TCI rebuild thread.

Following on from this thread: http://www.xjbikes.com/Forums/viewtopic ... rt=15.html
tumbleweed_biff has sent me his dead TCI to (hopefully) repair.



So we have this:




The already diagnosed blown driver transistor:


Underside:




Time to remove the ceramic thick film board and the electrolytic capacitors:



The main part of the board is fine:



Not so good for the board under the heatsinks:


So that's it for tonight, tomorrow will be time to clean up the board with acetone and re-solder the existing components in the undamaged area.

The damaged are will need a lot of work, as once the board has been burnt it will conduct instead of insulate causing future failures. Options to consider are isolating that section, or carving out the board.

I may repair the undamaged section first, and make the transistors a separate board/heatsink. That would be the quickest way to test that the circuitry is functional before committing too much time.

Common faults on these TCI units are normally fractured solder joints and aged/leaking electrolytic capacitors. This sort of failure will probably have an external cause (currently unknown).

 

Masochist.

Good job though, keep it up. This should be in "FAQ Suggestions."

 

how much do you charge to do something like this? i cant find anyone here in the states close to me who wants to even attempt this

 

awesome. also wondering about what someone would charge for this. im sure that if you wanted to, you could have quite a few people coming to you wanting work like such.

keep it up and keep up with the updates!

 

I am not charging for this, for a few simple reasons.
I offered my services.
This repair (if sucsessful) would not be viable for an hourly rate anyway.

Once you charge and accept jobs, you are obliged to produce the goods rapidly and honor a warranty on something that is 30 years old.

That means I can (since my last post in this thread) get out for a ride to a pub and eat lamb (on the spit) rolls while listening to a bush band!
as well as a few social events.


I will not be looking for any jobs until I have my Pumba project past stage 3.

 

When Darkfibre had looked at this he told me it was not economically viable to repair but that as a project to write up and such it was worthwhile to him with no guarantee of success but a good probability in his estimation.

I just recalled seeing something about such in the past and a search found that at one time, Robert offered a service to do this but I don't know if he still does. Also, I doubt he would resurrect something this far gone. If he no longer does, perhaps he is willing to publish his learnings from the various TCI's he has worked on? For a long time he was accepting the donation of dead TCI's for learning and repair ... I suggest a PM to him might be a good way to go. He said at the time ('09) that he hadn't figured out how to get into the epoxy'ed box of the xj700, ... my answer was a dremel type tool with a cutting wheel to remove and then I will have to make a custom box - the old one was a total loss ...

I also recall someone mentioning a business which did such, but I can't find a reference to such. As I recall, it referred to dealing with cold solder joints, etc., as opposed to a catastrophic event such as mine experienced (which I suspect was a result of the PO using automotive coils and allowing them to hang loose and short against the engine block - or maybe any of a dozen plus hacks of the wiring harness).

For those of you with the funds, I suggest trying Chacal to see if he has any for your bike in his stock. For me, financially, Darkfibre's offer to try as a project was a God-send. I hope his efforts and write-ups will be a blessing for members of the forum for years to come. In my case, my electrical/circuit skills are about a 0.5 out of 10, logic 9 of 10, soldering 2 of 10, steady hands 2 of 10, so such a project is far beyond the scope of my ability even with such a write up.

 

well either way i would love to see how this project turns out.

 

Board is now cleaned up (ignoring the burnt areas) and resoldered. Most of the caps have been replaced, I just have to get some 0.47uF caps that I thought I had on hand

Should be able to then bench test it without the output transistors to see if the unit is functional very soon. Then its time to work out the neatest way to mount the transistors.

 

How about a parts list with part/component numbers when you are done? Great post and take/post lots of step by step pics!

 

Bench test without transistors was successful.

Bench test with the modern VB921z ignition driver transistors also successful.

Of to buy some connectors to make an adapter harness to test it on Pumba (750 Seca).

Results and more pics later.

 

OK.

New capacitors and a resolder, powered up with no smoke!




And we have an output pulse to drive the transistors




So its time to cut away the tracks around the burnt area.



One of the good things about the VB291 ignition driver is the fact the case is insulated, unlike the original transistor. Some odd modifications due to the lack of surface tracks:



Tracks jumped over with tinned copper wire and silicone tubing.



And done (yes, it's messy )

 

WOW!
Messy? Who cares?! I don't see how it could have been neater given the damage. The fact is, you resurrected it and upgraded the transistors at the same time. Something far beyond my abilities.

 

http://www.youtube.com/watch?v=1ejDsxEODiI
http://www.youtube.com/watch?v=uow4Ndget4U

 

For something that looks as bad as that did when you started I never thought it would work. Good job

MN

 

Now that is is working, I just have to support the components that I have floating in the air or over time the vibration from the bike will cause fractures at the solder connections. I just have to get my hot melt glue gun back from a mate.

Tumbleweed_biff, do you have the cover still as the one I am using on it at the moment is not a spare.

NOTE: Do NOT swap TCI units between a 700X and a 750 Seca as the wiring connections are different - I made an adapter to do this.

For test purposes I made a harness adapter, and found the XJ700X TCI to perform close to standard over about 2,000rpm but very soggy under that.

Adapter:

 

Wow! Nice job man!

 

Darkfibre,
The box was a write-off. Not only was part of it melted from the heat event, but the thing was glued to the board all the way around and I had to cut the box apart with a Dremel to get to the board. I expected I was going to have to make one out of something. It will probably look like $@#% as I am not good with precision stuff. I was hoping to find a project box about the right size I could adapt ...

 

Wow this is awesome.. I am in aw of your know how..

 

I believe I may have mentioned this on another thread, but would it be possible or feasible to build a "new" TCI?

 

All things are possible.

The main hurdles I see are sourcing the main chips and the program and equipment to burn the program onto them. If you had that you'd probably be able to do it.

 

I am not sure, but I got the impression that there doesn't appear to be any "programming", that it is all hardware based? Robert is probably the best to ask as I believe he knows this better than anyone.

 

...or adapt another TCI from a similar air-cooled, 4-cyl, 2-valve/cylinder, 2-inductor pickup, pre-programmed ignition advance engine (doesn't even need to be a Yamaha)? The minimum wiring diagram shows 8 connections to the TCI. Is there a central repository for TCI connection schematics?

 

Much as I like this idea, keep in mind that

is a starting point.

Before testing this TCI on a different model, I first made sure the coils had the same current draw (primary resistance) and the crank trigger had the same shape and pattern.
If you are considering doing this, post up a thread and make sure I see it, as it IS possible in some cases, and I have some ideas.

As for making a new TCI, most aftermarket injection systems can be used for spark only, and the first thing that has to be solved is the fact the standard trigger system (one pulse per revolution, duplicated 180 degs out). All the new systems seem to need more pulses per revolution.

Someone produce a bolt up (plug and play) 36 - 1 trigger wheel and sensor and then things will move forward quickly.

 

FWIW .. Hyperpak in New Zealand have this to say about their TCI for bikes
"Hyperpaks offer greater reliability than the OEM product while supplying much improved bottom to high end performance, due to a more efficient timing curve and improved dwell time resulting in more spark energy, this also alows the use of aftermaket HT Coils. The Hyperpak will run on both standard and modified engines without having to touch jetting or change HT Coils. " are you contemplating modifying the curves ?

 

That may have started with me. It is my belief that the only "programming" in the TCI's is the laser trimming of resistors on the ceramic sub-board.

darkfibre, I was talking with you a while ago about some TCI stuff and I thought that I had found that the input triggerings functioned like a flip-flop. You were going to double check:

http://www.xjbikes.com/Forums/viewtopic ... rt=30.html

Now that you've got your test rig operating again, what did you find? Did it have to alternatly trigger each channel or could you repeatadly re-trigger the same channel?

Good work on the reconstruction!

Burnarr

 

Each channel seems to be Independent, I treat it as two separate systems sharing the board.

Unit is ready for its trip home.

 

I await its arrival with bated breath.

 

Thanks for checking. I don't know what I was looking at then... It wasn't the watchdog circuit. I'm sure it was something on the driver logic. Haha! That's why I should take notes!

Anyway, thanks again, and when I do get around to firing up my test rig again, I'll make sure I pay more attention next time! :idea:

 

According to their ebay offerings, Hyperpak fits the stock XJ series for a little over $200. Anyone have any experience with them?

 

You can fake multiple signal pulses from a cyclic signal source with a toy from Split-Second. He has a lot of neat signal modifier and boost/fuel controlling modules I've played with over the years on the Genseis Coupe 3.8 GDiT project.
I have one of these laying around the shop somewhere. You could use it to produce whatever tach signal you need for the firing trigger: https://splitsec.com/product/ta2-100-tach-adapter/
I've found a few fairly cheap options to produce 2-8 cylinder firing including using Speeduino with a couple of Bosch igniters. I've been pondering the possibility of a mini-turbo kit on my 550. Need to get the bike done first. LOL

 

For the money, that gizmo is at least halfway to one of several plug and play options:

https://www.xjbikes.com/forums/threads/aftermarket-ignition.45931/

 

Hello guys i don't know is this correct thread ask about my question about TCI/CDI box but i bought few months ago 1992 Yamaha xj900 to my cafe racer project and just wondering where should i look a new TCI/CDI box or do anyone have wiring diagram from that box? Is there any universal TCI/CDI system what you prefer to use with that bike? Bike is running nicely but just in case because i could not find similar parts from ebay or enywhere.

 

there are a couple of tci companies out there that sell aftermarket tci they run 300 to 400$ usd.

there is a web page about fixing the tci list of parts needed
https://www.xjcd.org/general-electrical

 

Hi All XJ900er's, I have been rebuilding my XJ900F on and off for a few years, so I decided to get to the guts of the TID14-37, here is the first design of my version of the unit. You will need to use the old HA1825 and the HSA-10B, I will keep you posted. Regards Arto.

 

Not knowing what you mean by the HA1825 and HSA-10B, what processors are you using? Is this going to be a programmable load/speed unit, ie mappable 3 dimension?
If there was a cheap commercially available mapped system you might sell a few, but there aren't thousands, or even hundreds of XJs sitting in sheds waiting for one of these. But, if you can make one for a four cylinder Yam, maybe you could roll it out to others....

 

Hi, this a personal project for my bike to get it operational, I know what it's like when these units have to be repaired and the mess the board becomes after soldering and re soldering internal bits a few times, with that added gunk they smear on every component. I have been through this twice before and the board looks pretty bad. So I thought I would share my work with the many other repairers that need a fresh board with new parts on it. The HA1825 & HSA-10B are the two main chips in the TID unit that you cannot get off the shelf, as I am still reverse engineering the functions of these chips, I hope I will be able to use other ways to create their functions. As for now I am creating a PCB board and the BOM list so it can be rebuilt as new plus the 2 chips from the old board, as these chips do not generally go bad.

I have already created a few clean Schematic diagrams so I can see how these units operate, I will share everything when I have tested a unit on my bike. Regards Arto.

 

That's pretty cool

 

Hi All, this is my latest picture, the last iteration had a small error in it, this should be the final pic, I am currently working out how the HA1825 works. Regards Arto.

 

Excellent! Almost time to get the old breadboard, sockets and wirewrap out!

 

Hi All, I have a picture of a die cut of the HA1825P chip, I do not remember where it originated from, the time line was 2018, I have looked for source online with no luck. I am still deciphering the chips functions. Regards Arto

 

https://www.xjbikes.com/forums/threads/ic-ignitor-wip.118221/page-3