dantheman

iTrader: (4)

Do whatever you need to do.

You still have my money for the latest version. Hang on to it, I trust all of you to make whatever changes are needed.

gsl-se addict

iTrader: (3)

Not sure, yet. I need to figure that out. My cost is the same as before, but since it is missing a key function, it would be fair to lower the price a bit.

Dom: Sorry we didn't catch the problem sooner. I am trying to see if we can somehow use the parts from the last design. However, it doesn't look possible. The problem is we need the output to happen some time before the something happens on the input. I have found a possible way to do it with logic, but it will take many ICs to make it work. The other option is a microcontroller.

Dan: Sounds good. I will do my best to correct the situation.

Eveyone else: Don't be shy. Let me know what you think. I want to do whatever that will make you guys happy.

Kent

Dom

Can I propose using a simple RC circuit on the J109 output:



to produce a wave like this:



that we could set 'Q3' (before it reaches 0V) to trigger 4538???

devitek

Just curious, but what about those of us that already got one? Personally, I don't mind constructing another one from parts, but I'd need the board...

Anyway, sorry I haven't been much help lately. I got an 87 a couple months ago and haven't hardly driven the 84. Not to mention the new baby and all. If there's anything I can help with, let me know.

MattG

iTrader: (2)

So, how long we talking on this more advanced board. Are we talking weeks or months.

I'm in a tough pickle, I really want to update my system, but I am far from that being on my top list of things to do on the 7, I HAVE to change the brake MC.

I am willing to wait a little bit longer, for the version you guys have been working on so hard. I dont doubt that it will never come through there is always constant work going on with it.

I think I will just wait and get the final version. I have faith that you guys will get it here soon.

Although I do think a offering a board to tide us over is a smart idea!

gsl-se addict

iTrader: (3)

Hey guys,

Dom: That would work if the pulses had fixed spacing. Unfortunately, they don't. If they did, a more accurate solution would be a clock (10kHz or so) and a counter. The counter could start at some value, count to zero, put the output high. When the J-109 went back high, we could force the output low causing the 2nd gen coil to fire. Basically, we want our output pulse to end where 1 is on that figure. We want our output pulse to start 2-4ms before that point. The only way I can see is to measure the previous pulse and use that as an esitmation of the next pulse. Count down until say 4ms remain, put output high. Then bring output low when the J-109 goes high. I think it would work, but takes either a bunch of logic/counters or a microcontroller to do the job.

Dave: No problem. I know you all have a life besides this. I do appreciate all the contribution all of you made. How about I just give you guys that helped testing the board you have now? I can then put what you paid against the next design. We can figure out the details then if that works for you.

Matt: Here is what I am thinking. It doesn't appear that the new design will be able to use these boards (for sure) and probably not most of the parts I have (except Vreg, wire, connectors, and maybe boxes). So, I think it would serve everyone best if I just send you guys these boards at reduced cost. I would like to get at least enough to cover my costs, if possible. Now since some of you guys have paid a long time ago, how about I offer you guys a double credit between what you paid and the reduced cost towards the next board? That way you guys can have something now, and spend little or nothing when the next board is complete. If you just want this board and not the advanced one, I could refund the difference between previous cost and the reduced cost.

As far as time, if microcontrollers are used, I can't do much until I return to the US (Oct). I then need to choose a contoller, build/buy a programmer, write the code, then fully test on bench and in car. Once it works, I would have a few boards made for testing. If successful, the remaining boards/parts would be ordered. With a micocontroller, it would be possible to have a CAS input as optional and possibly 2nd gen trailing ignition. The downside is if there needs to be code updates, you guys would need to send the chip or box to me to reprogram. It is also getting outside of the scope of this project. It may be easier for people to get a MegaSquirt if we go this complex. I could probably make something cheaper than the MS, but the MS already has been tested, has software, etc.

If I can come up with an easier way to do it, I could still design the board here and have Dom or someone build on breadboard and test it.

Anyway, it is up to you guys. I will continue to find a simple way to make it work. If I can't, I need to know if you guys want a real advanced setup (microcontroller or something). I am willing to go along whatever path you guys want.

Thanks,

Kent

Kentetsu

Kent- I think you're worrying too much about some things. I don't think that there is anybody here who didn't know that this was all experimental when they purchased the TT units. I feel that I got my money's worth either way, and if a new and improved version comes out and I feel the need to upgrade, then I'll happily pay the price for it. There will always be improvements on something like this. That doesn't mean that the previous product was no good and you should offer refunds or anything though. Anyway, just my two cents...

gsl-se addict

iTrader: (3)

Yeah, Ken. I was mainly talking about this current batch. People paid for a version with the max pulse limiting, but that function does not work correctly. I feel that it would be inorder to drop the price to just above what you got yours for. This circuit is nearly the same with the exception of the professionally constructed board. There have been people that paid several months ago (before the issue was discovered). I feel it is the least I can do to offer double the difference for what they paid and what the reduced price will be as a credit on the next batch (or a refund of the difference or a total refund if they decide they want no board at all).

I think something like this is a good plan for everyone. They get the circuit, I recover my costs, and it saves me from having to bring it all back with me when I move. I think it is probably a win-win solution. As it stands, the circuit works just like the last batch. Performance wise, it probably can't get much better. The only problem is with the coil packs getting too hot.

I also have a Mazda style compression tester in the works. It will be most likely microcontroller based. I am not sure if there would be interest or not, but I want to at least build one for myself as a project. If there is demand, I will have them available for everyone.
Realistically, I probably won't be able to finish the next batch of TT until I return since it will likely need a microcontroller. If I can come up with an easier solution, I can work on it here.

So, that is where we are right now. I am sorry that the new design did not work out. It is too bad that we spent so much time until the problem was discovered. I guess that is all part of making improvements. I must say thank you once again for all those that helped with testing and contributed to possible solutions of the problem. I couldn't have done it without you guys.

I will set a new price and start sending eveyone a PM about it tonight most likely. I do need to figure out a ballast resistor that will help with the heat problem, though. I could use some help testing it out.

Kent

t_g_farrell

iTrader: (1)

So if we had built one of the simple circuits that you posted way back, would we just need a ballast resistor added onto the coil to keep it cooler ?

BTW, great project, I learned a lot about how this all works. Maybe there's a cheap way to do this new circuit without an expensive processor and code. Since I code for a living, I try to stay away from it if I can :-)

gsl-se addict

iTrader: (3)

Yeah, t_g. That should work. The most ideal place would be just to replace the 2nd gen ballast resistor with one with higher resistance. We would just have to find the most ideal value. We want it high enough to keep current lower at low RPM to help with the heating issue, but if too high, the coil won't fully charge at high RPM.

When the coil enters current limiting mode, the current is about 7.5A. I am not sure if it controls the current or if the control is based on adjusting the coil - based on the stock configuration. I bet it is just on current to account for variances in battery voltage, coil resistance, and so on. The ballast resistor would reduce the time spent in the current limiting region since the coil will charge slower. The stock ballast is 0.35 Ohms, I believe. We could maybe try like 0.75Ohm or 1 Ohm. We would just choose the smallest value that still keeps the ignitor cool enough. It isn't the most elegant way to do it, but it should get the job done.

I am glad that you are learning a lot from this project. I know what you mean about the code and stuff. I would like a more simple solution, but I don't know. I know of another way to do it, but it would require timers, flip-flops, logic, and still would be difficult to change the charge time with RPM. A PIC or similar could do all this in one package. It would save space, soldering, and the cost would be about the same (some PICs aren't too much more than the several ICs it will replace). It just is a little more complex than I was hoping for, but it is up to you guys. I will follow the whatever you guys want.

Kent

gsl-se addict

iTrader: (3)

So, I have offered the guys that were buying these boards a few options for the route they want to take. The results are mixed. Some didn't want to wait (I gave full refunds), some want to just wait until the next design, and others would like this design at the reduced price and get the new design later. If you haven't responded, please do so. That way I can get a count of who wants what. For those of you that I put on a waiting list, spots are now open for all of you if you want this current design (or I can keep you on the list for the next batch).

I have found a couple sources for ballast resistors. In terms of selection and pricing, the US is the best place to get them. I can get them in the same style as the ones in the 2nd gen pack or I can get a different style in which we could put in series with the current ballast. Getting the 2nd gen style and then swaping is probably the best solution. It is mounted against the aluminum backing for a heatsink. I need to measure the 2nd gen ballast to see what kind of power rating resistor we can get in there (I believe the stock is 0.35Ohm, 5W). I am going to assemble a few TTs and ship Monday. These are to the people who said they would like this current board. Once we know a good ballast, I will order and ship to my mom's house. She can then send each of you a ballast resistor.

So, this is what I need:

- If someone could try a couple different ballast resistor values, that would be great. It can either be with the circuits I shipped last Sept. or this current batch. Some sort of coil temperature vs. performance would be ideal. Perhaps we can find a resistor that pretty much fixes the heat issue without hurting performance.

- Once this is done, I will order the resistors. I would like to try to get a fairly good count on these. I imagine that people with the old circuit will want one, the people who buy this circuit will get one, and probably the couple of people who built their own TT would like one. I get a price break for 10 resistors and a further break at 100. Perhaps I will just get 100 to make sure I have enough. I imagine if the resistor makes the coil pack reliable, there may be a good number of people who will want the current design.

Anyway, I am doing my best to get this situation worked out for the best. I am open to any suggestions if you think I should be doing things differently. If you have any questions/concerns, feel free to pm me and I will help you as much as possible.

Thanks.

- Kent

Rx Seven

iTrader: (12)

Well I have one from the Sept batch so I would be interested in a resistor. I paid for just the guts of a new version (no box) but I guess I don't really need it.

I wouldn't mind waiting until the next batch come out but if you need the funds I would take it and pay for a box.

But I would still buy one from the next batch whenever those get worked out

gsl-se addict

iTrader: (3)

Alex,

It is up to you. The current boards don't offer any real gain over what you have except they could last longer (more resistant to corrosion, elements, ..) and are prettier. :p Since you are looking at getting the next design, I would probably just stay with what you have. I will apply what you have paid toward the next design. I will get you guys setup with the resistor.

Kent

H4Inf

Hey Kent,

I can try some resistors if you like - what needs to be measured?, and what exact spec resistors do i need - have a look if they are available at www.dse.com.au or www.jaycar.com.au - these are my local stores :-)

Cheers,

Paul.

gsl-se addict

iTrader: (3)

Hey Paul,

That would be great. I will look through those sites to find what we need. The resistors are 5W, 0.35Ohm in the stock configuration. I am thinking using about 1Ohm may do the trick. Basically, any increase in resistance will help the heating. If the resistance is too high, the performance may decrease in the upper end. It would be nice to measure coil temperature. That jaycar site has some thermistors that may do the job. They change resistance based on temperature. We could attach to the coil pack, idle the car for some amount of time, read resistance. We can later convert the resistance to temperature using the manufacturers datasheet.

Basically we are looking for some value that will keep our coils running cool. Once I find what they have, I will give you some part numbers.

Kent

gsl-se addict

iTrader: (3)

Hey guys,

I think Paul is going to do a bit of testing with the ballast resistors. Is there anyone that can test as well? I think RadioShack may have a resistor that will do the job if you don't have other electronics shops around. I am assembling a couple of the current circuits for the people that want them. It is a bit more difficult to ship from my new place. The post office closes before I get home from work. I will have to either take a different bus into town and use that post office, or I can mail on Saturdays.

I tried to find resistors that are a drop in replacement for the 2nd gen ones. It is a bit hard to find in that configuration. I think the easier way is going to put the new resistor in series with the old one. We can mount the resistor on the other side of the coil pack. A nice advantage of this is there is more room on that side. This means we can use a resistor with a higher power rating. We may need a thermal epoxy or something to mount the resistor and allow for good heat transfer. We also have to ensure that eveything is connected so that it cannot sort out against the case. Heatshrink tube should take care of that part.

Kent

H4Inf

Hi Kent,

I should be able to test next week :-) Got to run off to work now!

Paul.

gsl-se addict

iTrader: (3)

Hi Paul,

No problem. I know you have your work and all. When you get ready, let me know. I can make a diagram or something on the placement of the ballast and maybe how to setup a thermistor if we want to try to get temperature data as well. We can start with about 0.5Ohm and work our way up to say 1Ohm or so. I am not sure what we will need exactly, but I am thinking something in that range should work. The exact value is probably not too critical. It would be nice if something like 0.5Ohm or 1Ohm would work nicely. I say this because people can get 1Ohm (10W, I think) resistors at RadioShack in packs of 2. They could use one resistor for 1Ohm, or put the two in parallel for 0.5Ohm. I think that it should probably work nicely. Certainly a small investment if it will save our coil packs (I think the RadioShack ones are like $1.59 for the 2 pack).

Kent

Kentetsu

If I can get some free time soon, I'll give this a shot too Kent. Busy weekend coming up though...

gsl-se addict

iTrader: (3)

I shipped a couple more TT boxes out the last 2 weekends. I am going to build a couple more for those people who said they wanted this version. Those who want this version should speak up soon as I am going to have to start shipping my supplies back to the US. I can do more when I return, but I will be in the process of visiting family, moving, and starting a new job, so I will probably be kind of busy for a couple weeks.

Thanks.

Kent

MattG

iTrader: (2)

How Do we know If we are part of those that have been shipped.

If I am Not, I would like to be!

Thanks Matt

gsl-se addict

iTrader: (3)

Hey Matt,

I sent all of you guys a pm a couple weeks back. I don't think that I received a reply from you. I hope I sent you one. Basically, I listed the options for those who have already paid:

1. Get the current version at rfeduced price plus and extra credit towards the next design
2. Wait until the next design
3. Full refund

I have had a few that have said they wanted this design now. I have sent the majority of those. It has only been a few people that have said this. Most want to just wait and others didn't want to wait and wanted a refund. I have a whole bunch of these circuits remaining for those who want one. As I said, I am going to start shipping things back to the US soon, so I need to know if you want one. Otherwise, there will be a delay until October or November. Matt, I will get you one ready to ship this weekend. Everyone else, just send me a pm if you want to get in on this batch.

One thing: This version will require a resistor to be added to the 2nd gen pack to help with the coil heating problem. I think a 10W, 1Ohm resistor will do the trick. RadioShack has them in 2 packs for like $1.59. It would only take a quick soldering job to do. I don't know the exact value that will work best since I need someone toi test in the car. I can't reproduce the pulse that the J-109 does exactly with the equipment I have.

Kent

MattG

iTrader: (2)

I never got a PM, but thats cool

I will go ahead and go with choice number one.

"Get the current version at reduced price plus and extra credit towards the next design"

Let me know if you need anything else. I assume it will come with instructions as to where to add this resistor.

-Matt

gsl-se addict

iTrader: (3)

Hmm. That's strange. I thought I had you on the list. Anyway, I will get you setup. For the ballast, I can do instructions and add a pic if someone does the mod and takes a pic. I could also take a pic of my coil pack and add a diagram of the connection. Should be fairly easy to do.

I will probably just assemble say like 10 more of these for now and then send the other stuff back. I can then send whatever TT boxes that don't sell now back to the US later. For those of you that have paid and didn't get a pm or you haven't paid, but want one from this batch, just send me a pm and I will get you setup.

Thanks.

Kent

H4Inf

Hey Kent,

Perhaps I should have made time to test the balast resistor. My coilpack just died on me hehe. Once the sun comes up tomorrow I will have a thorough look over everything, but as far as I could see tonight, nothing was loose.

I managed to drive home fine on trailing ignition, however there was absolutely no torque - any steep roads just did not work out, the car died lol.

I'll probably switch back to the stock ignition tomorrow - I should retard the timing by about 5 degrees when I do this, shouldnt I ?

Could you send me the full details on what I need to do for balast resistor. I can get photos when I do the mod and make up a guide.

Cheers,

Paul.