Honda CX500 Motorcycle CDI ignition unit

[stj]

test the thermister by reading the resistance while heating or cooling it.
the "coller" is just insulation - probably to raise it up a bit incase it gets hot.

[redwire]

I got an email response from the Finnbox G47 maker:
"Sorry to tell but I haven't made any cdi units for a long time. The schematic is the thing which I keep with me. It's not very complicated but unfortunately it's the most important part of the process. I think I have told enough in my blog how to make one."

So either we make a PC board or leave it.
OP, take your original PCB and check behind a bright light for knife cuts to the traces, from cutting off the potting compound.

The UF5408 is huge, it won't fit and slower 1N5408 was on the first prototype, then all changed to 1N4007. The HV coil is 430R nominal so it can't put out even 1A. The KILL switch just shorts out the coil.
My change would be some to faster recovery diodes or a little more 2A for the high voltage section.
Japan in the late 1970's did not have great HV semiconductors. Here there are three 600-800V diodes in series, giving 1,800-2,400 PIV which is quite high and slow at 3usec.

The heatshrink just has some resistor in it, to prevent shorts to the resistor next to it. The green thermistor is nearby on the board.

[roadrash]

Think I've found the finbox designers blog
https://tuumaustauko.blogspot.com/p/...itech-cdi.html
Need to use Google to translate Finnish t English

[redwire]

I did translate it and read it but it's some Rev. 1.2

I can't make out some resistors values. Other schematics have errors. Sigh.
R21 - goes to D12/C8 left TDC rtd, then to SCR-L gate, maybe 910R?
R19 - goes to D11 right TDC, then to R26, maybe 180R?
These are by where the wires come in and not in pics.
TH1, TH2 - thermistor, maybe 500R?


I don't see the point in reverse-engineering the box, posting pics on your blogs, youtube and then not giving out any schematic or parts list, if you're not going to make and sell the replacement modules? Am I wrong? Then you do improvements and post nothing specific.

[stj]

if it can be designed small enough using smd parts then you can get the chinese to make the boards for peanuts and even put the parts on it for you.
i think the limit for cheap boards is something like 4inch square.

[roadrash]

You guys a really great helping me with this dilemma.
I could just buy a Igntech one but Its too expensive for what it is.
Every bike made after 1980 had CDI ignition and now they are all dying. Won't be many running bikes from this era if there are not alternatives available but all manufacturers are not making schematics of these CDI's so they can be remade. God help modern bikes and cars when their ECU's fail in the future. They will never be classics that run.
Sorry about rant it's so frustrating this CDI issue. Hitachi should make the info available if they haven't destroyed it. I did some more searching and found a forum where a guy has made a working CDI and making all info available incuding PCB details.
Take a read here it's very interesting.
https://www.cx500forum.com/threads/s...options.24593/

[stj]

cars wont be a problem, their are several aftermarket ecu's available from megasquirt kits all the way upto the holly dominator beasts!!

[redwire]

I find little mistakes in most of the redo's of the CDI unit.
I have it figured out but I can't read two resistor values that Finnbox guy used, what a hassle.

I looked at doing an all SMT build but there are many high voltage parts which are big size, and it's high vibration and temperature swings. Through-hole is tougher and with no more potting compound, I would go that way with D2PAK SCR's because they look better for heat and vibration.
I never tried a chinese board stuffing house. How do they get the parts, and their SCR's are probably from the bowels of guangdong? LCSC is expensive.

[stj]

funny you should mention it, they use LCSC.
julien illet has done some vids about the process.
https://www.youtube.com/watch?v=xDl1AwAzREc

[roadrash]

I could do with 2 of those SCR';s I have one working one but the legs are broke. I put that Thermister I think you said it was in the freezer for a hour or so but its resistance only changes by 30 ohms. It is this little green thing with 351 marked on it you are talking about??. That PCB was rubbish I had quite a few problems with the plating in some of the holes falling out. I couldn't get it all back in but made sure I soldered the parts legs to both sides of the pcb holes.
Do you think we can get this working as it should? I could pour wax into the plastic tubes when I put them back together and use a hot glue gun to hold others parts stable and then seal it all water tight in some kind enclosure.
Its these unknown SCR's that are the problem right? Be nice if we can use easy to get off the shelf parts. I wish I knew more about what you are talking about but Motorcycles and engines are my specialist area.

[redwire]

For reliability, I would have a hard time using the original PCB.
I can lay out a new PCB, it's not that difficult for me. I've figured out all the parts. It doesn't all need to be potted. Like the Finnbox, a bare PCB and waterproof plastic box is good. Just the wires coming in and out need to be sealed up. Keeping it the same as the original design but better new parts. Let me know what you think.

About replacement parts:
The thermistors are the green disks marked "351" and not able to get. They are Hitachi 350 ohms at room temp. It is possible to substitute 470 ohm part such as Vishay NTCLE100E3471JB0, if the 82R is changed. Popular size MF52 or MF11 thermistors hard to find below 1k ohms.
The "351" should go from 350 ohms up to around 770 ohms when cold in a fridge, but touching it will warm it up fast. Freezer should be above 1,000 ohms. It's not important, I figured out the temperature compensation.

The SCR's: BT151-500R (NXP) 500V 12A which is popular for CDI but obsolete now, I would use ST TYN612RG 600V 12A.
The smaller TO-202 SCR is apparently the rev limiter. I think an ST X0402MF would work, 4A low voltage is fine. Finnbox went so small to a 1A TO-92 part I would not do it.

[stj]

it may be cheaper to use matching scr's given they are probably sold in 5's or 10's

[redwire]

The weird part is the Left and Right cylinder have different SCR-firing circuits, as far as the gate drive (resistors). The pickup coils are supposedly the same between Right and Left, so this leaves me puzzled why no symmetry. I can't get values for R19 or R21 from anyone's pictures.

The semiconductor shortage is affecting some SCR's. I see some ST parts out of stock until October at Mouser and Digi-Key. On a new PC board I'd go with D2PAK parts because they can run cooler with a little copper pour. Anything that sticks up high on a PC board can't take vibration and eventually breaks off, so SMD is better I think than TO-220 up in the air.

I'm drawing a PCB and have to find an enclosure.

[roadrash]

Sorry for nipping off for a bit. I got pursued by a friend to rebuild a 1960's Honda CB72 for him and I've just finished rebuilding the engine and now got it back in and trying to sort out the crappy wiring and switches.
Hey Redwire it's incredible how you understand all this CDI stuff. I wish I could be of more help.
If you could make working CDI unit that would be incredible.

[redwire]

What size is the original can? I couldn't find a waterproof box a decent size, popular is 4.53" x 2.56" x 1.58" (115x65x40mm) (i.e Bud PN-1321 too small) but then it jumps to 4.53" x 3.54" x 2.17" (115x90x55mm) which seems thick. Finnbox guy I guess people used Velcro to tie it down.

Did you get values for resistors R19, R21?

[roadrash]

Originally Posted by redwire

What size is the original can? I couldn't find a waterproof box a decent size, popular is 4.53" x 2.56" x 1.58" (115x65x40mm) (i.e Bud PN-1321 too small) but then it jumps to 4.53" x 3.54" x 2.17" (115x90x55mm) which seems thick. Finnbox guy I guess people used Velcro to tie it down.

Did you get values for resistors R19, R21?

Original can is 92x65 by 45mm deep.

I am having trouble locating R19 but if the one I think is it then its as follows.
R19 =180 ohms???
R20=610 ohms.
R21=910 ohms.

[roadrash]

in this schematic of my CDI box I got here:
http://rajamolor.blogspot.com/2008/0...schematic.html
its shows that those tubes with a resistor sealed inside like R20 for example, are marked as "CAL"
would these be for calibration done in production.

[redwire]

The heatshrink tubing is just insulation to prevent the part from shorting to parts next to it.
The CAL resistors are the ones sticking way out on pins past the edges of the board.

From your pics R20=220R (in the heatshrink) which matches other people's pics. Not 610R which does not exist as a EIA E-24 value. Where is that from? Who is 610.

The rajamolor schematic is so hard to read, R21 is referenced twice (one by CAL R24 and then one by R20 as well) looks like it is 110R or 910R. Can you check if there are two resistors getting mixed up.

[roadrash]

Originally Posted by redwire

The heatshrink tubing is just insulation to prevent the part from shorting to parts next to it.
The CAL resistors are the ones sticking way out on pins past the edges of the board.

From your pics R20=220R (in the heatshrink) which matches other people's pics. Not 610R which does not exist as a EIA E-24 value. Where is that from? Who is 610.

The rajamolor schematic is so hard to read, R21 is referenced twice (one by CAL R24 and then one by R20 as well) looks like it is 110R or 910R. Can you check if there are two resistors getting mixed up.

OMG Sorry Redwire that must have been a typo as your right R20 is red, red, brown, gold=220R.
Your right about the schematic being awkward to read. R21 is shown twice on PCB near R24 and its also next to C2. Oh and R19 is definitely 180R

[roadrash]

Originally Posted by roadrash

OMG Sorry Redwire that must have been a typo as your right R20 is red, red, brown, gold=220R.
Your right about the schematic being awkward to read. R21 is shown twice on PCB near R24 and its also next to C2. Oh and R19 is definitely 180R

Hey Redwire, thanks for all this help. i appreciate what your doing but please not at your expense. I am quite happy to cover any costs you incur. My CX500 could pack up at any time with a 1980 CDI unit and as I said it maybe whats causing the popping at exhaust when slowing off the throttle. Its definitely not the carburettors doing it and not the exhausts either. Can you make your own PCB's?
I have put the original PCB back together now so if making a new CDI is going to be a problem and the components like the SCR's I have on the original board are still available somewhere, I should be able to get this old one working again now.