ignition advance really stops advancing at 3300 rpm??

Looking in the FSM (for another thread), I realized that it gave spark advance information:

15° BTDC is base, ramping up at 1800rpm and finishing advancing at 37° at 3300 rpm.. Given that it's only 1/4 to 1/3 of the way to redline, that feels like the timing isn't all that well optimized over most of the RPM range.

I was under the impression that while the burn rate certain changed with rpm, the optimal advance also varied considerably with RPM to keep the peak cylinder pressure point in the 15-20° ATDC (and more or less linearly with RPM)... ??

 

The size of the combustion chamber also matters as the smaller it is the less time to complete the burning so I assume this is the case with small engines.
A dynamometer experiment can verify is there is any gain in advancing more; I assume if there was anything to be gained it would have been no reason for the factory not to do it.

 

Given the sophistication of the electronic ignition control of the time, I'm not surprised that it's simplistic, just surprised that they capped the advance at such a low rpm (vs. redline). Considering that in my riding, it's never below that point (4500 is usually as low as I keep it).

It was something I noticed, and was wondering if anyone on here knew of any reasons why.

 

I had to go back to some of my learning material to get a better description of what you are asking compared to my explanation. I don't explain things right sometimes. so try this. I've highlighted something that you have not taken in to consideration or may not know.

**********************

As an engine speeds up, the spark plug should fire earlier (timing should be advanced) for the simple reason that there is less time for the combustion to happen as the piston speeds increase. The engine tuner's job is to make that spark happen at just the right time throughout the rpm range. For those that want to get technical, it turns out that the spark should occur at a point so that the flame front travels through the mixture and burns it completely, and the heated, expanding combustion products reach a maximum pressure when the piston is about 15-20 degrees after TDC. With a dynamometer at his disposal, that job would be very easy if it weren't for all of the variables mentioned earlier. Fortunately, the effect those variables have is very well understood and they can be accounted for if the engine has the right management system.

Under normal circumstances, pressure has the most significant effect on the ignition timing. When the pressure of the air/fuel mixture increases, the flame front travels through it much faster as the front has less space to jump when traveling from molecule to molecule since the molecules are closer together. The same can be said if more of the mixture is in the cylinder or if the temperature is lower, as cold air is more dense than warm air. An increase in humidity will also act like an increase in pressure as the extra water molecules will help bridge the gaps between the air molecules, increasing the speed the flame front can travel. Since an increase in pressure results in a decreased combustion time, the ignition timing must be retarded as the pressure increases to avoid knocking.

As far as what makes the pressure change - the cylinder pressure increases as the load on the engine increases and/or if forced induction (turbocharger, supercharger) is used. When an engine is idling or under light load and there is little pressure, the ignition timing may be advanced. Once the engine experiences a load or forced induction kicks in and the pressure builds, the timing must be retarded.

Plus there are many factors that play havic on the flame front speed. Such as combustion chamber shape, type and size, bore dia, piston face shape, valve shrouding, plug location, fuel, plug type, intensity of spark, air, moisture and I'm sure more that I'm forgetting. I think you get the idea.

 

Yeah, I can see all of that. I'm wondering if the volumetric efficiency going up with rpm increases the overall cylinder pressures such that the extra advance isn't needed.

but then, our bikes don't take any of those other variables into account (Intake air temp, coolant temperature, manifold pressure, etc.). The more modern systems do a much better job of that. I'm somewhat familiar with the tuning tables in my WRX (I've researched them, but not actually varied them to try to tune, as I like my car being reliable). But the number of adjustment maps, and the considerably changes in timing over RPM (including a very strong continual increase in advance from 5000 rpm to redline, as the turbo's ability to provide boost tapers off) is definitely a change from the more or less static timing that the VFR appears to have over it's actual on-road, at-speed, operating range.

 

That's what basically happens. What you have to keep in mind is that an engine is a dynamic machine. That's why static compression doesn't really mean much compared to dynamic compression that the engine sees. There are many different changes that happens per cycle vs rpm. For example porting heads at 28" of water. The motor never really sees that as a constant or a max. I'm now studying a better way to port heads. There are somethings that have been around for years to making more power that people think are set in stone. Some know that to make more power or the type of power they want that those stones need breaking.

If you really want to learn how to break those stones then get any reading material by a gentleman by the name of David Vizard. He has even taught the guys at McClaren, Cosworth and others a few things. He has another seminar coming up in April. Though it cost a few big bucks to go. I think he has some vids on youtube that may be of help. He's well worth checking in to.

 

And from looking into how to tune a WRX, which can see anywhere from 5psi to 30psi MAP over a huge rpm range... things are VERY dynamic on some engines, but the timing and fueling are just as dynamic as those variables are. Actually a stock WRX ecu is overly aggressive at 5000-5200rpm by a couple degrees, and it's clearly evident when you increase boost (or use Cali fuel at full stock boost). Most tuners delay the onset of the timing ramp by a few hundred rpm to compensate. (as an example).

====

I've got a project that's still a year or two out from starting, but I think will be a lot of fun. I'm going to pick up a 5th gen VFR, and just go to town on it, suspension, engine, tuning, etc. More or less make it a road-legal race-prepped bike by lightening it, adjustable suspension, and then blueprinting the engine while likely upping the displacement and compression, porting the heads, etc. Some of this I'll have a shop do (boring out, for instance), but others like the porting I'd like to do.

The journey being the goal, not the destination.

If I have a dedicated project bike, maybe I'll leave the rest of the cars alone... (and have another bike that's not torn apart for my daily rider).

Also, this isn't a "cost effective" build, but just something that would be fun to do, for the sake of doing it. I just wish I had better access to a machine shop.

 

Tell me what is a "cost effective build".

Hey if your doing all of that. Then why not sonic test your cylinders to see what would be a safe over bore. Remember you have to take in to consideration of bore distortion. Then have some pistons made to where you can take advantage of a bit more stroke and a longer rod for less piston drag. That should keep you very busy just trying to locate pieces parts that you can use. The bad thing is there are not many choices when it comes to bike engines vs car engines. So you would have to get specs for several different honda mc engines to see where you might be able to get rods from. Then find some pistons that are close from Wisco or whom ever and have them adjust the pin location to fit. Hell I haven't even seen specs like that for bikes compared to cars. Seems like the mc factories don't give those out. Seems like the mc world doesn't have a call for after market parts like the automotive side does.

With autos you can build any engine anyway with all the after market stuff available. But then most bike motors have more power from the factory than what their owners can handle to begin with. Plus they are more sophisticated(race bred) than most car engines from the start. So the need isn't there.

Sounds like a fun project good luck with it.

A friend of mine had his WRX souped up a tad. Then it caught fire about a year later. He now has a Mazda Speed3 turbo. He's had some work done on it and it's putting out 300hp to the wheels. That thing feels like my GTO. Neat cars I'd like to roadrace one.

 

CBR919 pistons should be good for about 840cc (bore only). I hadn't really wanted to stroke it much, due to the peak cylinder speeds. If the cylinders can take the extra diameter of the 919 pistons, it's worth it (although since the pistons are 919cc engine pistons, they're used to a bit more stroke, obviously. But more stroke means a new crank, and I don't think I wanted to get that crazy on it. At least since that's not work I can do, vs. just paying to have others.

I want to experiment with the various engine coatings (piston skirts with low friction coatings, ceramic insulation on the crown, hydroscopic coatings on the crank to shed oil easier). Things to make it spin easier, without actually sacrificing low-end torque by lightening the flywheel. At these engine speeds, friction is a big deal, and the world has changed a lot over the last 10-20 years with regard to friction management.

 

Depending on crank journal size vs rod big size you might be able to offset grind the crank to increase the stroke. But that depends on a few things. Might be worth checking in to.

I haven't had much experience myself with coatings. I did ceramic-aluminum coat my first set of headers with Cermakrome from Techline. Worked really nice. Kept the heat in the pipes and the engine compartment was cooler. I've read a bit about piston coatings. Some pro some con. I agree less friction the faster the engine should spin up. The quicker the spin up the quicker the acceleration should be. Though many have not seen any real horse power gains per se. Power is heat and most engines lose about 30% power thru heat radiation. If it can be kept in the combustion chamber it should make more. At least that's the theory. Just never had a chance to play with it.

This site though does have some topics on it. You'd have to search for them. • Speed Talk

You'd have to register to get in to the "Advanced Engine Tech" section. The standard Engine Tech section is very good and may have topics on certain coatings that have/are being used. Great place for info. An engine is an engine, doesn't know if it's a car, motorcycle, truck or what ever. They all work the same as far as the basic principles. It's how you use those principles that makes the difference. Check out the site you'll find a lot of real good info there.

 

Just to clarify: the ignition systems on the early VF's have just those two advances - nothing else. They are literally a step at 3300rpm.

 

Jamie, on the VFR did that become a ramping between 1800 and 3300, or is it a step at each place?

 

I'm not 100% sure abot that, but I think it's a step (or maybe steps) as well. I've seen data captured for the First Gens but not the second. That would be interesting to know for sure.

 

I'm "assuming" it would advance like any other ignition system. Just because I've started out in second gear a few times with no noticeable hesitation or stumble.

I just looked in my FSM for gen1 at the timing section. It states that timing starts at 1300-1750 and ends 3100-3500rpms. There is one way to check it. I have an extra alt cover I got just for that. I haven't used it yet(just got last year). The book shows to take cover, put about 1" hole in it where the timing marks are. That way you can check timing on a running engine. It would be interesting to see if it does advance with rpms.

I'm sure you could change the starting and ending point of the advance by slotting the pulse gen mounting holes. This way you could move the pg's forward or back to advance or retard the initial and total timing. I don't see how you could adjust the curve though. Unless your an electrical wiz and modify the CDI box, IF it uses resistors for a moving timing curve. There would be no need for any of this on a stock machine I would assume.

Jamie what did you see in the data capture? Did you use something like a RacePack type data capture device? If so what else did you capture with it? Just curious.

Woody just get another alt cover, put the hole in it and see what happens. I think that would be the easiest/cheapest way to go.

 

At the moment, I'm not planning on doing any modifications along these lines on this bike (it's in too good of shape to hack into), but greatly enjoy learning about how they work. On a project bike, I'd probably toss the CDI unit eventually and switch over to a motec, just for the fun of being able to play with all the paramters. Also, I'd do that first on an engine I didn't mind blowing up...

 

Actually, to measure timing, it would probably be easier to use a scope on the crank pulse generator and the low voltage side of the coil. At least, if you have access to a scope, that is.

 

A guy on the V4HondaBBS captured the ignition advance, not me. He found there was a clean step at the stated rpm with no ramping up. I've wanted to capture this information myself but have never gotten around to it. Judging by the spark box circuitry I wouldn't expect much more sophistication than a simple step function.

 

I have stripped a spark box of a VF400 and opened another of a VF500.
The circuitry is not simple at all, mostly because both boxes contain a "black box" filled with resin. In the black box there is a hybrid PCB with some surface mount components and some components etched on it.
I had also tried to find out what was going on with the advance but can't recall much as my main goal at the time was to verify if the box was giving full advance.

Methodology is simple as long as you have the right equipment. I was using a USAG timing light which was capable of retarding the strobo light by a potentiometer and display the degrees. I had increased the rpm to 3500 then delayed the strobo light around 37 degrees and there I had the timing mark aligned with the case mark instead of the full advance mark. Taking some pairs of rpm/degrees and there you have the graph of advance.

 

Yeah, I'm thinking I'll just use the oscilloscope here at work to snag graphs at a bunch of different RPMs. I can probably rig up all the input signals needed (cam and pulse generator signals) with equipment I have here in the lab at work, and then can sweep it on the bench. That way I don't have to lug the scope home (and make it smell like exhaust). And it's not exactly an appropriate use of company equipment.... So I really shouldn't (but it's really, really tempting). I need a scope for home use anyway.

This might be useful: USB Oscilloscope - SparkFun Electronics

Certainly qualifies for low-cost. Will be a far cry from mine here at work, though:

MSO7104A Mixed Signal Oscilloscope: 1 GHz, 4 analog plus 16 digital channels [Discontinued] | Agilent

sigh...

 

Nothing better than the proper tool for the proper use. The oscilloscope will not be a straight forward solution but will do the job.
I could probably use a 20GS/s LeCroy WAVEMASTER 8600A that there is around but I would certainly prefer using the humble USAG timing light.