Roll Your Own CDI

[ivan H]

Clicl on the links under "build ur own CDI" at the end of any of Rohmell's posts for a couple of them. Cheers

 

[nightcruiser]

No mate, I dont, sorry, & I'm using a phone 2 go online which makes it a little harder. I did have them saved but lost the phone. I know they're a ways back. If I find them I'll let u know. B good 2 c how u go with the light. Cheers

I was working with a plan for 6V LED lighting off the white wire, but taking power from the CDI actually opens up a lot more doors as far as lighting goes. When I had my magneto connected to the oscilloscope the power coming off the top coil was huge compared to the bottom coil. That said, I worry about the smaller wire in the top coil, being about 1/3 the size of the bottom coil (white wire/headlight coil). I've been thinking about trying to make a system that would use the negative pulse from the top coil to create 12VDC for lighting. My fear is I will burn that coil, but you never know 'till you try... I'll definitely keep a spare magneto on hand, and will report back if I come up with something that works out....

 

[ivan H]

I think both Norm & Impression rewound stator coils & posted the wire guages & amount of windings. The white wire puts out about .3 of an amp I think, the cdi winding a lot less. From memory maybe the Rohmell posted cdi that charged a battery used the negative half of the sine wave, but I'm not sure on that. Good luck with it anyway. Cheers

 

[GearNut]

The following is what I have learned by reading many, many, many (did I say many?) threads regarding the power output of the white wire.
The white wire output averages 6 volts, 3 amps. Some are better than that, most are less.
Some folks have measured 16 volts + unloaded.

 

[Goat Herder]

Yeah the true voltage saturation is 6 volts for the most part. I concur with Gearnut.

 

[ivan H]

Yes, it is 6V. When read without a load the meter IS the only load it see's & most meters have an input impedence in the order of 10 or 20 million (mega) ohms per volt, hence the higher reading. Cheers

 

[Cavi Mike]

What I came up with. Bought the little box from Radio Shack and couldn't resist buying the little heatsink. Obviously it won't be super efficient since the majority of the heat needs to travel through the screw but it probably wouldn't get hot enough anyways. I just think it looks cool. Haven't tried it out but I sure hope it works, I only have the one SCR.

 

[Greg58]

Looks good, I like the heat shrink over the solder joints. I couldn't find 18 ohm resistors locally so I ordered some from ebay. I'm going to use the pill bottle for mlne, my wife can bring home any size I need from work.

 

[ivan H]

When mounting the heatsink 2 the SCR its adviseable to use a mica or silicone insulator to prevent the heatsink from being "live", as its connected to the anode of the scr. Cheers

 

[Greg58]

I finally got my resistors, so now I have everything to build two. I have two other recs that are rated at 35a but only 200v ac. The recs I plan to use are 500v.

 

[ivan H]

Yeah, good idea to. 200 V items would b running at close to 50% over their rated working voltage. The ones I use are 400V. I've been using 285V polypropylene charge caps because the local outlet has them on the shelf & they've held up well & done heaps of miles without a problem, but I should order some 400 V caps. I think Rohmell posted an add on circuit to experiment with for a retard feature at above mid rpm for the cdi ur making, I'll try finding & get back to u if ur interested. Cheers

 

[ivan H]

Hi Greg58, yeah, scroll thru page 14 (or others b4) of this thread & there's a pic showing layout of the simple cdi without retard feature, then scroll thru pages 15 & 17 & there's pics showing the same cdi with the retard feature added, showing values Rohmell recommends as a starting point for experimenting with. Cheers

 

[Greg58]

Well it works! I built the standard version. In the pic. I noticed I had the caps in series not parelell so I made the change before trying it. I tested it on the bike that has a bad magneto, the stock cdi is good but the voltage from the mag. is so weak it will not fire a plug. But with the " fire in a bottle " it started up, that tells us that this cdi is far better than the stock set up. I have a mag. coil on order then I'll see how good it works.

http://motorbicycling.com/album.php?albumid=259&pictureid=4455

 

[Greg58]

These are the components I used. 1. nec 2p4m scr 2. in5406 diodes 3. 18ohm 1w and 82ohm 2w resistors. 4. mm36r 1.0k 480v ac/ 1000v dc caps. 5. Emgo coil 24-71536. I'm going to run this cdi for a while before I give it two thumbs up. I upgraded the resistors after R/B had his 82ohm overheat.

 

[Greg58]

I did my first test run on the new cdi, it seams a little retarded on the bottom end but really comes on from 20 mph. Real good at wot. the diodes I used may be the reason but it flies on the top end.

 

[Greg58]

I think I have it right on the second cdi, I changed to a s2035k scr and added a .047uf cap. Great responce from idle to wot.
http://motorbicycling.com/album.php?albumid=259&pictureid=4522

 

[ivan H]

Cool. I take it that u now reach higher rpm than u did with the stock cdi as u say it flies wide open. The stock unit kinda acts as a rev limiter. Did it smooth it out also. Cheers

 

[Greg58]

so far it has out performed the stock cdi, I plan to ride tomorrow since i'm off. I plan to next start work on one of the retarding circuits.

 

[ivan H]

Cool. If u get it 2 start retarding from 2/3 throttle on it'll rev higher & smoother at hi revs, & it'll run cooller at hi rpm as more of the heat of combustion is transfered 2 the pipe, which is also benefitial if ur running an expansion chamber as sound waves travel quicker at higher temps, so it extends the "on the pipe" range up top. Cheers

 

[dracothered]

Ok can anyone tell me the difference between these two circuits and why the jumper are in different places? What does each one do?