dwell reducer circuit for toyota COPs
 

Attached is the schematic and PCB layout.
IMPORTANT: Use a TS914 opamp instead of an LM324.

The values shown are for reducing the dwell by 1.5 mS, which is appropriate for OBD2 NA cars and '99/'00 cars. It will also help 99/00 cars because they can't drive 2 COPs at the same time. The Toyota COPs want ~2.1mS dwell, and the 95.5-00 cars have 3.6mS dwell, at 14.4V.

Circuit has been tested on a 99 and on a 97.

You can get all the parts on Digikey.com.

I could have used a PIC to do the same thing but I can knock together an analog circuit quicker than a PIC program. If you want to make a PIC program you should use a Schmitt input with 1.2V logic low and 3-4V logic high threshold, and the output must drive a 180Ω grounded load to >1.5V, with <0.8V logic low. The input and output are both logic high for dwell, fall to fire.

I also found out that because the COPs draw 8-9A each peak, (total 16-18 A in waste spark mode), compared to 5.5A for the 99/00 ferinstance, the factory wiring has a lot of voltage drop (like 3.5V!). It would help lots to put a 10,000 uF / 16V / 105°C (or 2x4,700 uF) electrolytic cap across the 12V and GND lines that feed power to your COPs.

The PCB layout is 1.1" x 2.4".

You guys will have to do a group buy for PCBs. There are suppliers that will do PCB layouts for something like $30 for 4x4". I can provide the Gerbers, "panellized" so several PCBs can fit in one standard size panel (dimensions depend on the vendor).

Very cool.

Quick question. You say that the 95.5+ cars have a 3.6ms dwell, is that factory dwell?

Hi Jason,

you have written, that putting a 0,000 uF / 16V / 105°C (or 2x4,700 uF) electrolytic cap might already help a lot. Do you think, just puting in that cap would already help to get them fired up (because my problem was just at startup with the EMU (if running - the COPs were fine).

Thanks

Dayum, thanks for the tip. I'm going to hopefully stop at Fry's on my way home.

Jason, would you care if I got that board design printed? I'm certain I have a source that can do them.



edit: read...looks like it's ok.

Holy shit. I couldn't even understand this thread. I'm a :noob:

Im a bit confused. Is this board designed to be used with the factory ECU to lower dwell? Who is this targeting?

I understand the addition of a cap across the gnd and +12V lines, but is that necessary for people with MS's?

I ask because if there is in fact such a huge voltage drop in the OBD2 cars, it will explain why I need to run twice the dwell on the MS for the COPS as people with the OBD1 cars.

it's targeting people running Toyota COPs on the stock ECU.

AH. very good.

The cap across the gnd and 12v as well? The way Ben posted, makes me think otherwise.

above my head....

How was what I posted confuse anything? Lay a freakin cap across +12V and GND.

If you have to run excessive dwell, I'd bet you have shitty +12v or GND connections.

Its confusing because you run a MS and own a 91, but this board and from what I understand this CAP mod is for people running the stock ECU. Based on the paragraph where the CAP mod is mentioned it seems its for people with the 99/00 cars, but im uncertain if thats just used as an example.

So my question is, why would you want to do the CAP mod on your car?(based on your post)

As for my excessive dwell, I dont know whats causing it but its def not my GND or +12v because those are connected to the factory harness with fat wires and I have tried everything short of new coils to solve the dwell issue.

From what I can tell, the CAP is for all miatas running COPs on stock wiring because of the large voltage drop. The large capacitor is supposed to allow the COPs to fire at full amperage instead of letting the coils just deal with the drop.

Chad is correct. the Vdrop is caused by insufficient wire size in the factory harness.

OHHHH.

Sweet. I cant wait to try this then.

I still wonder though.
How is everyone else running cops before this discovery?

Edit: man how awesome it would be if this solves my issue. Not really an issue, just requires a higher dwell but I dont know yet how it will affect in boost running.

The NA 1.6 factory ECU does not suffer from this issue (the one needing the circuit)?

I installed my COPS yesterday and they appear to be working very well. I'm using a link ECU but I would assume that has the same dwell settings as the factory ECU.

I checked the voltage drop on a 99, didn't on a 97. I wouldn't expect it to be any better though.

I don't know what the factory dwell is on the OBD1 NA's nor the 1.6's.

As for those not running this circuit at all, at redline you will be forcing the COPs to dissipate 5W or more extra in each COP - that is a lot of power in that small space. If the COPs don't overheat and fail outright, you will be shortening their life. However, the factory wiring voltage drop is so huge it's actually helping you by reducing the current and heat in the COPs. :D

Said voltage drop must be appearing elsewhere in the wiring harness and whatever is connected to it will be seeing some of it, as well as the EMI (electrical interference) generated by it. Automotive electronics in general is very robust and is designed to take it (this circuit for example will work down to 7V), but the COP current could interfere with sensor outputs. You are better off grounding the COPs at the chassis or block, just not on a piece of wire that is used to ground something else, like a MAP sensor :)). It would also be a good idea to get your switched 12V from the main relay instead of the wire from the ECU. Despite this wiring change, the 10,000 uF capacitor will still reduce EMI. And don't power the COPs from the battery, that's just stupid. :giggle:

Not on a 99/00 - the problem there is that the factory ECU coil outputs don't have enough capability to drive the COPs in parallel (180Ω to > ~1.2V).

For kicks you could try mounting 2 COPs on non-paired cylinders (like 1 & 2) and grounding the unused coilpack outputs with some thin wire. :)

The round things beside the wire terminations on the PCB are holes. A good trick to connect wires to a PCB while provideing stress relief is to have a hole right next to where you solder the wire in. The hole has to be slightly bigger than the wire insulation diameter. I will post a pic later describing it.

Not an issue for those of us running MS standalone style right Jason?

Jason, thanks for the information.

I'm currently running COPs on my stock 99 ECU with no issues. I definitely want to do this mod though.

Braineak, I'd be interested in the circuit board if they are being printed (although I'd be even more interested in buying a built product).

Jason

The wiring is an issue. You should connect the COP power and gnd as I described earlier.

Hi Jason,

could you eventually do a little scheme, which wire off the wiring goes where on youre device? (It's because I am a electronical noob - it's like reading chinese for me) Thanks again!

Greets

Jason, I understand you're using an AEM; for those of us with AEMs, is this a mod we should consider doing too?

Thanks

Rafa

If this solves my dwell issue Jason, I may just have to send you a little something something.

Ground of the circuit should be connected to ECU ground, or the ground provided for the factory coilpacks. 12v can be any 12V switched. It won't care.

Do you mean using the Toy COPs? Yes, for high boost.
If you mean the power and gnd wiring for the COPs if you are using the AEM ... that's a yes too.

Hookers and blow? :D

Largest caps I could pick up at Frys was 1000mfd. They're 16V 105C electrolytic. There were 4 on the shelf, so I got all 4; 1 to go across each COP. Best I could do in a pinch.

^ ebay has 10000 but they are not the type you can lay down, and are audio grade. Like 8 bucks. Largest I could find in radioshack were 4700.

LOL Digikey has 10,000 uF / 16V Nichicon HE series for <$2 each:

http://search.digikey.com/scripts/Dk...me=493-1546-ND

LOL plus 5 dollar handling charge and shipping ;)

I bet I could pull one out of an old computer powersupply.

Yeah dammit, $5 shipping.
I think I'll get several and see if people want them at $3 to offset. Everybody wins.

I went and picked up two Nichicon 4700uf at radioshack for 10 bucks, I think they will do.

Well damn,

Thanks for putting all the time into this Jason and posting the schematic.

I'm one of the guys having issues with the COPs and EManage Blue. I know you were able to get COPs running on Bob's EMB equipped '99 so I'm not worried about a circuit working with my '91.

So, basically, I need to find out what the stock dwell is set at for my '91? Assuming I can do that, what changes would I have to make to the circuit for it to run properly with my '91?

Holy shit, why did i just read every post and still have no idea WTF is going on...and i just finished my COP harness.
Hate to sound like Sam, j/k, but seriously, can someone explain this to me in a ESL (English as a Second Language) type of ordeal.
I am running MS standalone on a 94, and my COP harness is a direct PnP into factory harness. Where would i insert this capacitor? Do i need one for each coil, or pair them up, or each on his own? Sorry, but i really do not get this shit, should have stayed on school, i know.
Thanks

If you had a 10000MFD cap you'd install it between the +12V and gnd that runs behind the head to the coils, before each splits to 2 leads. If you had a pair of 4700MFD caps you'd install it one at each paired harness.

Ben, i could do either, i got pretty good with the soldering gun, lol.
I think I may get in on your GB thing and just do the one before it splits. I just was not 100% sure I understand all this. I think i got it now. Thanks!

My miata is a '96 with an AEM F/IC (piggyback). I've got my COP's, not installed yet. After reading this thread it seems as though it would be a good idea to make one of these doo-dads before taking the plunge.

Unfortunatly, im no electrician and don't know what to make of the schematics. Any pointers?

Ignore the schematics, read what Ben said to me, we are pretty much in the same boat when it comes to your question
either buy the 10000MFD or buy two 4700's...that will determine where you will solder this in
I bought the 10000MFD and will put it in the main harness before it splits

Sorry I couldn't resist; this post is signed by Dan but it sure looks like it was written by me :giggle: with the ESL remark included :bowrofl:

It aint pretty, but it will work and be hidden behind the head.

Yup, now did you wires those caps in series or parallel? Did you pay attention to the anodes and cathodes?

Nah, I just tossed them in whichever way was convenient.

edit: Parallel, and yes. Just FYI, arrows point to ground on those caps.

This is a good read for anyone not savvy on electronics.

http://en.wikipedia.org/wiki/Series_...its#Capacitors
http://sol.sci.uop.edu/~jfalward/ser...lcircuits.html

Lol, looks like a snake that swallowed an animal.

Thanks for the info everyone!

Just acquired my TOY COPS and connectors today, and will put the unit together tomorrow.

93 NA 1.6 on EMB stock ECU. Anyone else having problems with their COPS on this set up?

Just checking if the CAPS are installed at the right location on Brianack's schematic?

Mini,

I've got an EM-B, and unfortuately it doesn't want to start at this point with the cops in place. I haven't tried to mod the harness with the cap yet.... but basically, if it doesn't start first time with the cops. Jumper over of the ignition wires (removing them from the EM-B) and see if the sucker starts on the just the stock ecu.

Dave,

Thanks Steelrat, guessing we are the only few left with the EMB setup. Hopes you could find some solution starting it up. Át the meantime, I'll watch for you updates.

Minime,

There's been a bit of a discussion on this issue with the EMB. Popstoy over on Mnet was able to get the COPs running without issue on a '99 with EMB using Jason's circuit and CAP. My CAP from Ben should be arriving in the next couple of days. I'll try to get it hooked up this weekend and see if the CAP solves the start-up problem. We would still need the dwell reducer to keep the COPs from overheating at high rpm.

Thanks FHS, what about the jumper setting that would need to be adjusted?

Do you think this could be why my COPs aren't working when all installed on the car?

Shouldn't need to adjust the jumper Mimi.

Braineack: Are you having any printed? I'm putting a parts list together and at this stage it'll be a breadboard jobby unless you've got something in the pipe..

Cheers,
Tony

I never saw an answer on the question of the 1.6 dwell, so let me state that on my '92, I measured it at appx. 5ms running last year before I pulled the ECU to go MS.

BTW, a friend has asked if I'd build one of these for his car. Did the group buy on the PCBs ever materialize? If not, I'll host a real one.

Joe, if someone can prove that Jason's circuit allows 90-93 stock ECU and EMB users to run COPs safely, I will singlehandedly make your GB happen.

Well, it seems to me that the only real difference would be the stock ECU dwell. I didn't realize the later cars used a shorter dwell, so this schematic is not one-size-fits all. The board, however, could be.

Looking at side A as an example, the amount of dwell reduction appears to be a simple factor of the RC time constant specified by R as (1/((1/R2)+(1/R15)))+R4 and C as C1. If I'm solving the circuit correctly, assuming Vc must be 2.5, I get t=1.419, which is pretty close to the 1.5ms that Jason claims. It's probably within the tolerances of the various parts, anyway.

One alternative would be to do away with the parallel R2 and R15 combo, and replace them with a single 50k pot.
*or*
We could simply publish a chart of known component values to achieve various amounts of delay. Shouldn't be too hard to figure out.

Wait, that's not right. R would be 1/((1/R2)+(1/R15)), which is 10.476k, which would make t = 0.726ms. Clearly I'm missing something...

Still, the basic design is universal, just find the right component values to plug in and it'll work with any year.

Joe Perez....'Did the group buy on the PCBs ever materialize? If not, I'll host a real one.'

I'd like to get in on a group buy too. I've already purchased my other parts from digi-key.
How will I find out?

Lee

Dwell reduction
 

Is this thread dead? I have a cop kit coming for my 92, which I plan to run w/o MS at first. I need dwell reduction, I believe. Is anything happening? Joe?

When I get some free time, I'll draw the board so that we can price them.

Before I do that, is everyone happy with the design as it stands? Does anybody feel the need to put potentiometers in so that the dwell can be precisely trimmed? Means you'll need an o-scope to set them up.