Announcement

Collapse
No announcement yet.

GSX400 Spark Advance Modification

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

    GSX400 Spark Advance Modification

    This thread is really only applicable to the 8 valve 400s that were available outside the US. And I suppose I should apologize in advance - always a dubious move - for not having been more scientific and all that. And no pictures.

    I haven't personally owned a Twin Swirl 400 all that long; about 10k and a year [6000 miles], but the bike is original bone stock so representative of what they were to begin with. I did spend some time years ago with a friend's 'X', but it was so shot that it was hard to tell what it was like originally. My 425 seemed lie a better bet.

    One of the first things that struck me was that the motor seemed a bit rough and over advanced in the 4000 - 4500 rpm range. 5K was smoother and 6k is like an electric bike. I don't recall the exact specs but the 2 valve motors had 35 degrees at 3500 and the specs for the 4 valve are 40 degrees at 4000. 10 degrees seems universal at idle.

    Without getting into a religious discussion about whether high octane gas burns slower than regular - I think they start out about the same but the high octane gets less exponential at the end - it seems rather odd that a fixed 40 degrees would be optimum at 4000 and still manage to be able to burn in time at 8000. You would think that 20 degrees at 4000 would be the equivalent timing. But throttle opening restrictions of the CV carbs and cam overlap and all those gas flow factors mean that the combustion characteristics below the 6000 to 10,000 power band aren't optimal and burning is slower.

    Today's motors have the possibility of any sort of advance you want to program, plus knock sensor overrides. This makes a mechanical advance seem primitive, which it is. As I see it, Suzuki took the same old advance that ran the points and just gave it a touch more throw and had it open at 4000 because that's what the motor could tolerate. Probably the same old springs and weights and such. It worked.

    I experimented a bit with gasolines and found that the motor tended to knock a lot WO in the 4000 - 4500 range unless it had 91 octane. 87 was hopeless, although it would sorta work up high. 89 was fine but still had the 4K knock problem when you wound it open when it was hot. I've run enough engines to know when a motor feels advanced beyond its needs. So I decided to take a spare advance mechanism I had and modify it to give 40 degrees by about 4700. In theory this would give about 30-35 in the bad spot which is more like a two valve, and hopefully make the power band more linear, with maybe a bit less vibration.

    The advance weight weighs 8 grams, but the bulk of that is around the pivot and not contributing much centrifugal force. I seat of the pants estimated that a half gram off the business end would probably be about right so I ground that much off the flat of the end so the stop would still fall in the right place. The weight is 3.5mm thick and it hit target at 3mm, starting at about the halfway point out from the pivot.

    The backup plan was that I could always fall back on the original setup if I didn't like it. The test ride I used has a nice long 'dyno hill' on it where you can whack a 400 wide open in 5th or 6th and do some analysis. From the vantage point of my ears and butt - the bottom line - I think it is hitting full advance at 4800 now and that means it must be low mid thirties at 4000 which actually seems to produce equivalent power but without the harshness. In fact, the whole low range from 3K up seems more linear and of course there wouldn't be any changes anywhere else. Gas mileage hasn't changed, that I can tell, but it shouldn't.

    And the harsh feeling and extra vibration at the 4K point is gone. It's not a big deal, but I can't see why I'd want to go back. Sure, on 91 octane there's no serious ping stock, but if you ever run into a situation where the gas isn't as wonderful as the headline number you'll be better off.
    '82 GS450T

    #2
    This is good stuff and I'm not sure why it would only apply to a 400 outside the US? I'm sure you don't have a motorcycle distributor machine so did you or do you plan to put a timing light to it to confirm the recurve?

    I remember doing some recurving with my old 327 Chevy many years back. Ironically the change from stock Chevy to the modified curve recommended in the hot rod magazine of the day (this was the late 70's) is remarkably close if not identical to the stock setting on the GS1100E.

    IIR correctly the stock 327 has 4deg initial with about 32 degree total accounting for 28 mechanical advance.
    The mod for to limit the mechanical to about 20 degrees and increase the initial to between 11-12 degree. Total advance remained the same but it sure accelerated much faster. I don't recall the stiffness of the springs I used, but I did have a kit with different ones.

    Comment


      #3
      Many years ago a GS tuner I knew said he would play with the weight of the mechanical advance units on the GS1100 to tune for different things , drag , street, race ect. I don't know if he modded them himself or if they were built that way, and I suppose like jetting you would have to have plenty of good info at hand to decide how to set it up for the application.

      I thought at the time it sounded weird, but over the years I now can see how this would be a benefit for the mechanically and technically inclined.

      Comment


        #4
        Originally posted by posplayr View Post
        This is good stuff and I'm not sure why it would only apply to a 400 outside the US? I'm sure you don't have a motorcycle distributor machine so did you or do you plan to put a timing light to it to confirm the recurve?

        I remember doing some recurving with my old 327 Chevy many years back. Ironically the change from stock Chevy to the modified curve recommended in the hot rod magazine of the day (this was the late 70's) is remarkably close if not identical to the stock setting on the GS1100E.

        IIR correctly the stock 327 has 4deg initial with about 32 degree total accounting for 28 mechanical advance.
        The mod for to limit the mechanical to about 20 degrees and increase the initial to between 11-12 degree. Total advance remained the same but it sure accelerated much faster. I don't recall the stiffness of the springs I used, but I did have a kit with different ones.
        While I did a bit of 'adjustment' to my 425 - which is functionally just a 400 - with header and pods and dynojet, I never felt the need to do much to the ignition. Maybe just keeping the points happy was enough of a challenge. It would run on most gas just fine, but I ended up stuck in Cabo during a gas shortage and all I could get for the run back up was Mexican 82 which is closer to what they ran Model As on. Lotsa sulfur. I had to back the timing way off and use half throttle and it still ran hot; the rings were never the same after that.

        The two valve head is a classic hemi and runs about 9:1. The four valve is 10:1 and the whole combustion chamber architecture is different. So the 450 and 500 motors have more in common with the original motors than the GSX 400s. I don't think there's much change to the carbs and pipes, but cams look pretty different, partially because they run out through rockers and probably snap the valves a bit quicker even though the duration is similar - I think.

        So I don't think the curve for the two valve is 'off'; if anything I seem to remember that the detonation surfaced when it hit the power band at six or so. I think I was right in thinking that 10:1 and 40 degrees was a bit optimistic at 4000 rpm.
        '82 GS450T

        Comment


          #5
          I have adjusted springs before, but never ground the weights down... How did you take exactly the same amount off of both weights? I guess looking at how it works maybe it's not all that critical?


          Life is too short to ride an L.

          Comment


            #6
            Originally posted by tkent02 View Post
            I have adjusted springs before, but never ground the weights down... How did you take exactly the same amount off of both weights? I guess looking at how it works maybe it's not all that critical?
            In this case, adjusting the springs to be stiffer would be a good trick. If you wanted to be critical about it, I suppose you should machine an exact amount off the weights, but I just laid them against a grinder and tried to get them as even as I could. As long as you end up with - in this case - about 7.5 grams instead of 8 you should be close enough. If you don't have a scale that accurate, the guideline of about .5mm removed from halfway out until the end is going to be close.

            I just ground a bit and weighed, and then ground some more. The mechanism will average the two weights anyway, so you could just take a whole gram off one and leave the other alone, but that seems pretty crude. OTOH, it wouldn't matter, that I could see.

            I know I should have done a real 'by the numbers' documentation and analysis, but I just wanted to see if the idea had any merit. I don't know if my calculations are exact, but I know that by the seat of the pants I can feel the advance finish at 4800 rpm, which is where I wanted it, and it runs exactly as before from there now. Exactly what the advance is at 4000 is hard to know, but if you take 30 degrees total advance and move it from a 2K to 4K to a 2k to 4.8K sweep it should be about 32 degrees. As I recall, that was the magic number for a Norton Commando motor which had a great sweet spot at 4000 rpm and a much bigger combustion chamber. What I like is that I can be in too high a gear and still pull from 4000 or less and not feel like the motor is pounding on itself so much. It's just more linear down there; no faster, but nicer to be with.
            '82 GS450T

            Comment


              #7
              You could use a timing light with manual advance/retard of the flash.
              ---- Dave
              79 GS850N - Might be a trike soon.
              80 GS850T Single HIF38 S.U. SH775, Tow bar, Pantera II. Gnarly workhorse & daily driver.
              79 XS650SE - Pragmatic Ratter - goes better than a manky old twin should.
              92 XJ900F - Fairly Stock, for now.

              Only a dog knows why a motorcyclist sticks his head out of a car window

              Comment


                #8
                Originally posted by Grimly View Post
                You could use a timing light with manual advance/retard of the flash.
                I'll get around to some measured numbers when I get a timing light to borrow. And a rainy day….it's been too sunny and warm to work on bikes lately.

                It also occurred to me that I could have ground down the tip of the advance weight instead; that would be a lot more controllable. But as it is, the mod is stealth - not that anybody ever looks at advance weights.
                '82 GS450T

                Comment

                Working...
                X