Fiat X19 SOHC Bike Carb vs DCNF Build and comparison

[James Bowen]

Hello,

It has been a while since I have posted on this excellent forum, but I have been waiting to be able to post some results.
Guy, you will probably remember the thread that we started on “SOHC carburetion and setting up.”

Here is the link: viewtopic.php?f=13&t=647&hilit=sohc+carburation

That rather long thread followed the tuning and developing of the engine set up I had at that time. I.e. Changing exhaust, and the Weber 40 DCNF carburettor jetting. To save time for other interested members and viewers, I have summarised what the set up was, and what we did:

1.) Standard 1498cc capacity

2.) Lightened flywheel, balanced together with clutch, pulley, crank, and the rod's balanced end to end.

3.) Cylinder head has standard valve sizes. 3 angle seat work, and ported/flowed by myself, after some study of “Vizard’s” books, and Guys wise words on this site.

4.) Compression ratio is 10.8:1, but standard flat topped pistons. (I'll admit now that I've skimmed the head to get this. In my defence, at the time, I did not know about high area squish bands, and other associated anomalies that are now detailed in Guy's Cylinder Head prep. series)

5.) Cat Cams steel billet camshaft Lift 10.75 mm Clearance (cl)=0
Duration 284 deg (0.1mm+cl)
252 deg (1.0mm+cl)
Peak angle (lobe centreline) 106 deg
Lift at TDC 3.50mm

6.) GC valve springs

7.) Goetze head gasket

8.) 4-1 manifold 1 3/8" primaries 23" long, 2" secondary. Single stainless steel box straight through type Jetex silencer.

The engine put out at the flywheel, 126 BHP at 7,600rpm (Though wheel HP was dropping all the time from its peak at 7,000 rpm), with Peak torque being 104 lbft at 5,600 rpm

Power plot from January 2006. Originaly set up by Track n Road after fitting 3D mapable KMS Ignition, and wasted spark ignition.

Map 2006 Track n Road.jpg (120.27 KiB) Viewed 20760 times

After looking at the Power plot, Guy suggested that the torque was a little low, and that the jetting of the carbs was suspect due to some small 130 air correctors being fitted. Guy suggested that either:

1.) The head was not flowing enough air.

2.) There was an exhaust black pressure problem, or adverse pressure waves.

3.) Camshaft not dialled in correctly.

Guy suggested a “step by step” tuning approach, which I endeavoured to follow, and can summarise here. Not wishing to remove the head just yet, I accurately this time, set the cam at split overlap. It had been 5 crank degrees advanced. I found that made no real difference though power seemed to come up earlier than before. I then swapped the silencer box to one of proven flow. I took the car to the rolling road, Peak torque now appeared to be more, and torque came in at 3,800 rpm rather than 4,000rpm, but then power dropped off much more quickly after 6,750rpm rather than holding on till 7,600 rpm as before. The fuelling issue was the same as before.

Power plot after altering the cam timing retarding it 5 cam degrees back to 'Straight up' Also a proven George Polley silencer was fitted

Map 25th Jan 2007 Southern carb.jpg (76.09 KiB) Viewed 20760 times

I’ll admit was pretty vexed. Measured power was down, though I don’t think it felt any slower. Guy suggested that we needed to look at the header length next. The design was 4 – 1 with primary lengths of 23”. I changed these to 30”. Though this made virtually no difference to top end power, it did significantly alter the low end performance for the better. The torque came up sooner, and the part throttle performance at low and mid range was much better. Fuelling.....Same.

I then popped in 34mm chokes as opposed to the 32’s. I had the car rolling roaded, and got pretty much the same results as before but with the engine holding on slightly longer up the rev range. Torque appeared to have dropped from the “Southern Roads” dyno session, but was more consistent or smoothed. I cannot be sure how accurate the figures are any way due to the different dyno’s used. But the RPM where peak power and torque was delivered remained constant, as did the fuelling issue.

Power plot after changing the primary header length to 30" and a change to 34mm chokes

Map Feb 2008 DCNF 34mm chokes Austec Racing.jpg (49.04 KiB) Viewed 20760 times

Peak power still 126 bhp though now at 7,225 rpm, and peak torque now 102 lbft still at 5,600 rpm


Next stage was maybe to look at the head flows, however before this, I wanted to be sure that the exhaust was definitely not the cause of the suspect jetting. Happily, just as I was planning this, Guy had received, and very kindly posted the results Dr K. Mullens analysis of a suitable exhaust for Steve Sharpe’s X19 inj.

The link to the exhaust build thread is here: viewtopic.php?f=13&t=1482&hilit=sohc+setting

The result of this in terms of a rolling road result that I can post, I am afraid to say was not measured. Subjectively, the mid range torque felt hugely improved. Top end was still the same as before. The engine didn’t really want to go much beyond 7,200rpm. The low end part throttle operation though, was much better. The hole that had been in the torque at 1,500rpm disappeared and made the car feel like standard in terms of tractability.

If you go back to the very start original “SOHC Setting up” thread, it originally was a question to Guy about using a sidedraft carburettor set up, and more specifically a set of motorcycle carburettors to provide fuelling. This post covers exactly that question. But before embarking on that road, in May 2009, I started to rebuild my engine.

I say rebuilt, however, when I put together my first engine, I compromised on some things that I now know I should have dealt with. One of those was a slight imperfection in the bore of one of the cylinders. Due to this, I got hold of a complete late model 14 bolt engine second hand. I re-used the conrods, crank, and flywheel assembly, as the flywheel had been lightened, and everything balanced previously. This time I bought some GC designed and supplied (originally, though second-hand to me) CP forged high compression pistons.

GC designed 'CP' high compression piston

Pistons.jpg (108.47 KiB) Viewed 20760 times

My original cylinder head, though ‘ported and 3 angle seats fitted’ (27mm by 25mm port runners) was put to one side due to it having been skimmed down to raise the compression ratio (I know what Guy thinks about that!) The new head was fitted with bronze guides, and had 39.5 mm Fiat Tipo inlet valves fitted. The valve throats were made 33.5mm, (85% of the valve diameter) and the port runners 30 – 31 mm. The valves had a 30 deg back grind. The standard 33mm exhausts valves were retained. 3 angle valve seats were cut 30/45/70. I followed Guys threads on deshrouding the combustion chambers and worked mine to a similar shape. I retained the same camshaft as before, and timed it in straight up at 106 deg. The engine was assembled together with the same Stainless Steel 4-2-1 exhaust and the same DCNF carburettors using the same jetting as before. The engine build went fairly smoothly, and owes a lot to the high number of downloads and print outs from Guy’s “How to ......” series in the GC Virtual workshop. I honed the bores with a Flexhone. The engine rings bedded in, in about 1.5 hours running time. Compression tests showed 218, 214, 216, 220 psi. The engine was run in carefully as per instructions then, we went to the Rolling Road again.

Below is the resultant plot, along with the “Pre-engine build and new exhaust plot” for comparison purposes.

Power plot after engine rebuild inc 39.5mm inlet valves and forged pistons. Everything else the same as before.

Map July 2009 DCNF's Austec Racing.jpg (50.75 KiB) Viewed 20760 times

Comparison of 36mm inlet valves vs new 39.5mm inlets Compression ratio should be about the same as is everything else

DCNF 36mm Valves vs 39.5mm Valves.jpg (117.72 KiB) Viewed 20760 times

I was a bit disappointed – again! Peak power was ‘only’ up to 130bhp. Though peak torque was a lot better at 110 lbft, it still wasn’t great. And to cap it all, the fuelling issue was still present, though to a much lesser extent. I ended up with 135 main jets and 180 air correctors. The plot you see here still has the 170 airs in. Again we found that the air correctors had almost as big effect on the peak torque at 5200rpm as at the top end. As I never got the AFR as I wanted it, I didn’t go thrashing about on my new motor.

OK, now I had had enough of developing the DCNF’s. I had tried most things, step by step, and spent time and money. I had still been hankering for the last 3 years after the sidedraft idea. I reckoned that they would release instant extra power and torque, maybe provide more consistent fuelling as well?

Why bike carbs? Well, I saw an article in a mag some time ago, extolling the virtues of the idea:

1.)Cheap second-hand.

2.)The CV (constant velocity) design with the slides providing a choke size relevant to the engine demand for better low end drivability.

3.) With a set of 40mm carbs, once they are open, there is no choke or other obstructions apart from the butterfly.

Well I can squash the idea that they are cheap! Not any cheaper to buy, overhaul, and set up on a manifold and tune, than a set of Weber DCOE’s or Dellorto DHLA’s, as you will see.

First off I needed some carburettors. I chose some Kawasaki ZX9R Keilin 40mm carbs from a 2001 machine. They were clean, and from a low mileage bike.

Kawasaki ZX9R E1 Carbs spaced to suit the SOHC inlet ports

Carbs 1.JPG (133.99 KiB) Viewed 20760 times

Next I needed a manifold. My idea was to have the bike carbs spaced so as to provide equal length and shape manifold runners. I designed the manifold to have as straight a shot into the inlet ports as possible, and carried on the inlet port angle in the head i.e. 9 degrees from perpendicular too the manifold face. This gave me a problem with fitting everything in between the engine and the rear bulkhead / cross member. The manifold could not be longer than 85mm overall.

Straight shot into the back of the valve

Manifold 1.JPG (60.72 KiB) Viewed 20760 times

The gap between the water pump and new inlet manifold, for the exhaust primary on no.1 cylinder is very small. The gap is even smaller when the bike carbs are attached. I used an old cylinder head to bolt everything too, and ensure that the manifolds would fit together. The manifold was made up by a company called “ALTISS ENGINEERING” and I can recommend them with no hesitation at all.

Equal length inlet manifold runners

Manifold 2.JPG (111.03 KiB) Viewed 20760 times

Not much space for the No.1 exhaust primary. Heat build up is a big issue. Bolting up the manifolds was an exercise in patience as well!

Manifold 3.JPG (102.73 KiB) Viewed 20760 times

Next job was to attach the bike carbs which I had spaced to suit the manifold, and sort out a throttle linkage. I could have just attached an extension cable to the existing throttle cable, and plugged that straight to the carbs throttle wheel. But, motorbikes have a small throttle spindle / wheel, because the twist grip on the handlebars has to be able to fully actuate the entire throttle range, without the rider re-adjusting his grip. That would have meant a very short throw on the accelerator pedal, and I would have to be super sensitive on the accelerator for tickling around town. Instead I devised a way of using the existing linkage on top of the cam cover. I then used a lever and link rod arrangement to turn a pulley wheel, and attached the throttle cables through a 90 degree turn down to the throttle wheel on the carbs.

Linkage looks a bit elaborate, but works progressively and well.

Linkage 1.JPG (128.49 KiB) Viewed 20760 times

Shaft runs in a bearing held in the black home made bracket.

Linkage 3.JPG (108.12 KiB) Viewed 20760 times

Linkage both opens, and shuts the butterflies for extra control, and safety

Linkage 2.JPG (83.85 KiB) Viewed 20760 times

I worked out the radius of the various levers, together with the length of arc the pulley needed to turn, to operate the throttle wheel through its entire range. The result is an accelerator pedal that is able to be fully depressed in order to fully operate the throttle butterflies. Incidentally, arranging the small lever so that it starts almost horizontal ensures that the first part of the throttle operation is slow moving. The link rod is moving more vertically than horizontally. This alters as the lever moves up to, and over the top of its arc, making the throttle act more quickly as full throttle is reached.

The next job was ensuring clean air. There is no way I will run any engine of mine without filtration. However, this proved to be, and is still an issue with fitting them in between the cross member and the carbs. I initially designed and built an airbox to go in the gap, but in the end I couldn’t get it to fit. I decided to go with some K and N air filters. The depth of them is 64mm, which means there is a gap of 39mm between the top plate and the end of the rampipes. Due to this I cannot try different rampipe lengths, so I may be missing out on lots of potential torque from length tuning.

Not used, but pic shows the rampipes

Airbox 1.JPG (111.06 KiB) Viewed 20760 times

I was quite proud of my home made effort. Didn't fit though....

Airbox 2.JPG (99.45 KiB) Viewed 20760 times

Another issue is that the air in that area is hot from being over the exhaust, and near the silencer. I have ducted air from one on the side ducts as a temporary measure, which actually seems to work quite well. I will need to sort a proper solution though. I have tried to protect the maifold, and the carb float chambers from heat build up, with exhaust wrap (minimum I can get away with) and heat reflective tape, and an aluminium heat shield between the silencer / exhaust pipes and bottom of the carbs.

Heat shield tape on the float chambers to reduce temperatures

Carbs 2.JPG (111.64 KiB) Viewed 20760 times

OK, well I have had the car on a rolling road, and below is the result. I went to a company that primarily deals with motorbikes. They also have a 2 wheel car chassis dyno for bike engine cars, and efforts such as mine. Unfortunately, the software does not convert to flywheel power and torque, so the figures you have below are at the wheel. Now I know that different rolling roads produce different wheel figures due to the different roller set ups, and of course different run conditions on the day. The figures put up are 105 lbft at 5,000rpm and 121 bhp at 7,000rpm. That’s about 9 lbft and 16-18 bhp more than recorded before at the wheels. How does that relate to previous dyno runs? Well I don’t know for sure until I take the car for a run at a different dyno that can work out flywheel figures.

Wheel Power 121 hp and 105 lbft of torque.

Map Jan 2010 Bike Carbs PDQ.jpg (67.16 KiB) Viewed 20760 times

The fuelling appears more consistent, but not brilliant, and is a bit lean 5,000 to 7,200rpm. The carbs are using the stock pilot jets, but mixture screws are ¾ turn out, as opposed to standard 2 turns. The needles have been dropped to lean the mid range. The main jets have been increased from 160 and 165 to 170 across the board.

For interest I have re-adjusted the DCNF wheel power figures up, to make the drag losses the same as I think this type of rolling road has for my car (It uses the same big single roller that the first rolling road used – ‘Track n road’ in Essex) If nothing else it serves to compare the curves even if the figures are too much or too little. Note how there is a gain EVERYWHERE!

Gain is everywhere along the torque curve. The curve also bears a very close likeness to Steve Sharpe's X19 Inj. prepped by GC and mapped by by Dr KM.

Bike carbs vs DCNF.jpg (151.57 KiB) Viewed 20760 times

X19 1500 atmo 07.148.pdf

Steve Sharpe's X19 Inj. Power plot

(12 KiB) Downloaded 575 times

DCNF

Engine.jpg (150.79 KiB) Viewed 20760 times

Bike Carbs

Engine 2.JPG (121.26 KiB) Viewed 20760 times

The car certainly feels worth the extra power. It goes like stink! Best of all though is the low end tractability. It runs about town like a standard motor, with no hint of bad behaviour. Starting is easy with a choke mechanism and the ECU controlled ignition that is already installed.

Are bike carbs better than DCOE’s? I do not know. I would not be so bold as to make that claim, for one I have no experience of the latter. Also, as you have read, there are many snag factors to this set up, and the DCOE’s have the benefit of being tried and tested. Also there is an off the shelf manifold, which appears to be proved to work.

For further development (Really?) I wonder about the camshaft. Power drops off after 7,000rpm. The engine is built to take 8,000rpm. Will I gain more than I lose by moving the torque up the rev range a bit? Will advancing the cam help? That usually loses power at the top end replacing it in the mid range?

Regards, James Bowen

[Guy Croft]

MODEL POST!

Well done James, some very interesting stuff there! Can't say I like 'inducting at underhood temperature' though!

No idea about the 'whys and wherefores' of bike carbs, sorry: that whole subject leaves me cold.

You need a GC head yuh know!

G

[James Bowen]

Thanks Guy. Your site is one of the only places on the web. that people can get actual data and results, rather than "My mates's car made 2,000hp with a new cam" I hope I've added a bit to that.

When comparing the dyno plot from Steve Sharpes car to my own, it would make me believe that exhaust design is an overiding factor in determining the shape of the torque curve. (?) Our different induction systems are chalk and cheese.

"Inducting at underhood temperature" will not be a feature for long. Certainaly there is already a cold air feed to that area, and a better one will be in place before track use.

As for Bike carbs. I'm not going to be an advocate for them against other forms of fuelling. A quick glance at the AFR maps of fuel injected cars versus carburettors (in whatever form) shows that fuel injection is the way to go, for an ultimate set up. Though as you yourself say on this site, carbs are still a proven way to good horsepower.

What I would say for bike carbs is that these particular ones are from an engine that is only 900cc in capacity, yet produces 143 BHP at 11,000rpm. (a staggering 159bhp per litre) The camshafts are 316 deg and 300 deg in duration. YET. The bike is a standard roadgoing production model. It has to pass emissions tests, etc. and pull from low rpm without any histrionics.

From what I have so far experienced with these carbs fitted, is that these low speed running characteristics, have been passed to my engine. Remember I have changed NOTHING else except the carbs and manifold. I have had my foot flat to the floor at 1,500 rpm in 3rd gear, and the engine has pulled cleanly, and with no hesitation or missing all the way through to the red line 7,300 rpm. Today, I was going up a hill in a residential street. 2nd gear, engine slowing to 1,300 rpm for each speed bump, and pulling cleanly away from each bump. (As good as my bog standard Ford Mondeo) This is seamingly without a comensurate trade off or penalty at high rpms. Beyond my own experiences with these carbs I cannot of course comment.

Head flow? Hmmm, I am sure it won't be as good as yours. I reckon I've got 96 bhp per litre at the moment. Torque per litre on par with the CVH posted in GC VW recently. Camshaft not as radical as it could be yet....I will let you know in due course.

Regards, James

[Rich Ellingham]

Truely excellent reading James, well done on all the trials that came your way and I thought I had a few issues.

Rich

[Guy Croft]

"Torque per litre on par with the CVH posted in GC VW recently"?

I have no wish to be picky but if I read your dyno right you've got around 103/105lbf ft on a 1500, the 1600CVH had 123/126 lbf ft (with/without filter).

There's a finite limit to 8V torque, I use this limit - about 126lbf ft on a 1600cc - as my 'aide memoire', so:

(1500/1600) x 126 = 118lbf ft.

That's what you've have if yours was on a par with the CVH. Am I missing something (I often do..!)?

G

[James Bowen]

Guy,

Only that the power and torque posted on mine is at the wheels. Though I have yet to get a dyno read out for the Flywheel figures, I have 'guesstimated' the Flywheel power and torque, based on the most conservative 'drag or run down' figures I have for my car from the numerous rolling road sessions it has had. Hence somewhere in the region of 118 lbft at 5,000rpm and about 145 hp at 7,000rpm.

OK, well I have had the car on a rolling road, and below is the result. I went to a company that primarily deals with motorbikes. They also have a 2 wheel car chassis dyno for bike engine cars, and efforts such as mine. Unfortunately, the software does not convert to flywheel power and torque, so the figures you have below are at the wheel. Now I know that different rolling roads produce different wheel figures due to the different roller set ups, and of course different run conditions on the day. The figures put up are 105 lbft at 5,000rpm and 121 bhp at 7,000rpm. That’s about 9 lbft and 16-18 bhp more than recorded before at the wheels. How does that relate to previous dyno runs? Well I don’t know for sure until I take the car for a run at a different dyno that can work out flywheel figures.

I have based the figures on the least amount drag recorded at any of the dynos used. I appreciate its only a ball park, but I don't think it will prove to be much different from the estimate I have given above.

Rich,

Thanks v. much. I appreciate your comments. Though I do enjoy working on the car and developing things myself. If I were to do things again? Well, hindsight is a wonderful thing.

Regards, James

[Guy Croft]

mmm... at the wheels of course. Silly me. Still, that 6lbf ft gap is the hardest bit. Don't forget Steve's is on a common plenum.

I've been thinking over your thread. It's a comparison between DCNF and sidedraft (bike) carbs. 90 degree bend with the DCNF. Maybe you'd have got the same gains on DCOE. When someone gives me a constructive explanation as to why variable venturi bike carbs might be are than car race carbs (DHLA, DCOE) I'll try and listen!

G

[James Bowen]

Guy,

The comparison with Steve's power plot is to show that the 'Shape' of the curve is the same. i.e. dips in torque etc, not to compare outputs.

When comparing the dyno plot from Steve Sharpes car to my own, it would make me believe that exhaust design is an overiding factor in determining the shape of the torque curve. (?) Our different induction systems are chalk and cheese

Already stated that I'm not making a case for the Bike Carbs over any other.

Are bike carbs better than DCOE’s? I do not know. I would not be so bold as to make that claim, for one I have no experience of the latter. Also, as you have read, there are many snag factors to this set up, and the DCOE’s have the benefit of being tried and tested. Also there is an off the shelf manifold, which appears to be proved to work

As you identify, the comparison in this thread is very much about the difference between a sidedraft and downdraft set up.

My car is used for Sprints, and trackdays, but mostly on the public road. (hence posting in "General Discussions - Road") I gave this set up a try rather than go a more conventional route because I wanted to. It works, I have gotten good results in terms of tractability and low / mid range, and netted a significant improvement over the DCNF's. I imagine a set of DCOE 45's on a race engine will produce more power than I have. With your permission, I am simply sharing my findings and backing it up with data. Members / readers are welcome to listen or dismiss as they want. At least they'll have a little bit more to base it on!

Regards, James

[Guy Croft]

Right - on James,

I do delve beyond GC Q&A from time to time!

G

[James Bowen]

Just a little update.......

Had the car rolling roaded again last week, to get a Flywheel figure as opposed to the Wheel figure of before. I used the same Rolling Road as the "DCNF just prior to Bike Carb swap" in the interests of getting a fair (as can be comparison)

The figures were a little down on what I was hoping, but none the less not bad:

145 Bhp at 7330 Rpm - Flywheel
115 lbft at 5335 Rpm - Flywheel

So 97 bhp per litre and 77 lbft per litre. And an increase of 15 Bhp and 5lbft over the DCNF's with No other modification

The car was also run up with and without the Air Filters for comparison purposes. The result was no difference to power or torque, except the fuelling was slightly more stable with the filters on.

Hope that puts a nice finish on the thread...

Regards, James