TW200 Forum banner
  • Hey Everyone! Enter your bike HERE to be a part of this months Bike of the Month Challenge!
1 - 13 of 13 Posts

·
Registered
Joined
·
6 Posts
Discussion Starter · #1 ·
Hi, I designed and made a CDI with no advance for my dad's '87 TW a long time ago. It worked to ride around my yard and around the block but really bogged at higher rpm. I'm updating that design with ignition advance. My problem is that my dad lives a few thousand miles away now. Does anyone in the Bothell, WA area have a spare 87-99 TW motor or basketcase whole motorcycle (i think the motor/electrical remained the same through that period of time?) that I could use to test my CDI? thanks
 

·
Registered
Joined
·
1,034 Posts
Go visit the Ole Guy! Take him fishing.

I wish my Dad was still someplace I could drive to.

PS... Sorry, I don't have a TW of that vintage. I'd drop it off at your place if I did.

Sent from my Pixel 3a XL using Tapatalk
 

·
Registered
Joined
·
1,034 Posts
I wasn't trying to suggest that you are not a great Son. Where does he live?

Sent from my Pixel 3a XL using Tapatalk
 

·
Registered
Joined
·
144 Posts
hi dabiz

i'm interested in your cdi project. especially that the timing advance aspect. i've been running fixed-timing cdi's on my tw for a few years now. 10btdc, so top end is about 50mph.

these are the cdi's i'm using

this is some of the testing i did

a separate timing advance (or even base timing shift) module would be super usefull. there's tons of $20 cdi's out there, the problem is just to get them to fire when ya want.

it would be way less engineering, as one is only dealing with the pulsar signal, and applicable to a wide range of motors.
 

·
Registered
Joined
·
6 Posts
Discussion Starter · #5 ·
My dad and a few TWs are in WI which is not a weekend trip for us. We plan to visit him this winter when he's not so busy though.

I wasn't trying to suggest that you are not a great Son. Where does he live?

Sent from my Pixel 3a XL using Tapatalk
 

·
Registered
Joined
·
6 Posts
Discussion Starter · #6 · (Edited)
I did the first half of this project (10+ yrs ago) just to get a basketcase TW idling. My goal was kickstartable, no battery, and stock timing. From memory, I used the pulser signal (your waveform thread post #8) backwards. I used negative pulse as the advance and the positive pulse as the idle ignition pulse which I assumed was around 9degrees BTDC since that's what the service manual showed. From old memory of the figure in the manual, the TW idles around 9 BTDC and advanced up to 30degrees BTDC. Why do the voltages in your figure only go up to 1V?

Back then, I breadboarded this circuit in Figure6 of STmicroelectronics AN819 and the motorcycle ran. I was able to ride the motorcycle around the yard and do a bunch of laps around the block until I realized I was leaving a trail of oil down the road. Of course I didn't take any notes on the component values I used to make this work because why would I do that :rolleyes:? I had also breadboarded a microcontroller to measure RPM which would work for 5-10 minutes but then poop itself (I'm assuming due to noise from firing 150+ volts through a breadboard). I moved to WA after this, dropped the motorcycle off with my dad (he was the actual TW enthusiast) and abandoned the project.

Anyway...today I know slightly more about electronics (tho I'm not an electrical engineer) than I did and still want it to be kickstartable, no battery required and with digital advance (digital is way easier to figure out for a non-EE haha). I have a circuit prototyped that works on the bench and correctly measures frequency/RPM from a waveform generator and calculates an advance but won't know until I plug it into an engine...there still might be too much noise. For now, it's designed to bypass the kickstand switch/clutch switch circuit and uses the run switch 'backwards'.

Per your suggestion, I might be able to make an 'advance unit' for an external CDI...it might not be that hard but i'd have to think about that and I have no motor for testing. What do your fixed advance CDIs use for a killswitch?

hi dabiz

i'm interested in your cdi project. especially that the timing advance aspect. i've been running fixed-timing cdi's on my tw for a few years now. 10btdc, so top end is about 50mph.

these are the cdi's i'm using

this is some of the testing i did

a separate timing advance (or even base timing shift) module would be super usefull. there's tons of $20 cdi's out there, the problem is just to get them to fire when ya want.

it would be way less engineering, as one is only dealing with the pulsar signal, and applicable to a wide range of motors.
 

·
Registered
Joined
·
6 Posts
Discussion Starter · #7 ·
What do you do with the brown and/or yellow wires from the stator since they aren't used by the fixed advance CDIs?

hi dabiz

i'm interested in your cdi project. especially that the timing advance aspect. i've been running fixed-timing cdi's on my tw for a few years now. 10btdc, so top end is about 50mph.

these are the cdi's i'm using

this is some of the testing i did

a separate timing advance (or even base timing shift) module would be super usefull. there's tons of $20 cdi's out there, the problem is just to get them to fire when ya want.

it would be way less engineering, as one is only dealing with the pulsar signal, and applicable to a wide range of motors.
 

·
Registered
Joined
·
144 Posts
What do you do with the brown and/or yellow wires from the stator since they aren't used by the fixed advance CDIs?

the red wire feeds the cdi, and the brown wire is capped. my theory about the brown wire is that it is there to provide quick power during the first couple cranking revolutions - note the small crank/idle increase of voltage in relation to the red wire. i remember reading that how a stator is wound would determine that. it's continuous with the red winding, so capping it isn't leaving voltage with nowhere to go.


yamaha knocks back the timing advance as a rpm limiter, tho it wouldn't have to be that way : )

the pulsar only puts out a tickle, but the "1 volt" in the waveform pictures is not really 1 volt, it's just "1" which indicates the clip level in audacity, the audio program i used. the volts are written in, measured with volmeter.

green+ white- pulsar polarity yields a falling-plus-at-zero-volts trigger, which is kindoff a standard as far as i can tell - ie the gy6 stuff, the tw, and if you think about it, points (tho points are mechanically advanced).

timing advance would be built off that reference signal.

the gy6 cdi's must take that ref (2x), and next revolution delay 350deg/advance10.

for cdi on/off, i run my kill switch backwards, with an internal mod that i forget exactly, something gets un-bonded in there.
 

·
Registered
Joined
·
6 Posts
Discussion Starter · #9 ·
I thought the brown/yellow wire was a signal to the cdi to switch the advance circuit. Looking at my notes, i measured it between 30 and 50VAC while idling and the charging coil maxed out my voltmeter over 150VAC. And on second review of the manual, I must have the 'EU' version because the brown/yellow wires are a separate coil from the red coil.
 

·
Registered
Joined
·
144 Posts
I thought the brown/yellow wire was a signal to the cdi to switch the advance circuit. Looking at my notes, i measured it between 30 and 50VAC while idling and the charging coil maxed out my voltmeter over 150VAC. And on second review of the manual, I must have the 'EU' version because the brown/yellow wires are a separate coil from the red coil.
i've heard lots of theories about"secondary" rpm timing triggers, but it never made sense to me.
there's a perfectly good pulsar signal, why complicate things?

frequency increases with rpm, reference that against a clock on a chip, or in analogue an r/c? circuit that delays more at lower freq's. done.

the ford eec is the automotive system i'm familiar with, and it definitively only senses rpm via the hall pickup in the dizzy. a vane for each cylinder, with one vane shorter as a ref.

gm and others use a crank position sensor, which i guess provides a bit more accuracy, because it's on the crank not the cam (turns 1/2 speed, so 1deg = 2deg in actuality).

not sure re the stators on your model. every diagram i've seen for any year shows them as one winding with a tap, and that's how mine metered. which is not to say yours isn't as described.
 

·
Registered
Joined
·
144 Posts

in that thread bustahump unhooks his brown wire and the bike runs fine, so something is going on with that wire...
 

·
Registered
Joined
·
2 Posts
Hello!
I own Yamaha XT225 1992 which is similar to TW200 and has similar ignition. Now I'm trying to understand how CDI box works. I mean not in principle (it runs your bike), but exactly. In slowmod's album there is a picture of original ACDC CDI board of TX200. So I thought you might help me)
Slowmod, do you still have that board? Could you (or anyone) share hi-res pictures of it? I'll try to reverse-engineer, at least draw a schematic.
I disassembled dead CDI from XT 225 of my friend (4jg version, 1993 and later), which is rather complicated. So I'm searching for something similar but more simple.
What I'm trying to understand right now is why there are two charge coils (XT225 has two separate charge coils, while TW200 has the combined into one with 3 wires). You said that you expect that it might be for some extra power on startup. That sounds logical, I'd like to check that idea.
Why I need that is to understand how a rotor can be combined with stator. XT225 has at least 3 old versions: before 91 (4 mfgnets, 2 coils), 91-92 (many magnets, tends to collapse, stator with star configuration, 2 charge coils), 93-00 (many magnets, covered, do not collapse anymore, stator with even more beams in star configuration, 2 charge coils). And each has CDI block. The wirk somehow if you combine them but I'd love to know for sure what happends.
Thank you!
 

·
Registered
Joined
·
770 Posts
The best way to make cheap CDIs work is to advance the static timing to 30-40 BTDC. Perhaps by drilling out the rivets that hold the center in the flywheel, repositioning and bolting or re-riveting. I was thinking about doing this when I was experimenting with cheap CDIs on my '87, before I decided instead to just rewire it to use a 2001-up CDI.
 
1 - 13 of 13 Posts
Top