RG500 CDI Technical details and modifications

NOTE: PRELIMINARY WEB PAGE SECTION UNDER CONSTRUCTION!
(I've sobered the page a bit to match style of other user contributions, Rob)

This is part of a few web pages I've been making to cover all that I've done with my RG500. Very soon I will finish all the other sections and link it all together. I can be emailed at marc@aems.com.au. I'm a bit snowed under with work right now, which is why I figured it's better to put all the stuff in one page!

I can be reached also at: Advanced Engine Management, 22 Melrich Rd, Bayswater, Victoria, Australia 3153. Ph: +61 3 9761 3161 Fax: +61 3 9761 3162.

Well, one day I was feeling especially constructive in my search for more power. I spent the next few weeks discovering how to pull the cdi box apart, and how to dissolve and remove the potting compound and 'rocks' that make it a black lump.

Then I spent bloody ages! working out a component overlay using digital photo's taken, and drew a schematic of the circuit, laying it out progressively as I learnt what each part of the circuit did. A signal generator, digital CRO and Microsoft Excel were very handy in simulating and charting the cdi's internal functions and timing output.

After I worked out what it all did, and which parts to tweak to change different advance curve areas, I borrowed a 'modified' cdi box from a friend, put it on my tester and charted its timing range which is changed by a trimpot. Now that I knew what a 'modified' cdi box did to the timing curve, it was easy to discover how to make the same modification to mine.

1. DISCLAIMER: I put this information here for perusal only! Don't dare do this unless you:
2. 1. Know what you're doing
3. 2. Are very happy to risk destroying your precious and expensive cdi :-)
4. 3. Not to mention happy to risk destroying your engine from ignition timing changes.
5. 4. Understand that nodoby but yourself is liable if it doesn't work, or your engine blows up and/or siezes.
6. 5. Have access to decent, reliable high octane fuel, or turn the advance down with crap fuel.
7. 6. Have access and can use a top quality soldering iron, signal generator and oscilliscope.
8. 7. etc....etc....!

Disassembling the cdi box completely (you don't need to do all this for normal timing change modifications, see below):

• With a hacksaw I cut around the 'top' of the box.
• Pried, wrenched, and generally, but carefully broke and removed the top plastic area off. Then the sides and bottom.
• Now I'm left with the cdi having no plastic surround, just a potted lump of semi-flexible black rubbery stuff. There are small quartz-looking rocks all through the potting compound, which didn't look like fun, but actually made it easier to dissolve apart!
• After a bit of experimentation and testing, I put the whole cdi, but not the wires, in a bath of 'General Purpose' paint thinners, bought from an automotive shop in the paint area. I used a metal tin covered in aluminium foil to prevent evaporation.
• Then I forgot (yes, actually forgot!) about it for 3-4 days. Luckily that seems to be just about the right time. The potting compound had swelled greatly, pressing against all the little rocks, and soaked up a good portion of the thinners. I lifted the cdi up by the cable, and nearly all the potting compound and rocks were left in a nice pile in the tin. More work with thinners and a blunt tool to scrape the remains off and it was done.
• The wires swelled a lot from soaking up the thinners, but soon dried and returned to normal, but the sheath was harder. Also the thinners didn't remove any of the colour codes or part numbers which was a real bonus!
• This is the point where I took pictures, made an overlay and starting working it out.

A schematic of the RG500 CDI in gif format (GIF 48Kb)

The overlay created to do the schematic (JPEG 240Kb)

My 'development' CDI with enabling switch and plug in resistor

After a lot of messing about on the bench I worked most of it out and started tweaking parts and mapping the new ignition curves. It's mainly the resistors on the right side of the picture, on stakes that make the changes. These were pretty easy to locate since I know they are on stakes for calibration in production.

Working off a simple drawing representing the 'base' ignition map I drew pictures of which part change did what. Note that the representative 'base' map is for reference only and is not a dead accurate picture of the actual stock RG500 curve. My curve change indications are general only and also not a dead accurate indication of curve shape change.

As they are resistors, placing another resistor in parallel (directly across) with it will reduce the resistance value. Or removing one leg of the resistor and placing another in series with it, or replacing it will increase the resistance value.

For reference: Resistance is measured in Ohms. R is Ohms, K is KiloOhms (x1,000), M is MegOhms (x1,000,000).

For resistors in parallel, the total resistance value is 1/(1/R1 + 1/R2). Series is simply R1+R2.

Timing changes by modifying component values

Increasing or decreasing R17	Increasing or decreasing R19	Increasing or decreasing R24
Increasing or decreasing C7, R29 or R33 have the same effect	Moving the crankshaft timing sensor	Refer to this for component locations

After bench testing my development cdi back to back with one that had already been modified, it was clear that only R19 was reduced by a trimpot on these aftermarket modified cdi's. The aftermarket modified cdi had one small square hole in the top right corner cut for the trimpot, and another larger square hole cut in the centre for another 'mystery' modification

But: the mystery modification in the center of the pcb area that had me going for quite a while! So I could make the timing change easily but was still trying to work out what else might prevent the timing curve desired. I knew that there was supposed to be a high speed retard setup on the cdi also, where it retarded the timing greatly at or around 10,000rpm. I discovered that as the crankshaft timing sensor's signal increases in frequency, it's amplitude is dropped in one section of the circuit causing an abrupt retard of the timing. So I made a small circuit to bypass this effect. Though, in dyno testing there was not a noticeable difference in power after this change.

Here's how the cdi was connected for the tests and charting done. Mail me to receive my calibration and timing data and charts that used this setup. The charts and timing graphs were done in Microsoft Excel V5.0

The end result!...

CDI Modifications

Component overlay with schematic of changes	Closeup of centre modification
The end result. This is a 'modified' cdi before filling with epoxy.	Closeup of trimpot

Rob Koopman ( Rob.Koopman@inter.NL.net ) ^{RG500 Index}