected into the loom and sensors. Initially I ran the system still using the distributor and checked that my ECU was behaving as expected. After plenty of test runs I had enough confidence to switch over the engine to the ECU and disconnected the distributor and was extremely pleased when I turned the key and the engine started! I repeated this a couple of times and the ECU seems to be working nicely, starting the engine very quickly and holding a nice idle.I just had a flat map in as a default with 10 deg advance for all rpm/load and so now need to look at getting hold of a better basic map to do some more testing and get the car out on the road. If anybody has any maps that might be suitable that would be really helpful?

Overall I am really pleased with the upgrades in particular the ignition controller as a lot of time went into the design, build and testing. Its also worth noting that the whole ignition setup ended up costing >£100 including the ECU, trigger wheel, sensors and loom. I have attached a link below to few pictures taken over the course of these upgrades.

https://www.dropbox.com/sc/5jx65al6m2ko5sz/AAB1CEiGHANjN599HmzeAr73a

I would be really pleased to hear any comments/questions and can't wait to get the car back on the road and along to a couple of events this year. Also I would be keen to see if there is any interest in the ECU / Ignition Controller as it would be great to see the design being used if there are any willing testers / guinea pigs out there?

[kevin the chicken]

Very clever stuff. Electrics and electronics are a mystery to me I'm afraid. I am sure there will be somebody interested to help you test out there.

[Peter Bell]

That sounds very interesting I had been wondering about megajolt but it seemed an expensive way of replacing a distributor. Not much help for a map I'm afraid but here are the advance figures for my distributor without a vac advance if thats any use. I have a piper 285 cam

Distributor advance, Bosch Hall effect.

RPM Advance °

1000 14

1500 18

2000 21

2500 24

3000 27

3500 30

4000 33

[fry61]

Some-one hiding in the top right corner of the Hants & Dorset area AND who can make electric smoke & mirrors work how they wish, must meet some-time soon.

 

The April meet for us is at Popham Airfield on the 9th, will you be mobile in the 2B then?

[Guest mjg86]

Unlikely to have it back on the road by the 9th as I'm limited to working on it at the weekends and this weekend is being spent at Brands Hatch watching the BTCC. Also as you can see from the later pictures most of the loom is still outside of the car so still plenty to get done! When/where is the following meet?

[Fossett]

Awesome Job!

[theduck]

Is the ecu completely your own design or is it based around something like megasjolt?

[brumster]

Looks his own. What microcontroller is that, Atmel?

[fry61]

Unlikely to have it back on the road by the 9th as I'm limited to working on it at the weekends and this weekend is being spent at Brands Hatch watching the BTCC. Also as you can see from the later pictures most of the loom is still outside of the car so still plenty to get done! When/where is the following meet?

 

May is Stoneleigh & then Wales with Nick (Wales/Ireland/IOM Sec ) following those two we sail for Spain ( to interrupt their bowls ? )

[Guest mjg86]

ECU is my own design, is based around a 32-bit PIC microcontroller.

 

Will definitely try and get the car along to one of those events, at least the UK based ones!

[Mrbarry]

Has been some time since I played with a PIC. Brings back memories!

[brumster]

Did you find your power supply was good against the noise of the car first time out, or did it take some development? Or did you have previous experience of automotive power supply design? Or did you base the design of that stage on something existing?

[richardm6994]

I'm reading this in complete awe of your electrickery skills!!!

The ecu would make a really good article for the club mag......

[Derek]

(The ecu would make a really good article for the club mag......)

 

Yes please?

 

PM me if you need any details or help, word is great for the text, please send photo's separately just put a note in the text (Photo 1 here)

 

Cheers

 

Derek

[Guest mjg86]

 

 

Did you find your power supply was good against the noise of the car first time out, or did it take some development? Or did you have previous experience of automotive power supply design? Or did you base the design of that stage on something existing?

 

I haven't designed this sort of power supply before but went into it assuming it would be a pretty harsh environment in terms of the noise, spikes and variability of the supply so designed against this. The signal from the crank position sensor is tiny at low rpm i.e. when cranking the engine, was seeing around 20mVp-p so which could easily be swamped by noise on the supply from driving the starter motor so was worried about this but the ECU seemed to work well and lock on to the engine's positions quickly. I did do the complete design of the ECU, didn't base it on anything existing, that would be cheating!

 

Can put something together for the magazine although it might be a couple of weeks until the car is finished so wait for that to get some end of build photos.