|
Post by buster on Apr 15, 2023 10:01:00 GMT
the techno wizardry bit, the CDI itself wiring is pretty straightforward, I'v got the black and white wire (joint under the tank) on the side stand monitoring terminal, the CDI is then programmed so that ground kills the engine, this allows the the stop switch on the throttle side to work. theres a jumper wire that needs to be cut in order to allow that terminal to work. next is the fandangled programming and understanding it, this is just a screenshot from the suppliers website and not what I'm running the various bits, easy one first, side stand blocking (kill switch for us) set to ground blocks ignition. the output voltage to the coil needs to be at the minimum to allow a fat blue spark at the plugs, I'v currently got this on 210v but may do more messing with that. too high and you could burn the coil out and get excessive wear at the plugs. the bit I'v really struggled with was understanding the sensor advance angle but I think I'v got it sussed now. it goes something like this, align the TDC mark on the crankcase with the centre of the two advance marks and then set the HALL sensor to trigger at this exact point. leaving everything in this position, measure the angle between the TDC mark on the crankcase and the TDC mark on the mag wheel, I make it 43.8 degrees. with all these things set you should now be free to programme the ignition curve, still tinkering a bit with that but the bikes running really nicely, I'd be happy enough to do derbyshire on it with the set up I'v got
|
|
|
Post by buster on May 27, 2023 16:13:47 GMT
oops cocked it up again so earlier in this conversion I'd tested the polarity of the magnet on a couple of 400/450 mag wheels, all came out as north pole. I'd therefore (stupidly) assumed they were all that way. they arent, some are south pole, including the one I have fitted. there is an upside to this though, I think the magnets signals were blending into one pulse at higher rpm, which suggests a longer pulse signal is necessary, a row of magnets should sort it. you can get an idea of the positions in this pic next problem was I couldnt get full advance, so I wrote to Felix, if anyone wants to explain this to me I'm all ears if I'm understanding this right (richtig) then the rising edge and falling edge (confusingly, when translated to english) refers to the fixed sensor and not the moving magnets. also if I shift the position of the first magnet to trigger slightly before the full advance marks I may be able to use lower advance figures when programming the cdi to obtain the correct advance. all thoughts welcome as my heads spinning
|
|
|
Post by buster on May 30, 2023 11:50:13 GMT
its official (not members on here, just to be clear) my magnet supplier seems to have twisted the laws of physics through 180 degrees to suit themselves an compass and the magnets they sent me (claiming they are south pole) pulling the pin apparently their fandangled machine is right and the planet is wrong...
|
|
|
Post by buster on Jun 1, 2023 10:02:48 GMT
hooray, finally got some magnets of the correct polarity, more by luck than management though, they werent sold by polarity but it just so happens that they are right. sadly no time to play today as I'v got a family wedding to go to, unless I can discretely bugger off after the grub...
|
|
|
Post by BritPete on Jun 1, 2023 14:14:45 GMT
My kind of wedding - leaving early
|
|
|
Post by buster on Jun 2, 2023 15:58:43 GMT
turns out neodymium is quite combustible
|
|
|
Post by buster on Jun 3, 2023 12:59:29 GMT
probably better to buy magnets of the correct thickness and not file them down, though it can be done despite what the interweb says, however attacking them with a flap wheel in the angle grinder is not such a good plan. the heat makes them even more brittle and also causes them to lose their magnetism to a greater extent, oh well live and learn. so one row of fairly untidy looking magnets, where you can screw them on is a bit limited because of the screws holding the inner magnets that generate power for the electrical system. but it works fine and is one continuous pulse signal, hopefully the widening of the angle here will reap benefits. also as Felix recommended the trigger point is now a few degrees before full advance. if I'v understood that right then I can knock the advance figures down in programming and yet get the advance that I need. more on that as I do it also got a bit paranoid that I wasnt going to have a bike for derbyshire, so I ordered another ignitech unit. I made some discoveries about that system in the process, they each have to be ordered and made for very specific applications. they are available to order for the 400's and are more or less just a plug in and go (need to sort a switched live from the ignition with a fuse) they come ready wired to plug into the coil and the pick up. the problem with that for me is that the pick up can fail and I dont think anyone can currently rewind one as the very fine wire doesnt seem to be available. the upshot of this is that I'v ordered one specifically for the sensor I'm using, Jiri at ignitech has specified that there needs to be a signal at 40 degrees BTDC and another signal at 5 degrees BTDC. so while I was messing about I made up another mag wheel to this spec.
|
|
|
Post by buster on Jun 4, 2023 14:05:46 GMT
lots of faffing around yesterday, it likes the longer signal (more magnets) but not too long, so magnet circled is now removed took it up the road with the cover off and it felt fantastic, like a well sorted 400 should, still had a bit of a stutter around 3000rpm at this point. so I spent the afternoon messing around with the program, first got the F mark to align at low revs, the only way I could do that with any consistency was to adjust the sensor advance back to 42 degrees. then upped the peak advance until the advance marks swung back nicely, then had a mess about to try and iron out the 3000rpm glitch. all this went really well and was feeling great at a standstill so cover and gear lever back on I grabbed my lid and went to start it. a few degrees of engine movement and bang, it stopped. I forgot how little clearance there is to the casing, no major damage done but the magnets didnt survive. so found some very thin magnets and thought I'd give them a go with JB Weld (supposed to be good to 288 degrees C, how hot does a mag wheel get?) not sure if this will work or be reliable but we'll see. also slightly concerned that the program is running into the red, I'll have to email Felix and see if this could damage the cdi
|
|
|
Post by buster on Jun 4, 2023 19:00:13 GMT
some success at last, the JB Weld has gone off enough for a test start up, adjusted the sensor a bit nearer to the new magnets and the bike fired up straight away on both pots. put the cover back on and clearance is now fine. I'll leave it until after work tomorrow to give it a test ride
|
|
|
Post by buster on Jun 5, 2023 17:26:26 GMT
quick test ride to get it up to temperature, Pendle Hill in the background everything feels good, pulls and rides like a 400 should and I seem to have got rid of the glitch at 3000rpm. whipped the casing off while it was still hot to inspect the JB weld and everything seems fine. I cant say I'm 100% happy with that as a fixing method for the magnets, it smacks a bit of gopping bodge but if it works it works, time will tell for sure
|
|
|
Post by buster on Jun 9, 2023 19:42:49 GMT
good job I didnt shout eureka again, nearly there though. so I went out this afternoon to do some yorkshire dales run research and all was going well, fantastic roads, bike seemed to be running really well until the fairly long climb out of Hawes, at first I thought it was running out of petrol, but no plenty in the tank what I think is happening is prolonged heat (fast run up to aysgarth) is reducing the magnets strength that act as a trigger for the Hall sensor. so basically had to wait for the bike to cool down and then managed to nurse it the 35 or so miles home. oh well better to find out now than in the middle of Derbyshire looks like I'll have to put some thicker magnets on, I'll find out more in the morning when the bikes cold again
|
|
|
Post by BritPete on Jun 9, 2023 20:17:34 GMT
I’ll bring a tow rope for the Derbyshire run
|
|
|
Post by buster on Jun 10, 2023 12:09:20 GMT
a quick check this morning and two of the magnets had more or less melted also coating them over the top with JB weld was probably a bad idea, it left them effectively in their own little oven, oops when I rolled the wheel past the sensor I was effectively getting two pulse signals (confusing the hell out of the CDI), at 4500rpm it came to and would run quite nicely, tricky ride home keeping the revs at that point but I made it. forgot to take a pic of the mag wheel I'v just done but its now basically got two magnets like these next to each other. and as a slight aside, if anyones ever wondered whats inside that little brass blimp, its nothing more exciting than a small neodymium magnet with a tiny steel shield on its underside (my numb finger for scale), I'd imagine honda were trying to shield the magnet (neodymium must have been quite exotic stuff back then?) from heat as best they could
|
|
|
Post by buster on Jun 10, 2023 18:30:26 GMT
pleased to report that the bikes ran faultlessly all afternoon, I'll consider that ready for derbyshire now, after a wash and a bit of polish
|
|
|
Post by buster on Jun 17, 2023 18:11:37 GMT
things I heard in derbyshire today from 400 superdream owners 'bloody'ell that thing goes dunnit' 'christ that thing pulls well' requesting the forums permission to be ever so slightly smug
|
|