XS1100 TCI inner workings

[Ranger_xs1100]

Deebs, Nice write up!

I’m sure you already know this, but others may not.

You can easily check those TO3 power transistors with an ohm meter.
Checking for forward and reverse bias out of the circuit.

On NPN transistors, clip the positive lead of the meter to the base of the transistor and the negative lead to the emitter and then to the collector.
Both forward bias readings should be low ohms. Reverse the leads and both readings (reverse bias) should be high ohms.

A simple but accurate test, however does not test for intermittent failures under load.

Keep up the good work Deebs!

[bikerphil]

Question, does this T-03 transistor for the 2H7 box have a more complex integrated circuit, or is it just a power transistor? I've replaced T-03's in musical amplifiers before and there was nothing special about those. Motorola might have made something compatible.

On a side note, out of 30 years of fiddling with these bikes for people, I've only seen one 2H7 box go bad, it doesn't seem to be a common problem and there should still be a decent amount available if needed.

[Radioguylogs]

I'm watching and learning. Thanks Deebs!

It's amazing there aren't more electrolytic capacitors failing in these boxes.

[DEEBS11]

Deebs, Nice write up!

I’m sure you already know this, but others may not.

You can easily check those TO3 power transistors with an ohm meter.
Checking for forward and reverse bias out of the circuit.

On NPN transistors, clip the positive lead of the meter to the base of the transistor and the negative lead to the emitter and then to the collector.
Both forward bias readings should be low ohms. Reverse the leads and both readings (reverse bias) should be high ohms.

A simple but accurate test, however does not test for intermittent failures under load.

Keep up the good work Deebs!

Yes Bob, that is a quick way to check a transistor. Electronically it's just two diodes with the anodes tied together like in this picture. The negative side of the diode is the cathode. I'm sure 99% of the Yamaha TCI modules never fail as Phil mentioned. Not bad for late 70's electronics. I'll bet the only ones that do are somehow cross-wired and blown up. The parts are replaceable once you get through the thick layer of clear conformal coating. BTW a diode acts like a water check valve but instead of water, it's electrons. A one-way valve. That's why the meter should read continuity in one direction but not in the opposite direction.

[DEEBS11]

Wonder if it wouldn't be easier to try mapping the input and output and running it with an Arduino?

LAB3, the Arduino is a small micro controller that has a ton of capability and certainly is a cool piece of equipment for processing small signal inputs and outputs. It's not really designed for outside use. You could enclose it but I would worry more about heat and certainly vibration. Also it's way more electronics than you need for an XS1100 ignition. As shown on the TCI schematic - there is no processing whatsoever. The sensors on the left side of the motor do all the thinking including the timing advancement. The TCI circuit is just a simple "step-up" to turn on and off the coils. It's basically a relay that can turn on and off super quickly (THAT would be the design challenge) It's so simple that it virtually lasts forever even in harsh environments. You don't see this kind of simple robust design with most of today's electronics.

[DEEBS11]

Here is what does all the thinking.

[DEEBS11]

If I was going to design a replacement I would start with something like this. I use these in most of my points ignition bikes. As you know, points do wear out over time. The contacts become burnt and they deform which causes ignition problems. The "condensor" or capacitor is around 220 Nano farads or 22 Micro farads (22uf). That is in just about every motorcycle that runs points. But as you can guess the replacement condensors of today are mostly Chinese crap and they don't last like original equipment. Thor Cap in England makes a quality replacement if you want to go in that direction. I do keep a Thor Cap in my tool kit just in case the Gammatronix module fails.

The Gammatronix module gets rid of the condensor altogether. It still uses the point contacts and they stay in place but now they only have to pass micro amps which makes the contacts last virtually forever. The points opening and closing are what activates this device. The challenge with the XS11 is the Hall Effect type sensors. According to the schematic, they run on 6 volts DC. You would need two 6 volt Gammatronix units. They make a dual 12v version for dual points as shown in the pictures.

[DEEBS11]

Here is a Gammatronix unit opened up. It uses a gate emitter which is the "super fast" switch we are looking for. A small transistor turns the gate on and off. Familiar? It's design is similar to the XS11 TCI.


https://www.infineon.com/dgdl/irgb14...535651cf0d240f

[DEEBS11]

This page out of the Gate specs shows the nature of the square wave input. Just like what the XS11 sensors could emulate.

[LAB3]

The Arduino was the first thing that popped into my head and yes, you don't NEED that kind of computing power unless you can find other things for it to do.

A few years back a guy got all excited about putting emergency call boxes all over a particular festival grounds I was frequenting. He built them out of Arduino chips and potted them to keep the elements out. Total waste of time and effort but I know for a fact they lasted at least two years outdoors.

[DEEBS11]

Here is the gate emitter and transistor from the circuit. I keep spares of course. A few bucks each.

[DEEBS11]

The Arduino was the first thing that popped into my head and yes, you don't NEED that kind of computing power unless you can find other things for it to do.

A few years back a guy got all excited about putting emergency call boxes all over a particular festival grounds I was frequenting. He built them out of Arduino chips and potted them to keep the elements out. Total waste of time and effort but I know for a fact they lasted at least two years outdoors.

Add the vibration of a motorcycle with an incredible heat source within a 2 foot proximity. Not to mention the electronic noise generated by the high tension wires to the spark plugs even with the 5K resistor attenuation and you could have problems. When I was younger I thought it would be cool to design and install a built-in garage door opener on my KZ1000 MKII. When the motor was off it worked great. Motor running - nothing. The massive noise generated in the high tension coils of the ignition interfered with that circuit so badly that the garage door receiver did not recognize the code generated by the transmitter. Beware of electronics that looked good on paper but fail in the actual environment. I have 40 years experience (actually next year 2024) repairing what looked good on paper. It was quite frustrating at times. My favorite line from a design engineer whos product failed in the field was "It's not supposed to do that."

[LAB3]

If not an Arduino then how about some other programmable system? Wouldn't it be neat to be able to play with ignition timing across the RPM range?

[DEEBS11]

Just to sum up what I found. There is nothing special inside the TCI. It is what is called in Electronics a "Dumb Box". No programming or logic control of any kind inside. Even the tip over switch is just a ground that shuts off the circuit very simply. It's merely just two heavy duty relays that turn on and off. All the "thinking" takes place on the left side of the motor with the vacuum advance. That is where the timing advance curve is created and possibly changed if you wanted to.

I believe Yamaha changed the curve twice in later years to give the bike a broader torque range in conjunction with different cams. I have an 80 motor in my 78 XS11 sidecar rig and I can't get over how much low-end it has compared to a KZ1000. Great for sidecar work.

[LAB3]

Just to sum up what I found. There is nothing special inside the TCI. It is what is called in Electronics a "Dumb Box". No programming or logic control of any kind inside. Even the tip over switch is just a ground that shuts off the circuit very simply. It's merely just two heavy duty relays that turn on and off. All the "thinking" takes place on the left side of the motor with the vacuum advance. That is where the timing advance curve is created and possibly changed if you wanted to.

I believe Yamaha changed the curve twice in later years to give the bike a broader torque range in conjunction with different cams. I have an 80 motor in my 78 XS11 sidecar rig and I can't get over how much low-end it has compared to a KZ1000. Great for sidecar work.

All I'm doing is kicking around ideas. That's the fun of being a philosopher, you get to tell people how to do things without the nasty process of proving it won't work.