|
SuzukiSavage.com
/cgi-bin/YaBB.pl General Category >> Rubber Side Down! >> That ignitor module /cgi-bin/YaBB.pl?num=1749758893 Message started by FinnHammer on 06/12/25 at 13:08:11 |
|
Title: That ignitor module Post by FinnHammer on 06/12/25 at 13:08:11 Since I fried that ignitor module with my tig welder, I have been increasingly concerned about the scarcity of suitable modules on the used market. I have one coming from EBay, no worry, but still..... One good thing about the 87-88 version of the module is that it is not potted, so that it is possible to get a rough idea about what it contains, and I see a lot of passive components, resistors, capacitors, diodes, a darlington transistor, a comparator and a microprocessor. This about sums it up, and the microprocessor already means that the unit is unrepairable, there is no way to retreive the firmware from it. And even if not, the surface of the parts is covered with ugly goo that makes it a mess to work with. It would probably not be all that hard to build one from scratch, though. See how messy the ignitor looks from inside: |
|
Title: Re: That ignitor module Post by FinnHammer on 06/12/25 at 13:08:42 But how does this system actually manage to calculate the right moment to fire the spark, with the right advance, when there is only one single timing event from the trigger coil. One thing is sure: The ignitor cannot fire the spark before the timing event, so the peak from the pick up coil has to be generated at, or before, the maximum timing advance position. Probably around 40deg BDC. When the comparator in the circuit has determined that the trigger has arrived, it will then be the task of the microprocessor to calculate the right delay from trigger to spark, and this delay must be calculated based on primarily the RPM of the engine. But there are other things to consider, and I will come to that in a moment. But the engine has to turn at least one revolution for the microprocessor to calculate the RPM, this means that at start up, cranking the engine by the starter motor, for that very first spark, the microprocessor is unable to calculate the timing of the spark: it has to be fixed in time, suitable to the low RPM at startup. I am certain that this is the reason for the black wire running from the starter solonoid to the ignitor: It tells the microprocessor that as long as this wire is energised, the timing of the spark is delayed, or retarded by a measure of time, to arrive at around 4deg before TDC, based on the known RPM as cranked by the starter motor. As soon as the starter is disengaged, the engine is running and the microprocessor is already up and running with the correct timing based on engine RPM. Look what the signal from the pickup coil looks like on a scope: |
|
Title: Re: That ignitor module Post by FinnHammer on 06/12/25 at 13:27:55 This is what I had sort of figured out on my own, then I engaged chatGBT for a sanity check, and things got really interesting pretty quick. I asked how a CDI or TAI system calculates the timing, and here is the first part of the discussion that followed: ChatGPT sagde: A CDI (Capacitor Discharge Ignition) system calculates timing advance based on engine operating parameters, primarily engine speed (RPM) and sometimes load (e.g., throttle position or manifold pressure), depending on whether it's a simple or more advanced system. Here’s how timing advance is calculated in a CDI system: ? 1. Trigger Source Detection CDI systems use a sensor to detect the position of the crankshaft or camshaft: Pickup coil (often near a magnet on the flywheel) Hall effect sensor Optical sensor This sensor sends a signal each time the engine reaches a specific position, often before top dead center (BTDC). [ch9201][ch65039] 2. Measure RPM (Timing Between Pulses) The CDI unit calculates the engine RPM by measuring the time between trigger pulses: Shorter time = higher RPM Longer time = lower RPM [ch9889] 3. Use a Timing Map or Logic CDI systems use one of the following to determine how much to advance the spark: Fixed advance curve (analog CDI): Uses resistors, capacitors, or basic circuitry to delay the discharge based on RPM. Digital advance map (digital CDI): A microcontroller reads a preprogrammed timing map. Input: RPM (and maybe throttle/load). Output: spark advance angle. The idea is: At low RPM, spark is near TDC (Top Dead Center) to prevent knocking and ensure smooth combustion. At high RPM, spark is advanced (occurs earlier BTDC) to allow more time for the air-fuel mixture to fully burn. Example: RPM Spark Advance 1000 5° BTDC 3000 15° BTDC 6000 30° BTDC ? 4. Trigger Capacitor Discharge at the Right Moment Once the correct timing is calculated, the CDI circuit: Charges the capacitor Waits for the right delay based on the timing advance Fires the SCR (silicon-controlled rectifier) or transistor Discharges the capacitor into the ignition coil [ch8594] produces high-voltage spark [ch9989] Summary A CDI system calculates timing advance by: 1. Detecting crankshaft position. 2. Measuring RPM (time between signals). 3. Applying a programmed or analog timing advance curve. 4. Firing the spark earlier at higher RPMs to optimize combustion. Let me know if you want a visual diagram or how it differs from TCI or ECU-based systems. Du sagde: I want to know how the signal from the pickup coil relates to tdc. ChatGPT sagde: Great question — understanding how the pickup coil signal relates to TDC (Top Dead Center) is key to grasping how timing works in a CDI ignition system. ? What Is the Pickup Coil? The pickup coil (also called the pulse generator) is an inductive sensor that generates a voltage signal when a magnet or metallic protrusion (on the flywheel or crankshaft) passes by it. This signal tells the CDI when the engine is at a specific rotational position. ? Relationship Between Pickup Signal and TDC The pickup coil doesn’t detect TDC directly. Instead, it detects a fixed point before TDC — typically 20–40 degrees BTDC (Before Top Dead Center), depending on how the flywheel or trigger rotor is designed. Here's how it works: 1. Trigger Magnet Position: A magnet or metal tab is mounted on the flywheel or a trigger rotor. This magnet is placed at a known angle before TDC (e.g., 30° BTDC). The exact angle is fixed mechanically. 2. Pickup Coil Sensing: When the magnet passes the pickup coil, the coil outputs a voltage spike (AC signal). This voltage rises and falls as the magnet approaches and leaves — usually the zero-crossing or peak of the signal is used as the trigger point. 3. CDI Receives the Pulse: The CDI now knows the crankshaft is exactly at 30° BTDC (or whatever the magnet angle is). From this point, the CDI calculates a delay (in microseconds) to fire the spark at the correct time — maybe immediately (if it wants to fire at 30° BTDC) or later (say, 10° BTDC). ? Example: Let's say your pickup coil is triggered at 30° BTDC: At idle (1000 RPM), the CDI might delay the spark to fire at 10° BTDC [ch8594] waits ~3 milliseconds after the pickup signal. At high RPM (6000 RPM), the CDI might fire at 25° BTDC [ch8594] waits ~0.28 milliseconds after the pickup signal. So the pickup coil provides a fixed reference point, and the CDI determines spark timing by delaying from that known point. ? Visualization pgsql KopiérRediger Crank Angle (°) ---> TDC [ch8592]---10°---20°---30° [ch8592] Trigger magnet here [ch8593] Pickup coil signal sent to CDI [ch8595] CDI calculates delay [ch8594] fires spark at desired BTDC [ch9989] Summary The pickup coil signal marks a known crank angle (usually BTDC, not TDC). The signal is generated when a magnet on the flywheel passes the coil. The CDI uses this signal as a reference to compute when to fire the spark, based on engine RPM and timing advance logic. TDC is calculated by knowing the fixed angle offset between the pickup trigger point and TDC. Let me know if you want help calculating spark timing delay in microseconds for a given RPM and advance angle — it's a fun bit of math. Here is a link to the chat for anyone interested, It is a good half hour read: https://chatgpt.com/share/684b273e-5dc8-800e-99df-41a7802e9d39 Cheers, Finn Hammer |
|
Title: Re: That ignitor module Post by Needles on 06/12/25 at 16:21:20 Wow! It just occurred to me, after seeing the board for the CDI in the pictures, that many of these are well over 20 years old. I restore old Ham radios. One thing you ALWAYS have to do with old equipment is replace all the electrolytic capacitors. (They're the ones that look like little cans.) Looking at the picture, all the electrolytics I see are swollen. They've either failed, or are failing. They are also pretty cheap. Vibration and heat don't help any, either. On that board, I'd be tempted to replace the bad caps, along with any resistors that look like they've overheated, and see if that works. Electronics components fail in a specific pattern: movable parts fail first, like relays or switches. Next to go is usually the active components, like transistors or ICs, but those are pretty rare, since they are really pieces of rock with tiny wires printed on them. They're pretty durable, unless the capacitors shorted and burned them out. By the way, even potted units can be repaired. You just have to use a heat gun or air bonder soldering station to "overcure" the epoxy used to pot the components. It's a PITA, but when I worked for Motorola, we did it all the time. You can replace the epoxy, or just use silicone rtv when you're done. |
|
Title: Re: That ignitor module Post by Axman88 on 06/12/25 at 20:31:11 77585F5F79505C5C5443310 wrote:
Thanks for sharing the pulser signal trace, very interesting! I would hypothesize that the single reluctor bar that we see on the LS650 Stator is placed such that it will trigger the spark with the max. possible advance. (You could check this mechanically by noting when the pickup coil is aligned with the beginning and end of the reluctor lump, and seeing how many degrees of rotation from each of these points till TDC) https://www.thebonnevilleshop.com/ignition-timing/ The specs say timing for a LS650 should be 5 deg BTDC below 2000 rpm and 30 deg BTDC above 4000 rpm. ( page 194 for a 1987) https://www.scribd.com/document/362846253/suzuki-ls650-savage-boulevard-s40-1986-2009-pdf Then, I would guess that the engine will add delay, reducing advance, based on rpm. I would guess the ICM doesn't trigger on the first pulse, but waits until it has a good idea of timing. It might be that the ignition controller is timing between when it sees the beginning of the reluctor to the end of its passing, (which I would suppose are the negative / positive spikes in the signal you showed us), and getting rpms from that, or even looking at the ripple, which is a function of the poles on the 3 phase stator which are generating power pulses at regular spaced intervals. But, I would imagine that the rpm calculation come from the full cycle, which I would imagine is from negative going pulse to following negative pulse. We can see that in 1996, Suzuki changed things a little, the reluctor node for '96 until the eng of production grew quite a bit longer, which suggests to my mind that Suzuki was using the start and end of its passing for some purpose, and wanted a longer interval to sample. https://www.ebay.com/itm/146200094003 Can you show us the signal for the full cycle, from negative pulse to negative pulse? Also tell us the vertical scale on the scope trace, and about what rpm the engine was spinning at? Thanks again for sharing this investigation. |
|
Title: Re: That ignitor module Post by Axman88 on 06/12/25 at 20:51:40 I'm no expert, but I think the skinny red diodes are Zeners and have something to do with overvoltage protection. They might have sacrificed themselves. |
|
Title: Re: That ignitor module Post by FinnHammer on 06/12/25 at 22:26:56 5B62777B7422221A0 wrote:
That was my first thought, but i diode tested them, and all diodes show normal 0.6ish volt diode drop in reverse, so I figured they are ok. Needles: the igniter bit the grass along with the also electronic decompression relay after I had done some tig welding on the side stand. I was careless there, I see that now, but nothing bad happened when I welded the header last year, so I did not give it any thought before too late. I actually don't see any swelling on the 'lytics, but they are heavily smeared in with conformity coating. |
|
Title: Re: That ignitor module Post by zevenenergie on 06/13/25 at 03:14:03 It's obvious that you know a lot about electronics :) What is your opinion the cause of your damage to the electronics? Was it stray currents or what was the high frequency starting of the arc? As for chatGPT I was calculating something yesterday that I know a lot about, and it is astonishing that GPT gives results including calculations that make absolutely no sense without doubting itself, once it has lost its way it just continues with the mistake it made. So if you want to find out something that you know nothing about, you will never be sure. My experience is that it is unsafe to ask follow-up questions to answers. It is better to formulate a whole new question. GPT has difficulty combining mathematical problems with technical insight. And I have to do my best to formulate the question in such a way that GPT understands the ecenty of it. But it is better not to confront GPT with too long chains of thoughts. It is already a lot better than it was half a year ago and it is a wonderful instrument when it comes to language. But with technical matters you have to use it as a tool. It can be wrong and it will be. ( the same goes for me ) 8-) We are all poor people groping in the dark so to speak, except when I am sitting on the beach with my fresh orange juice. [smiley=vrolijk_1.gif] |
|
Title: Re: That ignitor module Post by FinnHammer on 06/13/25 at 04:42:24 My guess is the starter arc, which is several kV Cheers, Finn Hammer |
|
Title: Re: That ignitor module Post by Needles on 06/13/25 at 06:00:05 I'd still replace those caps. 20 years is long enough for some of the electrolyte to have evaporated, especially if the bike spent any length of time in storage. I'd bet most of those are out-of-spec, whether they're bulging or not. BUT, if the transistor and IC is shot, you'd be wasting your time. If worse comes to worst you can always get an aftermarket programmable TCI unit. I don't see a microprocessor. The HA17903 is the comparator, but the M59201P is listed as a linear amplifier. The circuit likely has a set spark advance determined by some of the resistors. That part is discontinued, but it is used in lots of ignition modules, like Yamaha 650 twins, golf carts, and many motorcycles and other small engines with TCI. That's just in case you end up having to actually rebuild it. ;) |
|
Title: Re: That ignitor module Post by FinnHammer on 06/13/25 at 13:19:46 Needles, Thanks for your input. I found that the M59201P lisyed as different things, microprocessor one place, probably also as linear amplifier, but I picked the micro as the most probable, but that just relays how I would have done it if I was in charge. Here is a simplified schematic, from the Haynes manual. |
|
Title: Re: That ignitor module Post by Axman88 on 06/14/25 at 10:18:42 I'm pushing your response over to this thread from the other regarding Rev limiting, for more discussion. This thread seemed more appropriate. 4D626565436A66666E790B0 wrote:
I have been wondering about the following for quite a long time, and had a hard time finding anyone who seemed knowledgeable about it. I also own a Suzuki GZ250, which is another, smaller, 4 stroke single. It's service manual contains a very similar simplified diagram of the Ignition Control Module, which I've attached. But the manual goes on to describe a test procedure, using a multimeter equipped with a Peak Voltage Adaptor, which is basically a capacitor to convert short DC pulses into a sustained DC peak value. The manual says that a healthy ICM will be putting out pulses of greater than 200V DC. I attached a pdf from the manual, which can be found here: https://redirect.viglink.com/?format=go&jsonp=vglnk_174992082767911&key=30854a6a3b193bb73f39bd2305adfa54&libId=mbwgjexd0100qqcr000ULjx3a05p6rawd&loc=https%3A%2F%2Fwww.gz250bike.com%2Fforum%2Fshowthread.php%3Ft%3D28&v=1&opt=true&out=https%3A%2F%2Fapp.box.com%2Fs%2Faaokv6nqyhq0hsw71req&title=GZ250%20Service%2C%20Owners%20and%20Part%20Breakdown%20Manuals%20-%20Download!%20-%20GZ%20250%20Forums&txt=Service%20Manual%20-%20PDF%20File I saw a similar simplified diagram for the ICM in a manual for a Honda Shadow I was working on, which specified similar expected pulse amplitudes, on the order of 200V DC, and Honda was calling their system a High Energy Ignition. Is this evidence that the Suzuki ICM for the GZ250, (and possibly the LS650) is actively generating high voltage pulses, or is the >200V DC just the result of the inductive kick when the primary circuit opens? What does the polarity of the meter in the diagnostic diagram tell us, if anything? |
|
SuzukiSavage.com » Powered by YaBB 2.2! YaBB © 2000-2007. All Rights Reserved. |