Topic: cb750 intermittent power loss on cylinders 1/4 electronic Martek ignition

[catsoup]

I have a 78 cb750f which is basically stock (carbs, filter, exhaust) however it has Martek 440 electronic ignition and coils. 23k miles, at least 10k of which are known to have been on the martek, I dont know when it was installed by po

Few days ago, bike had been running fine, parked it in town, got back on a few hours later and started riding home, without warning I lose power from cylinders 1 and 4. Park it, take off the tank look at the coils, notice that the black/white wire going into the coil for 1 and 4 had become somewhat loose. plug it back in, all is well. I ride about 100 miles over the next few days, and then suddenly lose power from 1 and 4 again. I checked the coils / connections, seem okay. try taking it out again, and the same thing happens after a few thousand feet.  so, today i decided I would try to figure it out. here is what I have done.

Read manual, and used the search function finding quite a few people having had similar problems, specifically this http://forums.sohc4.net/index.php?topic=70895.0 and this http://forums.sohc4.net/index.php?topic=85004.0

removed coils, cleaned all connections at the coil, removed caps and tested coils as outlined here http://www.sohc4.net/index.php/howdoitestmycoils/
I measured 5 ohms on one coil(2/3) and closer to 4 on the other (1/4) with a crappy multimeter (culprit?)
cut the spark wires back 1/4inch to insure good connection when screwing the caps back on.

removed and examined plugs. photos here http://www.flickr.com/photos/hanback/7537679806/#in/photostream
note I cleaned plug 4 before I thought to take photos, it was nearly identical to plug 1, with slightly more soot than the others.
Cleaned all plugs and reinstalled

opened points cover and examined martek 440 unit for any obvious signs of damage, found none. removed and reinstalled as outlined in the martek440 instructions found on sohcChopper. ((((Is there a way to test these units? ---culprit? )))) Examined spark advancer unit, small possibility of one of the springs being less stiff than the other - marginal difference. (culprit?)
followed wires from ignition to the connections at the rear brake indicator, disconnected, cleaned, reconnected.

After all of this I started it up, it idled and revved fine for 5-10 minutes, and then it lost 1/4 again.

I can tell that it is 1/4 each time because they are noticeably cooler at the header.

I rebuilt my carbs less than 1000 miles ago, synched them less than 300 miles ago. Adjusted valves and cam chain at the same time. The plugs are no more than 1500 miles old, and are the ND ones. cleaned the air filter recently. It had been running in tip top shape for a while now. I was going to change the oil this week as I am at around 1300 miles on it, and had planned to ride a few hundred before this issue came up. I have been careful with routine maintenance, and have tried not to overlook anything.

So, as I have no extra coils, points, ignition systems or anything like that, and very little money to spend on this, is there a way I can isolate the issue further? Are there connections I should check that I have overlooked, is there a way to test my ignition system? I feel like I've hit a wall.

[catsoup]

one thing i forgot, not sure how relevant. about 300 miles ago I noticed the bike was running a little lean, I had synched the carbs not too long before that, including adjusting the mixture screws. I would get some popping on deceleration and occasionally it didnt feel like it had the power I was accustomed to, so I turned the mix screws out 1/2 turns and checked / synched the carbs again. Fixed it all up, am i running too rich now? this is probably not related


edit update

I checked the resistance of the coils between the two "gozinta" terminals (black/white and yellow or blue), removed from the system, using a fluke 88 automotive multimeter, and got around 3.5-3.6 ohms on each coil. Are these coils supposed to be that low?

they are marked 4-81 on the bottom, if that makes a difference. (manufactured march 81?)

[catsoup]

another update.
found this thread http://forums.sohc4.net/index.php?topic=17711.msg184209#msg184209
TwoTired recommended this test

1- Disconnect both the Yellow and Blue wire at the coil.

2- Check Black wire TO the coils with the key on.  Red probe on Black.  Black probe on a green wire or engine case where there is no paint.
If you don't have Battery potential (12V) or very close to it (do a compare to that at the battery terminals), the coils and ignition can't work.
Then fix the problem here.  Otherwise...

3- Reconnect the Black wires to the coils.  Now measure the disconnected Yellow and Blue wires each with the red meter probe and the Black probe on a green wire or engine case where there is no paint.
Again you should have Battery potential (12V) or very close to it.  If not, and you had voltage at the Black wire on the previous test, the coil(s) is bad.

4- Next lay #4 connected spark plug on the motor or wherever the spark plug body can touch a battery return path (engine, frame, Green wires, etc.)  With an extra wire, used as a temporary jumper, bridge the Blue wire to engine, frame, Green wires, etc.  Each time you remove the jumper, the spark plug should spark at the gap.

5- Next lay #3 connected spark plug on the motor or wherever the spark plug body can touch a battery return path (engine, frame, Green wires, etc.)  With an extra wire, used as a temporary jumper, bridge the Yellow wire to engine, frame, Green wires, etc.  Each time you remove the jumper, the spark plug should spark at the gap.

If you got the plugs to spark, you can rule out a coil problem and focus on a DYNA issue.  If not, there could be coil, lead, plug cap, or spark plug issues.

6 & 7- Assuming tests 4 and 5 were positive, reconnect the Yellow and Blue wires at the coil, and find out where these wires connect to the DYNA.  If you can disconnect there, repeat test 4 & 5 with the wires that would have connected to the Dyna.
If you still got plug sparking, most likely the DYNA or its attachement electrically to engine, frame, Green wires, etc. is compromised.  If not, you have an interconnect problem between Dyna and coils which you must track down and correct.

I followed these steps and found the coils seemed to check out fine, passing all tests.
one question, in step 3, I measured from the coil side of the yellow and blue wires, not the wire-harness side. they checked out fine this way, but showed no voltage when measuring the yellow and blue from the the wiring harness. I think this is correct, just clarifying.

So this seems to focus me on the Martek ignition.
could the slightly suspicious spark-advancer spring be culprit here, or should I face facts and start looking for a new ignition.
Or, should I get a set of points and be done with electronic ignition. (can I do this, are my coils 3.5 ohm?)

[Spanner 1]

Need a Martek  1-4 pickup to swap out, no other option at this point IMO. 

[1080]

http://forums.sohc4.net/index.php?topic=33317.msg344309#msg344309

[wrenchmuch]

The Martek's are seal units . There is no way to replace components inside without destroying the unit . It sounds like you have a fault that is temperature dependant . Coils can experience the same sort of fault but out of the two devices the Martek is the more likely culprit . You can try checking the Martek's ground but a problem here would likely affect 2/3 as well . Martek made 2 types that I'm aware of one came with just the point replacement unit and all fit under the points cover . The second came with the same unit as the first but also had amplifier (black box) that you mounted externally from the points cover . The box allows the use of coils with lower resistance ( higher load ) . Your 3Ω coils will be too much for points . The points won't last very long . There is still the possibility that you have a wiring problem somewhere . If there is it would most likely be the Yellow or blue to the coil (can't remember witch colour is 1/4) from the points . If you find the Martek is pooched I would be interested in buying the rotor from you . I'm not sure but if you have the amplified version it might be possible to use the black box with standard points to reduce point current like Hondaman does with his ignition amplifier . A  PM to him would confirm this . He also sells a resistor pack that mounts to the coils to reduce their load . This would make them safe for points .

[catsoup]

Need a Martek  1-4 pickup to swap out, no other option at this point IMO. 

Is this actually an option? the unit appears to be completely sealed.

http://forums.sohc4.net/index.php?topic=33317.msg344309#msg344309

I had read this thread, and tried poking at it while it was running to see if I could simulate the problem, but couldn't get it to fail that way. I am trying to isolate the problem as much as is possible before I buy any parts.

The Martek's are seal units . There is no way to replace components inside without destroying the unit . It sounds like you have a fault that is temperature dependant . Coils can experience the same sort of fault but out of the two devices the Martek is the more likely culprit . You can try checking the Martek's ground but a problem here would likely affect 2/3 as well . Martek made 2 types that I'm aware of one came with just the point replacement unit and all fit under the points cover . The second came with the same unit as the first but also had amplifier (black box) that you mounted externally from the points cover . The box allows the use of coils with lower resistance ( higher load ) . Your 3Ω coils will be too much for points . The points won't last very long . There is still the possibility that you have a wiring problem somewhere . If there is it would most likely be the Yellow or blue to the coil (can't remember witch colour is 1/4) from the points . If you find the Martek is pooched I would be interested in buying the rotor from you . I'm not sure but if you have the amplified version it might be possible to use the black box with standard points to reduce point current like Hondaman does with his ignition amplifier . A  PM to him would confirm this . He also sells a resistor pack that mounts to the coils to reduce their load . This would make them safe for points .

I believe I have the standard version without the amplifier, where does the amplifier wire in, maybe I missed it.

I have checked the continuity along the yellow and blue wires from the coil to the martek wiring, and it seems okay, though I realize that doesnt rule out a fault somewhere. I just checked again and they are good as best I can tell. no jumped or grounded connections.

If it comes to replacing the martek but not the coils, would I want to use an other Martek unit, or a Dyna unit, or some other electronic unit? is there any reason not to do this?


OKAY so here is my next idea. I am going to run it until it gets hot enough to fail, once it fails I am going to repeat the test that TwoTired recommended. is there anything else I should check while the bike is in a state of failure?

[wrenchmuch]

This is the Matek I have. In the second pick you can see the LED. There are two pairs of them. One pair for 1/4 and one for 2/3. The rotor breaks the beam as it spins. Run it till it fails again  then disconnect the trigger wires. Connect trigger wires reversed. Yellow to blue and red to yellow. Remove rotor , turn it 180 degrees and re install the rotor. Start the bike. If the martek is at fault the problem should move to cyl 2/3. If everything works let it run a while in case things had a chance to cool down.

[Spanner 1]

Didn't know it was a one piece unit... duh , so no swapping 1-4 side 

[catsoup]

This is the Matek I have. In the second pick you can see the LED. There are two pairs of them. One pair for 1/4 and one for 2/3. The rotor breaks the beam as it spins. Run it till it fails again  then disconnect the trigger wires. Connect trigger wires reversed. Yellow to blue and red to yellow. Remove rotor , turn it 180 degrees and re install the rotor. Start the bike. If the martek is at fault the problem should move to cyl 2/3. If everything works let it run a while in case things had a chance to cool down.

Okay, few updates.
When checking for shorts, I found that if I wiggled the connections near the brake light switch, the entire engine shut off. upon further investigation, I found that the little three way clip where the black wire from the coil connects to the black wire from the brake switch was faulty. I disassembled the little box connector and found that it was pretty crappy so I soldered it together. Took some photos to share in case someone stumbles across this in the future and wants clarification. http://www.flickr.com/photos/hanback/7544073442/#in/photostream/
Turned out however that while that was a problem, it was not the problem.

I did as you instructed, flipping the rotor, switching the wires, and sure enough, 1/4 fire 2/3 cold.  So, that settles it, the Martek unit is at fault?

Now, what do I replace it with, another Martek? what are my options here? I'd like to change as few elements as possible, and at as little cost as possible.

[crazypj]

I had a Martek 440 fail and rode about 250miles on 2 cylinders.
Switched to Piranha and never had another problem
I get the feeling they only keep working if you use lower rpm and don't go far?

[Hasenkopf]

it would be interesting to hack up an old one, maybe do some reverse engineering, and maybe make one working unit from two bad ones....just sayin....

[catsoup]

it would be interesting to hack up an old one, maybe do some reverse engineering, and maybe make one working unit from two bad ones....just sayin....

I think Hondaman has done some significant disassembly of these units, at least thats the impression I got. He could probably say if this is possible. my guess is no.

I am concerned that if I get another Martek, it will just eventually fail. But I have 3.5 ohm coils, so I can't just get points can I? I would need to get additional resistors, and may as well get the hondaman box. But then at that point it would be the same price as buying a Dyna system... thoughts?

[cb650PK]

If your Martek has no black box, it's one optical units and they are known to fail. I just replaced my friends Martek 440 with Dyna ignition. His failed on 2/3 cylinder.

[wrenchmuch]

Martek's aren't made anymore ( probably knew that). Anything you buy is used and is a gamble. I currently run a Dyna s with Dyna 3 ohm coils. No resistor.  I've read here on the forum somewhere that the coils are too much load for the Dyna triggers. So far so good. I have the Hondaman resistors on my wish list or I might try to make it myself. Points definitely won't like 3 ohm coils. There are a lot of electronic ignition systems for this engine now.