Exhaust Anti-Reversion Cones - Tried Them?

[geodoc]

My Green Guzzi project bike got a dyno run a few months ago (jetting & ignition advance curve tweaks). The torque curve from that confirmed what I'd felt by the seat o' the pants - torque dip from 3500 to 5500RPM. The bike has a home brewed 2 into 1 exhaust:

This winter I decided to try adding a cross-over pipe, just below the alternator cover. This seemed to help (seat of the pants again). A trip to the dyno will tell the tale. thinking to about further exhaust massage for a (hopeful) even flatter curve by employing "anti-reversion" collars in the exhaust. These are installed in the exhaust pipes right next to the heads:





I made these with the advice of someone who has used a similar design on a Triumph Trident with noticeable effect (much better throttle response a the low end of the power band). The ID of these ones is the same as the exhaust port at the upstream end and diverges slightly towards the outlet, or downstream end.

Here's an interesting discussion about them on a XS-650 forum:

http://xs650temp.proboards.com/index.cg ... hread=8722

I'm wondering is anyone on the board has fooled with anti-reversion devices before and if so - what effect did the have? - what dimensions did you use relative to exhaust port and pipe ID? - what sort of exhaust system did you use them on?

Thanks,

GD

[Tonerjockey]

Hey GD

Dunno bout no anti-cones but the bike looks great!

Beautiful paint!!!

Alex

[Chris R]

FWIW, I have tried anti-reversion cones and valves as well - rimflow intake and exhaust, same principle - and in both cases they did help. The improvement was measureable. The reversion cones required some fiddling in regard to length. Where the cones are in relation to the exhaust valve seems to be the critical thing. I tried different steps - that did not make too much difference. I did not have the time or money to evaluate them one component at a time so don't know where the gains came from. Exhaust tuning is time consuming and can be pricey - unfortunately there is power in that pipe.

Chris R

[geodoc]

Heh Chris,

Yeah, I hear you on the exhaust tuning project. In my case I'm not really shooting for exhaust heaven, rather just avoiding exhaust hell. Plan to just keep the existing configuration inspired by the Manel Segarra rig below and tweak with - 1st: cross-over pipe, 2nd: anti reversion cones, 3rd: "ditching" the exhaust (and maybe intake) valves and ports.

All this to be confirmed with dyno runs to see the effect of each change.

Interesting that you said the length of the cones didn't make too much difference. I made them purposely a bit long so I could shorten and check for any difference.

I have noticed that Manel's bike in later iterations (like what he was riding when he won the endurance event at SPA last summer) has lost the 2 into 1 and he's using a 2 into 1 into 2 set-up with reverse cones.

GD

FWIW, I have tried anti-reversion cones and valves as well - rimflow intake and exhaust, same principle - and in both cases they did help. The improvement was measureable. The reversion cones required some fiddling in regard to length. Where the cones are in relation to the exhaust valve seems to be the critical thing. I tried different steps - that did not make too much difference. I did not have the time or money to evaluate them one component at a time so don't know where the gains came from. Exhaust tuning is time consuming and can be pricey - unfortunately there is power in that pipe.

Chris R

[GuzziMoto]

First, nice bike.
I have some experience with them on Buells. They are popular with the Harley crowd. Part of that is because they tend to use primary pipes that are too large (they like the look) and the inserts help them make more power. Search the HD/Buell performance sites, I imagine they are still being used.
For max power I prefer a 2 into 1 set up with a merge collector. Length and diameter of the header pipes is where the key is, but a merge collector helps you get the most out of what you have.
My favorite set up was for a 750 Ducati and was made by a guy named Herb Gebler. He was/is a cool guy. He made us a beautiful set with stepped diameter primaries and two different merge collectors with different lengths for different tracks.
But that was 15 years ago, merge collectors were new then.

[guzzigray]

Hi George. For what it's worth I've done some research on pipe tuning for my racer. My limited understanding is this: primary i.d should be no bigger than the port otherwise you lose velocity. In the case of my 1100 that's 38mm. The stock Sport 1100 headers I use have a 38mm primary for approx 2" out of the port and then step up to 41mm i.d. so in effect a built in anti reversion primary. Seems to work as I get good power throughout the rev range and no power dip (low 80's rwhp). To answer your question no I haven't tried cones but the factory set up I have seems to back up your theory.

Also it seems 2 into 1 is good for power but is peaky while a 2 into 2 has a good spread of power and that large bore primaries are good for top end, smaller ones for torque. Guzzi 45mm O.D pipes seems to have the best of both worlds.

http://bmwmotorcycletech.info/InExTuning.htm

http://www.rbracing-rsr.com/lsr21.htm

[geodoc]

Heh Andrew,

Well it seems like Guzzi may have been a step ahead on their exhaust tuning out of the box on the Sport 1100. I'm running 850 LM "mid-valves" and the exh port diameter is 36mm at the outlet (the heads / ports were "stage II modified by Mike Rich - material removed, but not added which is his stage III). Header ID is 38mm. The cones in the photos are about an inch long - not the 2 inch like you said the Sport 1100 has, but a point to start at. The cones are pretty easy to make.

Hope to get this to the dyno in the next few weeks and test out the exhaust with cross-over compared to none as with the last (carb & ignition tuning) runs. Then install the cones and have a look at that.

BTW (spam-ish plug)

I used Sport 1100 NUSD forks on this unit and YSS RZ 366 TRL shocks when I built it. Despite a whole lot of mucking around with different rate springs, various weight fork oil and the range of damping adjustments on the forks, the ride was consistently awful - jarring and horribly uncomfortable. This was true for the shocks as well. I sent the shocks back to YSS (via Todd) for adjustment, but saw little effect there too - jarring and uncomfortable like the forks.

I have a friend in Asheville NC with a machine shop and Guzzi owner who has done some work for a fellow there named Rick Tennenbaum, owner of Cogent Dynamics, a motorcycle suspension outfit. I called Rick up and was impressed with the depth of knowledge and amount of pondering he has done regarding bike suspension, so sent him the shocks and forks for modification with hope of improving things.

http://motocd.com/

I got the forks and shocks back a few days ago which luckily coincided with 2 days of no rain (shocking in Vancouver Winter!). The difference was utterly amazing - the bike is transformed. I mean the gates of bike suspension heaven opened up and a chorus of compression, rebound and spring rate angles sang their lungs out. You could almost hear the forks and shocks laughing in contempt at the pitiful attempt from those downhill washboard bumps coming up to the stop sign around the corner from the shop to jar my eyeball out of their sockets like before. But ............ perhaps I exaggerate, though not by much.

He installed Cogent Dynamics damper units to replace the stock items and came up with a means to allow damping adjustment using the stock adjuster barrels (still compression on one side and rebound on the other, though both damper units are active in both directions unlike the stock units). Also changed springs to .90 kg./mm. Shocks were modified to decrease rebound setting and different fluid, re-sealed and re-charged. Sag was spot-on right out of the box. The price came to $1050 including shipping back to me, which seems pretty reasonable.

Hi George. For what it's worth I've done some research on pipe tuning for my racer. My limited understanding is this: primary i.d should be no bigger than the port otherwise you lose velocity. In the case of my 1100 that's 38mm. The stock Sport 1100 headers I use have a 38mm primary for approx 2" out of the port and then step up to 41mm i.d. so in effect a built in anti reversion primary. Seems to work as I get good power throughout the rev range and no power dip (low 80's rwhp). To answer your question no I haven't tried cones but the factory set up I have seems to back up your theory.

Also it seems 2 into 1 is good for power but is peaky while a 2 into 2 has a good spread of power and that large bore primaries are good for top end, smaller ones for torque. Guzzi 45mm O.D pipes seems to have the best of both worlds.

http://bmwmotorcycletech.info/InExTuning.htm

http://www.rbracing-rsr.com/lsr21.htm

[guzzigray]

Hi George. Good news on the suspension and the price seems reasonable.

Couple of corrections on my prior post: the stepped primary is more like 1" before it steps up. I.d of the header is 42mm not 41mm.

I had a torque dip in the same range as you and it was down to lean jetting. What are you running??

[geodoc]

pj 57
nj 266 ab
needle k5 lowest / richest notch
mj 150
slides 60/3

The dyno that it got set up on had no gas sensor, we just ran it and adjusted for best power and throttle response

Hi George. Good news on the suspension and the price seems reasonable.

Couple of corrections on my prior post: the stepped primary is more like 1" before it steps up. I.d of the header is 42mm not 41mm.

I had a torque dip in the same range as you and it was down to lean jetting. What are you running??

[Chris R]

Interesting that you said the length of the cones didn't make too much difference. I made them purposely a bit long so I could shorten and check for any difference.

What I meant was where the end of the tube is in relation to the exhaust seat seemed to be the important dimension. The space from the end of the cone to where you bolt it on did not seem to enter into. Also the amount of step didn't seem too critical to me but then I didn't have the time to really try much out. You can do some evaluation with the flow bench (or just a vacuum cleaner and manometer) by running it in the reverse sense and seeing what stops flow. My understanding of the way it should work is that the step stops flow at the sub-sonic level but has as little effect at the sonic level as possible (or so you hope).

That being said, I think the anti-reversion thing is a bit of a black art. I know that Grumpy Jenkins had some success with them- take a look at his Chevrolet Racing engine book.

You might make some spacers that fit over the od of the cones you made and play with the volume in that gap just to see.

Also it seems 2 into 1 is good for power but is peaky while a 2 into 2 has a good spread of power and that large bore primaries are good for top end, smaller ones for torque.

That is pretty much it - you really have to tune to the cam you chose. You can optimize for a specific RPM and that is about it and 2 into 1s will be peaker. Unfortunately a lot of cutting and trying is about all you can do. A four gas meter is definitely useful.

As usual I admire your talent as a craftsman George - very beautiful.