Saturday, November 18, 2017

Sunday, November 12, 2017

My revised muffler baffle for Minnie

This is the baffle I knocked up prior to riding to the FOIM.  I realised I was later than I expected at that point, so it was quick.  Not pretty, but very effective as it happens.  Many years ago Mark Harris who was Madaz at that time told me a baffle needs to fully obstruct the linear flow of the gas to really work, but without that obstruction being overly restrictive as such.  The previous one was a shorter tube much the same size with an open end, like a Staintune baffle would be.  It really didn't make that much difference.

Not sure how this will work on the dyno.  Not sure that I want or care to find out.

An end for Minnie's makeover

At the start of the week I had a "to do" list for Minnie to get her ready for the Festival Of Italian Motorcycles today.

"Front brake work?" at the top literally meant "does the front brake work?  With the second front disc adding a corresponding caliper, the demand for front brake fluid displacement increased by 100%.  The original single disc front master cylinder is 13mm, the original dual disc front master is either 15 or 16mm, depending on model.  My 851 had a 15mm master originally, which I replaced with a 16mm (from my Sport 1100i after I crashed it I think).  I was curious to know how it would work with the 13mm, as I'd never tried it before.  My feeling was it would just have a lot of travel and probably some good feel.  But the reality was you could pull the lever back into the bars with little discernable increase in pressure.  Like it had air in it - I spent ages bleeding it thinking I'd screwed that up somehow.  Whereas on the road it gave a lot of travel before it finally started to stop, without a lot of feel as to when it was going to stop.  So, curiosity answered, larger front master needed.  That was Monday's decision.

The 15mm master is 31% bigger, the 16mm 50% bigger.  I went looking for my 15, but couldn't find it.  I figured the best way to find it was to buy and fit a new 16mm master (it worked).  Because the look for this bike is the "coffin" style  masters, I had to buy a new one as the Sport 1100i 16mm now on the 851 is a remote reservoir style.  The new one turned out to be the new style with the larger fluid reservoir, meaning the lovely Chinese billet reservoir caps I'd bought the week before didn't fit anymore.  Great.  And the pivot pin, which is cad pacified (gold zinc) plated on the originals, is now silver and I've spent a heap getting all the fasteners replated so I had to refit the "not so shiny like the rest of the fasteners" original.  Hmmmmmm.  Did have a nice new lever though.

But, it worked.  Funny how a 100% increase in fluid demand is happily dealt with by a 50% increase in delivery.  The very cool floating cast iron discs don't like sintered pads, so I went through my stash of old single pin original pads and found a couple of pairs that were bead blasted and fitted.  Stops better than it used to with the single disc, but with the organic style pads you just don't get that initial bite that I do enjoy so much.  Maybe some new Ferodo Platinums will help.  More money.

So that was #1 ticked off the list.

The speed sensor for the Acewell was the next issue.   With the caliper adapters there wasn't anywhere to fit a little bracket like I had previously.  I liked the little bracket, as it is a serviceable solution (as in you can remove and refit).  My mounting as below is not - double sided taping it to the bottom of the fork leg.  The sort of thing that customers do that annoys me.  I drilled and tapped a thread into the disc carrier and fitted the little magnet and away we went.  Easy.  Compared to the 120/70-17 circumference of 1860mm, the 130/60-16 measured at 1735mm.  But after riding it around with the iPhone zip tied to the handlebar clamp and the speedo app running, I increased the setting to 1760mm (1.5%) to bring the speedo in on the underside of accurate.  #2.

The tacho drive is due to this being an SS engine, which has a cable drive tacho.  I do have a Monster blanking plug somewhere (at least one in an engine), but laziness had kept me from moving it any closer to this engine.  I found a little blue rubber cap that fitted just fine, and it has resisted bailing for quite some time now.  But, I figured I'd make it a little nicer and I like machining stuff, so turned up a little cap and screwed it on with an old cable collar (I always cut them off old cables just in case).

The rocker covers were looking a bit crappy, but I didn't have any paint close to hand that I thought might be a good fit.  Well, I have some gold that is possibly a good match for the Paso rockers covers that I always liked, but didn't think it'd match the rest of the bike.  I didn't want anything bright, but didn't have any shades of real grey so I gave them a coat of cold gal.  With the rocker covers bead blasted and heated with the heat gun the cold gal dried at it hit them, the finish is rather matt and coarse and lighter than I expected, but it's there and that gets it ticked off the list.

This engine has the D on the timing belt covers and the DUCATI on the alternator and clutch covers, so I scraped, rubbed and polished the paint off.  I thought it was a nice little detail.

I needed a new clutch lever and reservoir cap (to match the new original on the brake master) and the master body had a fair bit of scraping along the road damage near the pivot.  I figured the best solution was a complete new master.  Easy.

I was looking for something to fashion a brake line bracket from to hold the front brake hose at the lower triple.  I usually find old horn mounts are good for this sort of stuff, and I stumbled across a nice gold zinc one.  Just lovely.  A quick bend and on.  Another one down.

The last point on the list - frame bolt caps - had me loking for the rubber caps that go into the tubes for the engine/frame bolts, and it took me to a tub that I thought (possibly correctly) contained mostly parts of the disassembled 400SS.  And in that tub, I found the 15mm front brake master cylinder.  Bugger.

At least I know where it is, so now I can lose it again for the next time I want to use it.

To make the inside of the muffler end cap a little less obvious I tried to clean up the inside with some scotchbrite (which didn't work as hoped polish wise), then ran some masking tape around the inside of the outer and the machined end cap.  I had machined the end cap prior to giving it to Ken when he made the muffler so I could fit a baffle as used in the last muffler.  It didn't make a lot of difference sound wise, so I made another with a smaller and longer internal tube and then folded a piece of sheet metal into a u shape and welded it on.  Suitably low rent, and effective I must say.  I don't really know if it's that loud, but the high outlet certainly gets into my helmet.

And today at the FOIM, after a lot of them had left.  I think it might be finished at this point.  

Now I can pull it apart again.

Tuesday, November 7, 2017

Ducati 2V cam profile comparisons

I made up a fairly basic jig to hold a 2V head and a couple of dial gauges so I could plot out some 2V cam profiles.  I used a spare 400 head I have which is a 3 bearing style head, as opposed to the later 620/695/800 2 bearing head.  

For test samples I had the following, with their Ducati specs:

I borrowed some P (Montjuich, Laguna Seca, Santamonica) and 8J (900ie) from Chris Boucher and Peter Nuss on the TT1/F1 forum, so a big thank you to them.  The 2S 620 cams had been modified to go into the 3 bearing head as used in the previous installment of the Minnie the 400 debacle.  I didn't have any 800 cams (3X and 3U?), and they wouldn't fit into these heads anyway.

The rig was very much analogue.  I did look at some digital dial gauges with outputs, but the cost put me off that pretty quickly.  I did the testing at nights, which involved me sitting down, reading the degree wheel and dial gauges every two degrees of cam rotation and typing that into my spreadsheet. It started out as fun, but after the first couple it got a bit dreary, and by the last two was just downright tedious.  I was very glad to see the end of the job.

Because I was reading the degree wheel every two degrees, I have cam lift values corresponding to every four degrees of crank rotation.  I wasn't going to take readings every degree of the degree wheel - I would have lost my mind.

The result is the following.

Separating the Pantah head cams out (vertical cylinder carb at rear of head), the following has the Pantah (labelled PH in the chart), F1 and P cams.  The F1 cam is the same profile at the R, but ground 8 or so degrees advanced.  I had a set at one time that had OE and MV stamped on them.  To accommodate the higher inlet valve lift at overlap (inlet valve opening) that advance brings, the F1 pistons have comparatively deeper valve reliefs than the later 750 pistons.  From memory they have higher crowns too for more compression.  Although both would be less than that required for the P cams fitted to the Montjuich, Laguna Seca and Santamonica models.

Immediately obvious is the extra duration and valve lift of the P cam.  It's what I would call old school, and what would have been referred to back in the day as "race", or even "full race" if you were going for effect.  Cue the oohs and aahs and air sucked through teeth.  As with most of the old school race stuff, it's kind of crappy now that good ports and cams are well understood.  The greatly increased valve lift at overlap (exhaust closing, inlet opening) is what requires far deeper valve reliefs.

I don't understand what it is about the F1 cam that made them want to use that over the original Pantah cam.  Similar 1mm lift duration and less lift doesn't really seem like an advancement to me.  Bruce Meyers told me the Pantah cams were better in his experience, and now I can see visually that I would expect that to be the case.

Next the "Paso" style engines, which simply means vertical head rotated with inlet at the front.  Although I think it was the prototype Elefant that first displayed the reversed head in 1983.  Anyway, they started out with the R cam, then with the 906 came the HT cam, the HZ in the ST2 in 1997 and the 8J in the 900SSie in 1998.  Also available for them were the Vee Two cams, the 210 grind being the same as the DP 06090 and Gia.Co.Moto GM09 and the 212 grind.

The 2S grind, which was fitted to the new 620 engine of 2002, was the first of the two bearing cam fitments.  This cam was only fitted to the 620.  The 800 also used the two bearing cam with similar duration to the R cam, but with 1.5 - 1.8mm more lift.  This was marked 3X (also 3U?).  It specs like quite a good cam and worked well in the 800, which made 900 ish hp with the 900 size valves, but 15 or so degrees less duration.  This cam was then used in the M695 and M400ie.

First up, R vs 2S.

Dyno comparison from Minnie the 400.  Red is R @ 108 inlet centreline, blue 2S @ 112 inlet centreline and green 2S @ 107 inlet centreline.

R versus HT.  The R cam was also used in the W head M900 engines and the Cagiva Elefants and Gran Canyons.

Dyno comaprison from my M750 with ported, big valve heads.  I later realised that I had taken some ignition advance out of it which did hurt the power at the top end.  But I assume both curves would have been similarly affected.  It was nowhere near as much change as I expected (or had hoped).

This is a 750SSie I did prior to my M750.  Red is R, blue is HT.  I did quite a bit of tuning to it with the HT cams, both fuel and ignition advance.

R vs 8J

Dyno comparison from a Cagiva Gran Canyon, which is a 900ie engine with the W heads - R cams and 41/35mm valves.  We fitted 8J cams to it set to 106 degrees inlet centreline in the hope it would make some more power.  Not enough, as it turned out.  It was still 8 or so down on the 900SSie engine with 43/38mm valve V heads.

The 900 cams - HT, HZ (ST2) and 8J.  The ST2 cam has gained a reputation for being the best, and of working well in a 900 carb engine, but I've never seen any dyno proof of that.

There's really not a great deal of difference between the 8J and HZ.  The 8J inlet peaks a little earlier and longer.  If you compare the area under the curves, again there's little difference.  

If we add the two Vee Two cams to the above, you get a somewhat confusing graph, but the two Vee Two cams stand out for their similar exhaust profiles to the HZ and 8J and their similar inlet profiles to each other.  The 210 definitely works at a lower rpm range than the 212.  Both of these cams don't match their specs, as below:

HT vs 210.  The inlet duration is not much different, and the 210 certainly has a lot more area under both curves.  But the lesser exhaust duration is what hurts it at the top end.

Dyno comparison of a 900SS, fitted with Omrae slip on mufflers and otherwise std.  Well, I believe the after run is the first of the tuning runs.  Red is HT, blue 210.  It certainly picked up the midrange, which is what the long manifold carb motors are all about.

That's it for direct comparisons.  I'll finish by quoting Bruce Meyers, who has a lot of experience playing with these engines.  He posted this in relation to hotting up the 800 engines on a forum some years ago, and I think it's a good piece for reference:

--900SS Carby cams: give a nice increase from 4000 rpm up with avg +4 rwhp over stock cam.

--ST2 cams: Net a bit more on top end rwhp with no loss of midrange.

--900SSie cams: Net alot more on top, maybe 2-4 more rwhp than carby cams but a slight loss 
at 4500 rpm. I think that dip could be tuned out with a good pipe.

--V2 212 cams: Exactly the same as 900SSie cam results

--V2 210 cams: Exactly the same as ST2 cam results

Tuesday, October 10, 2017

Minnie the 400 Monster gets 620 cams: One has one's arse handed to oneself yet again.

Some time ago I bought some 620 cams and sent them out to Ian Drysdale to have them modified to suit the earlier 3 bearing heads so I could try them in other things - namely my 400.  I've also made a rig to hold a spare 400 head I have that allows me to read cam lift as the cam is rotated.  It's all very analogue (ie, me sitting there looking at a degree wheel and dial gauges) and gives me valve lift every 4 degrees of crank rotation.  You'll see more of that in the future.  This way I could compare the 400/600/750 cam (labelled R) and the 620 cam (2S), as below.

Spec wise, the 2S is like the R, but with the inlet closing 15 or so degrees earlier and with more lift on both inlet and exhaust.  In profile, the exhaust is a bit bigger and has an earlier peak while the inlet is as expected.  The reason I wanted to try the 620 cam is that the 400 has a long flat power curve at the top end - it's pretty flat from 9 to 11.  As such, I was wondering if it was over cammed, and figured the 620 cam should be better and a good illustrator of why.  What I was hoping to see on the dyno with the 620 cam was peak power of the same or more, the peak moved down the rpm range and the power dropping away after the peak.

Spec comparison as below:

I fitted the 620 cams at the inlet centreline spec of 112 degrees, and with the drilled pullies I could advance them to 107 to see what happened.  I was also considering retarding them on the dyno as well, but didn't bother in the end.  As usual, the results sucked all the motivation right out of me.

Red is R cam at 108 degrees, blue is 2S cam at 112 degrees and green is 2S cam at 107 degrees.  The midrange improvement is between 5,700 and 7,100 rpm, and the top end loss starts at 8,500 rpm.  Given we're only dealing with 38 hp at the peak, we're not talking great differences here.

As an aside, the runs below are all done with an original 2-2 header and the Megacycle mufflers.  I pulled the 2-1 a few weeks ago and jammed the original horizontal header onto the modified vertical header.  A dyno run showed the same power as where we started all that time ago, and with much the same richness as all the std airbox lid (both snorkels fitted) runs for the 2-1.

Mixture wise, it didn't do much at all.  Still too rich (std 2001 my M400 carbs and jetting).  Top graph is horizontal, second vertical from memory.

To lean it out a bit I thought I'd try to usual trick of pulling a snorkle, which had the usual effect, as below.

Still generally too rich, but better.

As the graphs show, what I wanted to see and what I did see were two different things.  I have become accustomed to dyno disappointment with this bike.

But my mechanical sympathy has me cringing as I rev it over 9,000 rpm looking to use the last of those 38 hp, it's not something I like doing when there's no performance gain to be had.  When I first made the 2-1 and fitted the Acewell dash I was running it up to 8,000 rpm or so and it felt great.  Dyno shows that the 2-1 with Danmoto muffler turns to crap around 7,000 rpm, but under that, and especially through the midrange, it's as good or better than the 2-2.  With the Megacycle muffler it holds the power a bit longer, and definitely felt better on the road between 8 and 9,000 rpm when I held it open just to see how it felt.  But I wouldn't normally go that high riding it around, so I'm a bit caught in the dyno number loop here as it would be.  Or, more to the point (let's be honest here), I didn't expect the things I did to work so badly nor require so many excuses.

Now with the 620 cams it's better again in the midrange, and I think I'll go back to the 2-1 and one snorkel and either tolerate the Danmoto noise or get another muffler made.  I have a design in mind, whether it comes to fruition or not is another matter.

As a comparison to where it was all std, it's not too bad.  Mufflers, cams and one snorkel out of the airbox lid in red, all std in blue.  As a graph, it looks great until you read the numbers on the LH side and think "oh".  Its got another 4 - 5 hp for most of the rev range, which, when you've only got 37 all up, is at least 10%, and more as you go down in rpm.  Woohoo!  Doesn't make it much faster though.  It's slow.  Kinda fun, but still way slow.

Wednesday, September 20, 2017

Minnie the 400 Monster: Not getting anywhere

Man, this thing is just kicking my arse.

What often happens with a project bike of mine is that at some point I go off on a pointless tangent and just dig myself a great big hole.  No point breaking tradition here.

My custom job introduced an obvious issue with the 2 - 1 exhaust, and it has occurred to me that adding the fairing may have played with the fuelling as well.  I've certainly been caught out on that before.

I'd be thinking about what to do about it, and had been getting close to buying another std horizontal header to modify to work with the ceramic coated vertical header and then get a couple of mufflers made to suit.

Thinking about what I had seen with the dyno runs so far, and Dave's comments that i should rejet it to fix the top end richness, I had a look back at some of the runs to see if there was anything to give me hope.  I know from lots of previous experience that top end richness on the Dynojet really kills the power.  Maybe it's richness past the torque peak, as it doesn't seem to affect the power as much through the middle.  A couple of runs caught my attention.

Red is the 2-1 header with the Danmoto muffler and no baffle, blue is the same with one of the snorkels removed to lean it out.  The improvement in top end power with mixture a bit over 13:1 (good for dyno power, but a bit lean on the road imo) was about 4hp.

4hp is about what I'm down from the original headers.  Red is std headers and two Megacycle mufflers, blue is 2-1 and one Megacycle muffler.  An extra 4hp through the top end would be just great, especially with the midrange improvement.

So the obvious was to lean it out.  When I fitted this 400 engine I set these carbs up with the 2001 M400 jetting, which is the same as the M600.  From my video of the slides lifting through the air filter lid, I figured the rich dip was main jet related.  Std is 132.5, and I had 122.5 and 127.5 on hand to choose from.  I was concerned 127.5 wouldn't be enough of a change, so went 122.5.

Easier said than done.  The fairing mounting brackets I made bolt off the bottom of the ignition switch, and all the wiring for the dash, including the original loom sections I needed to hide, are jammed on top of the airbox under the ignition switch.  All of which had to be removed to get the airbox out to access the float bowls.  I'd forgotten about all that guff when I allocated a couple of weekend hours to a main jet change, so it was straight into tanty time from the get go.  Only got myself to blame, on so many fronts.

It's fitted with a heater kit too, so the float bowls are hard to far enough out of the way to get into the mains.

After I had it back together I went for a ride.  Well, first I had to make the dash work again, as disconnecting it made it lose the wheel diameter and the tacho seemed a bit screwy, reading a bit too high.

On the road it felt great, revving out nicely compared to how it had felt before.   Brimming with (over) confidence, I took the chance for an early morning trip to the dyno and typically it went tits up.  On the way there I tried some WOT and it felt pretty crap - flat and surgy - through the midrange.  Turns out it dynod just like it felt.  Crap.

Compared to some previous runs, it gets a bit murky.  The dyno has just had the air/fuel cell replaced, so I don't know if that has influenced anything.  Doesn't explain the difference in how the bike felt though.  Anyway, red is with the leaner 122.5 main jets and one snorkel removed.  Blue is 132.5 main jets and about 5 degrees more ignition advance over std.  Green is the same, but with std ignition advance.  So I really don't know what is going on with it.

The problem is where to go next.  I want to fit some 620 cams to see what happens to the power curve.  I'm hoping that with the early inlet closing, they'll peak the power lower and reduce the need for so many revs.  But, the fact the factory used the 800 cam in the M400ie instead of the 620 cams makes me wonder if that too is just folly.

Before I do that, I want to make sure the result is not tainted by an idiotic exhaust or a wacky fairing.  Plus I have no idea what the original jetting was like with the original headers, which may itself have not been good.  You know where that's going.

Either way, the airbox needs to come out and the 132.5 main jets go back in again which means fighting with the airbox and wiring and other crap and really, I'd have been far better off just leaving the jetting the way it was and fitting the original headers I have.  Joy.

I can always refit the other exhaust when it needs to look good I guess.  I'm even going off that look too.

Friday, August 25, 2017

Minnie the 400 Monster: Muffler comparison test with the 2 -1 header (ie, why is it so bad)

I had the feeling that the Danmoto muffler I had modified for Minnie to go with my 2 -1 header set was the cause of the poor top end.  Getting back to the dyno with a Megacycle muffler fitted proved to be a bit of an organisational nightmare, but I finally got there.

Red is the original headers and two of the Megacycle mufflers.  Blue the 2-1 header with the Danmoto muffler from the previous.  Green is the same as that run, but with one of the Megacycle mufflers instead of the Danmoto.  Bit richer, more power - possibly different wideband sensor to the last runs (wacky?).  Yellow is with Megacycle muffler and one of the airbox lid snorkels removed.  Nowhere near as lean as the runs from last time with a snorkel out, not sure why.  Maybe that was no air filter as well.

All of which means it dyno'd like it felt with the Megacycle muffler - better, and pulled through the top end nicer.  Still down on power though.

An aside here is that when the bike was running the std 2 - 2 header set I didn't have a tacho fitted, and I'm sure I never went over 7,000 rpm.  So that big jump in the std header curve, which happens at around 7,200 rpm, I may never have felt.

I took a std muffler with me for these dyno runs, and some Kokusan ignition units to replace the Ignitech.  I have had intermittant cylinder dropping at idle when hot with this bike for years, across 2 engines now so I know it's not engine related.  I thought I had swapped the Ignitech out without success in the past, but I must say that the issue has not reappeared with these Kokusan units fitted in the couple of times I have ridden Minnie since.  To be fair to the Ignitech, it is the unit I originally bought back in 2006 or so.  It's had a good run, and the new version has a dedicated MAP sensor input which works much better than trying to mimic a TPS with a map sensor.  Maybe I just needed a reason to not feel bad about updating it.

Std muffler in blue, Danmoto in red and Megacycle in yellow.  The air/fuel is a bit unexpected in as much as the red line seems a bit lean, but it may be a physically different sensor (I had two to choose from).  Also you can see that an open muffler leans it out.  I had thought that maybe these carbs would richen up with an open muffler due to the way CV carbs work, but I'm still not convinced this was a good test.  Danmoto is still better than std!  For an illustration of the hollowness of that victory, see the following graph.

As from the previous graph, blue is 2 - 1 header with std muffler, red is 2-1 header with Danmoto muffler and green is way back when with std 2 - 2 headers and std mufflers (all with 108 cam timing).  Yay!  The line for a shot to the pills starts behind me.

Anyway, onward into the slurry.  Red is std muffler as before, blue is Megacycle muffler with Ignitech ignition, green is Megacycle muffler with Kokusan ignition and yellow is Megacycle muffler with Kokusan ignition and one snorkel pulled from the airbox lid.  Helps with the richness.

Comparing headers only (std versus my 2 - 1), first up std mufflers.  

And the Megacycle mufflers.  The point where the two curves cross is bang on 7,000 rpm.

So, some conclusions of sorts.

1/  The Danmoto muffler I really like doesn't work.

2/  The 2-1 header I made doesn't work.

3/  It's too rich at the top end.  From my test of watching the slides lift through the airbox lid I think i'll call it a main jet issue.  If pulling the other snorkel fixed that without hurting the midrange I'd be all over it, but as that wishful plan didn't pan out I'll ignore it for as long as I can't be arsed pulling the airbox out.

4/  Original headers don't work anymore due to the lack of muffler mounts due to the Ducabike rearsets. And the vertical header has been shortened, so any main header will need to be cut, bent and welded as this 2 - 1 header was for them to join.  I did find another original vertical header at work, so I could fit that.

5/  One has dug oneself a hole.

I think I'll need to refit some 2 - 2 headers, as the next step is some 620 cams modified to fit the old 3 bearings heads.