Topic: 3rd Gear Issue has me stumped

[akabek]

I have a 1975 CB550 with about 200KM on it this season without issue.  Over the last week I have been adjusting my carb jetting and in the last two days I noticed if I accelerate hard the bike sometimes experiences sudden hesitate 4-5 times until it red lines.  My chain tension and timing both check out.  The hesitations feel like fuel starvation but when I experienced that last year the bike would stop accelerating at a certain point.  In this case the bike reaches the red line with 4-5 jolts along the way.

Below is some info on my bike setup

Dynoman DP592 SV pistons 10.5:1
Pamco electronic ignition, coils, wires
Ported & decked head, valve job
CB650 cam
MAC 4-1 header with Cone Engineering muffler
Keihin CR26 carbs, stacks and ITG socks

The reason I am adjusting my jetting is the bike accelerated from 0 to red line is strong but while cruising below 5000 RPM the bike bogs if I snap the throttle open.

Any thoughts why the bike is hesitating during hard acceleration?

[HondaMan]

I would suggest trying to lean it out a bit in the zone just before it bogs. For example, if is feels strong at 5000 RMP but bogs temporarily at 6000 RPM, it is running too rich at 5000. The fuel buildup prevents it being able to reach past the mid-throttle zone of the carbs, then. This leaning-out can be achieved with the needles in the slides.

This all presumes you have either the stock airbox in place, or velocity stacks. If you have pod air filters, then THEY are what is causing this problem, because they are disturbing the airflow, making it roil and disrupt the fuel lift unevenly at the mainjet tip. I've seen this many, many times...

[akabek]

I am using the stacks that came with the CR26 carbs.

Tonight I reverted back to a jetting that was a non issue in the past. 

[akabek]

I spoke too soon.  This morning I pulled a plug and it was black.  I will adjust the needle height one or two clips to see if that removes the hesitations.

My current setup uses CR26s, stacks and ITG foam socks.  The bike accelerates great from 0 to red line but bogs if I wack the throttle open while cruising below 5,000 RPM.  My understanding is that this air filter setup does not provide enough vacuum for the carbs to respond.  If I remove the stacks and go to pods the air flow will probably be more disruptive.  What air filtration options do I have to get good throttle response for street riding with CR26 carbs.  I would like to be able to snap the throttle while cruising and get good throttle response.  Would the best option be to build some type of a air filter system using stacks and a box or 1 large filter shared by all the stacks?

[HondaMan]

I spoke too soon.  This morning I pulled a plug and it was black.  I will adjust the needle height one or two clips to see if that removes the hesitations.

My current setup uses CR26s, stacks and ITG foam socks.  The bike accelerates great from 0 to red line but bogs if I wack the throttle open while cruising below 5,000 RPM.  My understanding is that this air filter setup does not provide enough vacuum for the carbs to respond.  If I remove the stacks and go to pods the air flow will probably be more disruptive.  What air filtration options do I have to get good throttle response for street riding with CR26 carbs.  I would like to be able to snap the throttle while cruising and get good throttle response.  Would the best option be to build some type of a air filter system using stacks and a box or 1 large filter shared by all the stacks?

Try running without the foam socks...

[akabek]

Try running without the foam socks...
[/quote]

Running the bike with just the stacks would be the true test but I prefer a filtered setup for street use.  I may never get it perfect but I would like it a little better then what I have now.  Tonight I swapped the stacks for K&N pods.  I also leaned out my jetting but it started to rain so no test ride.  I also have a new set of Denso plugs I should install to test out.  Hopefully my hesitation will be gone as I work through the jetting.

[FunJimmy]

Running the bike with just the stacks would be the true test but I prefer a filtered setup for street use.  I may never get it perfect but I would like it a little better then what I have now.  Tonight I swapped the stacks for K&N pods.  I also leaned out my jetting but it started to rain so no test ride.  I also have a new set of Denso plugs I should install to test out.  Hopefully my hesitation will be gone as I work through the jetting.

I think that suggestion was purely for the purpose of trouble shooting. At the end of the day you're going to find that the Panco ignition isn't providing a long enough duration spark. If I where in your situation I would keep your coils and switch to points with the HondaMan transistorized ignition.

[akabek]

Jimmy, I figured removing the filters were for testing purposes but if the situation improved I still want filters for the intake.

I am still considering the C5 ignition.  I would like to do away with the mechanical advance.  I already modded the springs because they were worn so who knows how consistent it will be over time.   

[pamcopete]

Running the bike with just the stacks would be the true test but I prefer a filtered setup for street use.  I may never get it perfect but I would like it a little better then what I have now.  Tonight I swapped the stacks for K&N pods.  I also leaned out my jetting but it started to rain so no test ride.  I also have a new set of Denso plugs I should install to test out.  Hopefully my hesitation will be gone as I work through the jetting.

I think that suggestion was purely for the purpose of trouble shooting. At the end of the day you're going to find that the Panco ignition isn't providing a long enough duration spark. If I where in your situation I would keep your coils and switch to points with the HondaMan transistorized ignition.

How long is the PAMCO spark duration, how long is the HondaMan spark duration and how long do you thing the spark duration should  be? What equipment and method are you using to measure the spark duration? Is the spark duration dependent on the coil that you use?

[FunJimmy]

How long is the PAMCO spark duration, how long is the HondaMan spark duration and how long do you thing the spark duration should  be? What equipment and method are you using to measure the spark duration? Is the spark duration dependent on the coil that you use?

I'm just the consumer. You're the manufacturer. This information should be forthcoming from manufactures. Please fill in the blanks.

My conclusions are anecdotal plus a bit of forum reading. I've use both the Dyna S and Pamco ignitions and have been chasing down jetting variables ever since. When I switched to a multi-spark ignition my bike miraculously started to behave and carburete is I would expect. Especially with NEW modern carbs. I'm assuming that the multi-spark has a simular effect as long spark duration, but I could be wrong.

Here's a bit of the ending available on the forum.

Several of you have asked me about my "opinions" on the topic of Dyna S and Martek "points replacing" ignitions, and why I don't like them with the stock Honda coils. Here's the analysis, from an engineering standpoint, of what happens when you install a Dyna S or Martek trigger, just replacing the points, and keep the stock Honda coils. The sheet below is the notes and graphs from those analyses (the other pix is the oscilloscope waveform of the Honda coil, showing it's characteristic 1.5mS discharge time). While the image is too small here, if you want to see it better, e-mail me and I'll send you a full-size scan of it. This analysis is the summation that urged me to develop the now-popular, transistorized "Hondaman Ignition" that makes the most of Honda's original design.

The "bottom line" here is: a Dyna S or Martek trigger, if simply installed in place of the original points, will cause HV spark droop, starting at 3000 RPM. The stock points will not let the spark droop until in excess of 12000 RPM. The Honda coils' discharge time is 1.5mS, fixed. If the coil is allowed a full discharge before recharging, the droop will not occur because the magnetic field stays at max flux. If, however, the OFF time of the "points" is too short, the discharge is not complete, which will cause a strong resistance to the current reversal when the trigger "closes" again, and the resulting coil flux is lower (about 40% lower at 6000 RPM) than normal, which makes the HV begin to droop.

I first noticed this issue in my test setups at the beginning of this "Hondaman Ignition" project. The OFF time of the Dyna S is determined by the width of the magnet that is embedded in the advancer's "cam". This is a mere 27% of the equivalent points cam's OFF dwell, in physical degrees (Martek is similar). This causes a very short OFF time for the coil's discharge, especially acute as the RPM rises. This is probably the reason that the Dyna S manual states that you will receive "about the same spark voltage as the stock setup". If measured below 3000 RPM, as these graphs show, this is true. But, the droop at 6500 RPM is almost 40% and it gets worse above that. If the Dyna III "amplifier" module is added, this problem gets better, because that amplifier stretches the OFF time somewhat. Dyna's answer, of course, is to use their coils also, which actually generate about 25% more spark voltage at 8000 RPM than the Honda coils. But, that's a whole system, and that sort of performance should be expected in an expensive "set" like that!

The "Hondaman" Transistorized Ignition simply lets the points switch the coils through transistors, letting some additional electronics absorb the back-emf ("kickback" of the coils)  that would normally arc the points and cause their wear. But, in addition, the Hondaman unit also switches these spikes much faster (600% faster) than the open points and stock condensor, so the coil's discharge voltage remains at the full 100% value all the way to 20,000 RPM (should you be so lucky!), reducing spark droop considerably in everyday driving. This allows for a longer spark at all RPMs than the Dyna S can provide.

It's not so much that I don't LIKE the Dyna S, I actually do like them, in the right application. But, that's not on my bike, anymore. 

[pamcopete]

Hondamans excellent write up compares the Dyna system to his points amplifier system which has a distinct points closed coil on, points open, coil off, called the dwell angle whereas the Dyna system relies on the width of the trigger magnet.

The PAMCO system is more akin to the Hondaman setup because it uses two magnets, one to turn the coil on and the second one to turn the coil off, at which time the coil fires. The PAMCO system also uses an IGBT transistor with a built in clamp just as Hondaman describes in his system. The IGBT transistor is purpose designed and built for ignition coil driving. So, the fundamental difference between the PAMCO and the Hondaman system is the trigger source. One being the stock points and the other an Integrated circuit Hall effect sensor. They both simply turn the coil on and off in response to their respective triggers, so with a given coil, they will have the same spark duration. If you are looking for greater spark duration, then the use of a lower primary resistance coil, such as the "Ultimate" coil that comes with the PAMCO kit will do that for you as well as any other high quality 2.5 to 3 Ohm coil such as the Dyna coil.

In both the case of the Hondaman system and the PAMCO, they just turn the coil on and off. The spark duration would be dependent on type of coil, wire, cap or plug resistance, battery voltage and other external factors. The PAMCO has a fixed dwell angle of 120 degrees wheras the Hondaman system uses the stock points dwell angle of 185 degrees. Whether or not the longer dwell angle would increase the spark duration would depend on the same external factors as well as engine RPM. In our tests we decided to go with the 120 degree dwell for cooler running coils and less total power consumption with the lower resistance coils. There is a significant penalty of higher temperature and greater power consumption when using a 180+ degree dwell angle with a coil with less resistance than the stock 4.5 Ohms. We did not observe any degradation in spark duration with the 120 degree dwell using a lower resistance coil than stock.

[Duke McDukiedook]

Be aware that using a lower resistance coil on a 750 or 550 will bring other problems in the electrical system. Namely, burning out kill switches.

[HondaMan]

Running the bike with just the stacks would be the true test but I prefer a filtered setup for street use.  I may never get it perfect but I would like it a little better then what I have now.  Tonight I swapped the stacks for K&N pods.  I also leaned out my jetting but it started to rain so no test ride.  I also have a new set of Denso plugs I should install to test out.  Hopefully my hesitation will be gone as I work through the jetting.

I think that suggestion was purely for the purpose of trouble shooting. At the end of the day you're going to find that the Panco ignition isn't providing a long enough duration spark. If I where in your situation I would keep your coils and switch to points with the HondaMan transistorized ignition.

How long is the PAMCO spark duration, how long is the HondaMan spark duration and how long do you thing the spark duration should  be? What equipment and method are you using to measure the spark duration? Is the spark duration dependent on the coil that you use?

The Transistor Ignition spark duration is 1.5mS from 400 RPM to 11,000 RPM with OEM coils made for the 1969-1976 "K" bikes (4.6 ohm primaries). With the "F" bikes' coils (4.3 ohm primaries) the duration is 1.45mS. With the Dyna 3-ohm coils made before 2004 (I have a set of those) the duration is 1.1mS unless a 1-ohm Resistor Pack is added, which boosts it to 1.2-ish mS. The Dyna 3-ohm coils after 2004 yield a shorter spark, around 1.15mS with the same 1-ohm Resistor Pack added. The coils from PartsNmore for the early bikes (4.55 ohm primaries) make 1.55mS spark, and their post-1975 coils (4.3 ohm primaries) make 1.4mS spark. These are all measured using an oscilloscope in my test rig: pictures of the setup are in the Build Thread archives when we developed it here in 2006.

Points alone (on the early 4.6 ohm primary coils) generate about 1.5mS at idle speeds, dropping off to about 1.2-1.3mS (depending on the condition of the condensor) at 6000 RPM, which is all the faster I will run an unloaded SOHC4 engine: it's hard on the rods to go more than that with no load on the engine. This causes less combustion at higher RPM, noticed as a power loss.

You can think of it this way: for a given switching interval (aka "dwell") the spark duration shortens as the primary resistance becomes less. With a little more capacitance in the condensor, this will stretch it out slightly (which was why we tested the old Suzy Waterbike units to get all 0.26uF condensors instead of their .22-.26uF "range" of parts). If the open/close points (or switching) time is 50/50 per revolution, the maximum dwell of the coil will occur IF the condensor is tuned to the coil's inductance: if you lengthen the condensor (i.e., increase the capacitance) with RPM then the duration of the spark stays longer and does not shorten at higher speeds. Too much condensor (capacitance) makes for sluggish low-speed operation, though.

Now you know about one of the 'tricks' inside my design...

[pamcopete]

Be aware that using a lower resistance coil on a 750 or 550 will bring other problems in the electrical system. Namely, burning out kill switches.

The 120 degree dwell of the PAMCO reduces the average current for a low resistance coil to about the same as points with the stock coil. You can also wire the coils directly to switched battery and bypass the kill switch because the coils will not draw any current if the PAMCO is switched off via the kill switch. The PAMCO itself only draws about .020 Amps (20 mA).

[FunJimmy]

Hondamans excellent write up compares the Dyna system to his points amplifier system which has a distinct points closed coil on, points open, coil off, called the dwell angle whereas the Dyna system relies on the width of the trigger magnet.

The PAMCO system is more akin to the Hondaman setup because it uses two magnets, one to turn the coil on and the second one to turn the coil off, at which time the coil fires. The PAMCO system also uses an IGBT transistor with a built in clamp just as Hondaman describes in his system. The IGBT transistor is purpose designed and built for ignition coil driving. So, the fundamental difference between the PAMCO and the Hondaman system is the trigger source. One being the stock points and the other an Integrated circuit Hall effect sensor. They both simply turn the coil on and off in response to their respective triggers, so with a given coil, they will have the same spark duration. If you are looking for greater spark duration, then the use of a lower primary resistance coil, such as the "Ultimate" coil that comes with the PAMCO kit will do that for you as well as any other high quality 2.5 to 3 Ohm coil such as the Dyna coil.

In both the case of the Hondaman system and the PAMCO, they just turn the coil on and off. The spark duration would be dependent on type of coil, wire, cap or plug resistance, battery voltage and other external factors. The PAMCO has a fixed dwell angle of 120 degrees wheras the Hondaman system uses the stock points dwell angle of 185 degrees. Whether or not the longer dwell angle would increase the spark duration would depend on the same external factors as well as engine RPM. In our tests we decided to go with the 120 degree dwell for cooler running coils and less total power consumption with the lower resistance coils. There is a significant penalty of higher temperature and greater power consumption when using a 180+ degree dwell angle with a coil with less resistance than the stock 4.5 Ohms. We did not observe any degradation in spark duration with the 120 degree dwell using a lower resistance coil than stock.

Thanks Pete. That's great information.

[akabek]

A couple of days ago I decided to remove my filtered stacks and install K&N pods on my CR26 carbs.  I adjusted my jetting to the following: Main 105, pilot 65, needle YY7 5th position, Air jet 220 and mix screw 1 turn out.  This is probably the strongest acceleration I have experienced to date.  The bike still bogs if I open the throttle while cruising below 5000 RPM but the big problem is the hesitations while accelerating.  What I have found is that this happens only in 3rd gear.  I will have to check if this also occurs when the engine is cold or while slowly accelerating but during hard acceleration it happens.

I am running the Pamco Ultimate ignition
New Denso plugs
I pulled the fuel petcock out and there is no dirt in the filter screen.  The symptoms are similar to fuel starvation but more like jolts.
I have one clear filter on both gas lines.  I think I will remove them to see how that works.

I strapped a GoPro to a helmet and took some video.  The sound has a lot of wind in it but you can hear the hesitation I experience.  If you list to the gear changes you will see if only happens in 3rd gear.

To avoid watching the entire video here is where the hesitations happen
1:48
3:12
5:50
6:32
8:18
8:41
9:39
10:07
11:35

   

[pamcopete]

Spark plug caps, high resistance (>5K) or open.

[akabek]

Spark plug caps, high resistance (>5K) or open.

Pete, I am not sure what you mean by your response.  I have the Ultimate Pamco kit and it came with the recommended caps, wire and coils.  Are you thinking my caps are faulty?  If my caps are faulty wouldn't this occur in other gears?  If you watch the video there are a few cases where I drop down to 1st gear and the skipping happens in 3rd gear.

[pamcopete]

Spark plug caps, high resistance (>5K) or open.

Pete, I am not sure what you mean by your response.  I have the Ultimate Pamco kit and it came with the recommended caps, wire and coils.  Are you thinking my caps are faulty?  If my caps are faulty wouldn't this occur in other gears?  If you watch the video there are a few cases where I drop down to 1st gear and the skipping happens in 3rd gear.

If the caps are new, then they are probably not faulty, but it's possible the wires are not making contact so they are open. Even with open caps, you are getting a much higher voltage than with the stock coils, so the higher voltage would jump the open and you wouldn't notice the difference unless the engine is under heavy load, like pulling in third gear. It's just a suggestion as I experienced something similar in my trusty CB450 and thought it was carb related but turned out to be an open in the right side cap which was not making contact with the wire. It also showed up with intermittent strobe light flicker even though the engine did not exhibit a miss at idle and higher RPM's under a no load condition. The condition will also show up with slightly fouled spark plugs in the offending cylinder because the weaker spark due to the open prevents complete combustion. Just a thought. Doesn't hurt to check.

If you are going to check for this condition, then try to not disturb the wire to cap relationship because the problem may correct itself just as you are checking for it. The open can also be in the wire to coil connection, so just do a resistance check between each of the caps associated with each coil.

[akabek]

If the caps are new, then they are probably not faulty, but it's possible the wires are not making contact so they are open. Even with open caps, you are getting a much higher voltage than with the stock coils, so the higher voltage would jump the open and you wouldn't notice the difference unless the engine is under heavy load, like pulling in third gear. It's just a suggestion as I experienced something similar in my trusty CB450 and thought it was carb related but turned out to be an open in the right side cap which was not making contact with the wire. It also showed up with intermittent strobe light flicker even though the engine did not exhibit a miss at idle and higher RPM's under a no load condition. The condition will also show up with slightly fouled spark plugs in the offending cylinder because the weaker spark due to the open prevents complete combustion. Just a thought. Doesn't hurt to check.

If you are going to check for this condition, then try to not disturb the wire to cap relationship because the problem may correct itself just as you are checking for it. The open can also be in the wire to coil connection, so just do a resistance check between each of the caps associated with each coil.
[/quote]

I installed the Pamco kit a year ago and I may have 3-400 miles on the bike since then.  I will check the resistance between each cap.  I still think if this is the case is would happen in other gears.

[pamcopete]

akabek,

Well, if you were skeptical about the caps you are really going to love this one:

With the problem occurring in third gear, that could means that it occurs with a particular combination of engine speed to road speed and that could mean it is caused by air turbulence at that road speed with that particular throttle opening affecting the air intake or the bowl vent tube location. Some carb installations have a hose to vent the carbs away from the turbulence and the intake to the air box is likewise pointed in a certain direction, usually down, to minimize the effect of turbulence. I know you are not going to like this theory, but it fits the situation in that the carb setup is not stock. 

[akabek]

I was going to remove the wires from the coil to check resistance but they are screwed in very snug.  I don't really want to disrupt them.  If my wires or caps were the problem I would think I would have issues in other gears.  If you watch the video I posted the engine shows hesitation during full acceleration in 3rd gear and medium acceleration around 9:39.  It will check again but I don't think speed or temperature impacts it.

The CR26 carbs and Pamco ignition were installed a year ago.  I did not start experiencing the hesitations until last week.  I did swap out the stacks and pods a few days ago but the hesitation occurs with the stacks or pods.

What are the chances my problem has to do with the clutch or transmission?  Is there way to test it?

[pamcopete]

akabek,


It's not necessary to remove the caps from the wires or the wires from the coils to check for continuity. Just measure the resistance from one cap to the other on each coil. Typical resistance with 5K caps would be about 25K Ohms which includes the 5K + 5K = 10K and about 15K for the coil itself. If it was clutch slippage, the revs would go up at the point of slippage.

A dyno test would provide a lot of information. First of all, there would be no wind, so the theory about air turbulence would be seen there. Secondly, you could see what the torque is when the problem occurs as the theory about the caps would occur at high or maximum torque when a healthy spark is most important.

I had a dyno test done on one of my bikes a couple of months ago and learned a few things about the bike. It cost me $100.

A dyno test would also show the effect of those new carbs and you could try pods or stacks to see which is best. Most of the dyno shops will allow you to make three or four runs with changes to each for the same price.

[Brian G]

FWIW, at 8:41 in the video (and later on as well IIRC), I think I hear the miss during gentle acceleration as well so it may not be entirely engine load dependent. Only in third gear - that's interesting as the engine can be under identical loads in different gears.

I see why you mention clutch or transmission issues - I'm pretty sure I hear the revs increase momentarily during the 8:40 faltering - almost as if the load disappears for a microsecond.

This is an interesting one - not to mention frustrating for you on what appears to be an otherwise immaculate bike! 'Notify' clicked!

Regards,

[akabek]

akabek,

It's not necessary to remove the caps from the wires or the wires from the coils to check for continuity. Just measure the resistance from one cap to the other on each coil. Typical resistance with 5K caps would be about 25K Ohms which includes the 5K + 5K = 10K and about 15K for the coil itself. If it was clutch slippage, the revs would go up at the point of slippage.

I was not going to remove the caps.  I did not want to remove the wires from the coils.  If I can measure from the coil to the end of the cap that would be easy enough the next time I remove the tank. 

I am going to play around a bit more with the jetting.  What I may do is remove the pods and try straight stacks and see how that works.  When I took the video I had to switch to reserve.  It is a long shot but maybe I was experiencing fuel starvation which is what it felt like.  I may empty out my tank and refill it with fresh fuel.