Note: because of the 5 images/post limitation this will be spread over four posts.
Jetting a 1036 (or 1038) Concentric for a Gold Star Charles Falco
The 1½" AMAL GP supplied on Clubman Gold Stars has an almost unrestricted bore for flowing large quantities of air so is excellent for the race track. However, the lack a throttle stop and way the float bowl is mounted make it less than ideal for use on the street. Three common replacements for the GP are a 1000-Series AMALConcentric, 2000-Series AMALConcentric (also known as Concentric Mark II), or Mikuni VM, all of these in either 36 mm (1.417") or 38 mm (1.496").
A 1000-Series Concentric is often the favored replacement on a Gold Star because it looks "more original" than the other two alternatives. However, a problem is nearly all 1000-Series Concentrics originally were supplied for 2-strokes so they have different needles, needle jets, jet holder, spray tube, and air jet passages than used on 4-strokes. Unless otherwise noted, what follows deals only with modifying 1000-Series Concentrics, and when the word "Concentric" is used alone it only refers to this version of these carburetors.
Differences between 2-Stroke and 4-Stroke Amal Concentrics:
At the top left of the next photograph are the two easily-changed 2-stroke components (needle jet and jet holder), and at the bottom left are the 4-stroke counterparts.
Note that the 4-stroke needle jet has a hole cross-drilled in it and the jet holder (622/128) is slightly longer than the 2-stroke versions. Although not shown here, the outlet ends of the jets are different as well. The next composite photograph shows the spray tubes of a pair of 932 Concentrics, with a slant-cut 2-stroke spray on the left a flat-top 4-stroke on the right.
Some carburetors, such as the 2000-Series Concentric Mark II, have an interchangeable air jet in the central passage below the main bore that can be switched in order to preferentially affect the mixture at high engine speed. Not obvious, and rarely mentioned, is that the air jet passage in a 2-stroke 1000-Series Concentric body has a diameter of 0.193" over most of its length, the same as for a 4-stroke body, but an "air jet" restriction of 0.089" at the inside end, whereas 4-stroke bodies do not have this restriction. This is shown in the next photograph, that also shows the replaceable pilot jet found in all 2-stroke bodies but only in 4-stroke bodies made shortly after the introduction of the Concentric. An important note that applies to Concentrics of all Series on any motorcycle is that a poor float bowl gasket might not seem important since the fuel level is well below the surface. However, the pilot circuit draws fuel from a passage in the float bowl so if the gasket allows an air leak at the joint at that location the pilot mixture will be affected.
The fixed restriction in the air jet passage serves the same function as a replaceable air jet in the Mark II, and its reduced size affects the mixture at high rpm more than at low rpm. In addition, since the Welch plug has been removed the two pilot holes that emerge into the main bore on either side of the back edge of the slide can be seen. The smaller one is 0.026"-dia. (#71 drill bit) and the larger 0.035"-dia (#65).
Below I show that good performance can be achieved on a Gold Star even without changing the spray tube or drilling out the restriction in the air jet passage. I also show what performance is achieved if changes are made to the main body to convert it to a 4-stroke body. These two approaches are termed Modification I and Modification II later in this document.
Flow Bench Measurements
I made a series of flow bench measurements on 932 Concentrics with flat-top ('standard' 4-stroke), notched (Norton) and angled (2-stroke) spray tubes, the latter with and without a 0.089" restriction in the air jet passage. These measurements showed a 2-stroke spray tube increases the pressure drop at the top of the spray tube over that of a standard 4-stroke spray tube at all throttle settings (e.g. by ~21% at half-throttle). The same effect, although reduced in size, is present with the 0.089" restriction in the air jet passage (~15% at half-throttle). In essence, this means a smaller main jet would supply the same amount of fuel in any carburetor with a 2-stroke spray tube, although there will be a difference in the size of the jet depending on whether or not the body also has the 0.089" restriction. A smaller main jet also would be required if only the spray tube is changed to a 4-stroke but the air jet restriction left in place. Also, although the spray tube increases the depression over the entire range of throttle settings, the shape of the curve affects the details of the jetting at other settings as well.
As indicated by the next photograph, turbulence in the smooth-bore AMAL GP on the right would not be the same as in a Concentric, which would explain the measured 8% difference in air flow between the two.
The above composite photograph of a 1038 Concentric and 1½" GP were shot from the same height so the inlet trumpet on the Concentric at the left really is much larger than that on the GP, and the more complex path the air has to follow when entering a Concentric due to the location of the air jet and pilot air passages can be seen. In any case, if maximum h.p. at full throttle is the only consideration it would be better to use a GP than a Concentric.
Clubman Gold Stars came with a 1½" AMAL GP so the closest Concentric is a 1038 (38mm/1.496"). However, the inlet track of my DBD Clubman head is 36.8mm/1.449" so a 1038 would result in 0.024" step up at the head, whereas a 1036 would have a 0.016" step down. The step up can be clearly seen through the bore of the carburetor when a 1038 is attached to the head, as shown in the next photograph.
In the case of the GP, which my measurements show flows ~8% more than the 1038, the 0.026" step up was noted by the factory to provide a slight increase in h.p. at maximum throttle. However, I'm unaware of any data showing whether this same effect occurs with the Concentric.
------------ to be continued ------------------------
Another consideration in selecting a carburetor is that an inlet port obstruction, such as from a step-up or a gasket with too-small bore, can have a serious effect on the starting and low speed running of an engine. The next photograph of an old carburetor shows how the mixture from the pilot holes hugs the bottom edge of the carburetor at least as far as the head, where a step-up can cause coalescence of the emulsion which would upset the mixture. This effect would be different in a GP since it introduces its pilot mixture through the spray tube rather than through holes in the floor of the bore.
Presumably, the main reason for switching from a GP to a Concentric is to obtain more "civilized" behavior, which includes ease of starting and ability to idle smoothly, both of which might (or might not) be compromised with a 1038. For what it's worth, although my flow bench measurements found a 1036 flows 4.9% less air at full throttle than a 1038, and thus the engine would produce that much less h.p. at full throttle, I have both a 1036 and a 1038 but I chose to use the 1036 on my 'Competition'-model Gold Star.
Fuel Height in Float Bowl
Although not required, it's a good idea to replace a nylon float needle with a Viton-tipped one. Although these are available in brass as well as aluminum, the lighter weight of the Al version is essential for a Concentric, especially if also using a new 'stay-up' float.
AMAL specifies that the fuel height should be in the range 0.170"-0.240" (4.3 - 6.1 mm) below the top edge of the bowl. It might seem that because the design of the carburetor is "Concentric" that tilting it would leave the fuel level unchanged in the center where the jets are located. This would be the case only if the inlet were blocked. However, the inlet is not blocked in operation.
Because the float is not symmetric, and because tilting the carburetor causes the fuel to be higher at the outlet end of the carburetor than at the inlet end, this closes the float needle sooner and lowers the overall fuel level. On a Gold Star this is a significant effect. Because of this the level at the center of the bowl has to be set when the carburetor is tilted at the same 15° angle as when it is mounted on the bike
The carburetor on a DBD Gold Star is at a downdraft angle of 15°. This raises the fuel level on the engine side by ~ 5 mm, equivalently lowering the level at the center by several mm because it closes the inlet needle that much sooner. This means if the fuel level is set to the proper 4.3-6.1 mm when the carburetor is vertical it would be double that when mounted on the Gold Star. Such a discrepancy has a serious negative effect on the operation of the carburetor, one symptom of which is little or no response to adjustment of the pilot mixture screw.
Again, the fuel level must be set to 4.3-6.1 mm below the top edge of the gasket (i.e. bottom of body) at the centerline of the carburetor when it is tilted at the ~15° mounting angle to the head.
Variations in Marked Sizes of Jets
Two cautions are needed about marked jet sizes. As for the needle jet, wear or mis-manufacture by only 0.0005" is enough to completely wear out the jet (or make it too lean, if manufactured slightly too small). The needle jet in my Catalina experienced this much wear during a 1200-mile ride so the needle jet definitely is not a set-and-forget component. Although the needle jet used for both modifications discussed below is stamped '.106' I previously had measured a large number of NOS jets using two different calibrated bore micrometers and found them to be 0.1065", which is how they were originally listed by AMAL.
As for the main jet, a decade ago I measured a dozen new and used jets of the same size on my flow bench and found 75% of them to be within +/-1 size of being correct. Unfortunately, the other 25% flowed too much or too little by as much as 3½ sizes. At one point when determining the jetting for the 1036 Concentric I installed a jet marked two sizes leaner than the one that had been in it but the air/fuel gauge showed the nominally-smaller jet actually flowed a half-size richer. Keep this in mind if your jetting becomes too rich or too lean when changing to a jet whose marking makes it seem it "should" be correct.
Unfortunately, without a bore gauge for the needle jet and an air/fuel gauge for the mixture all that can be done is to be aware of such issues if encountering behavior that doesn't seem to make sense when changing either jet.
Determining the Settings
To determine the best settings for both Modification I and Modification II I mounted a Bosch wide-band sensor in a Gold Star pipe in the location shown in the next photograph.
I connected the Bosch wide-band sensor along with a throttle position sensor and an inductive clamp for rpm to an Innovate LMA-2 and from there to an Innovate LM-1 air/fuel meter for logging data on the road at 12 Hz for up to 44 minutes per session. I downloaded this data after each run, analyzed it using Innovate's Logworks 3, and used that information to alter the jetting for the next run.
λ vs. AFR
From data published over the past several decades most engines produce maximum power for λ=0.82-0.88 where λ=1.0 is the stoichiometric Air Fuel Ratio (AFR) of the fuel. This ratio is 14.7:1 for the "pure" gasoline used for much of the published dyno work, but 14.0:1 for E10 and 13.8:1 for E15. However, relatively little dyno data exists yet for current ethanol blends so it's not clear at this time if the same λ=0.82-0.88 range will hold for them.
Further compounding the uncertainty is that despite what is marked on the pump the actual content of most fuels is an unknown. For example, pump gasoline in the U.S. can contain "up to" 10% ethanol, and E15 can contain anywhere from 10.5% to 15% ethanol. To compound it even further, the 10:1 piston in my 'Competition' Gold Star needs higher octane than the maximum 91 (R+M/2) available to me so I added 4.5 oz./gal. of Race Gas™ octane booster to increase it to 100 to eliminate pre-ignition. The effect of this octane booster on the stoichiometric ratio of the fuel is another unknown.
Despite the above uncertainties, and irrespective of the fuel used, a wide-band oxygen sensor detects the crossover point from an excess of oxygen in the exhaust to a deficit, i.e. stoichiometry. The Innovate LM-1 control unit then displays the stoichiometric value of that fuel as λ=1.0 and, since the sensor has been calibrated in "pure" air, the λ values it displays are accurate for richer (λ<1) and leaner (λ>1) mixtures of that fuel as well. Although λ is a more "fundamental" value because it doesn't depend on the fuel being used, the Innovate LM1- control unit allows it to be multiplied by 14.7, 13.8, or whatever and display the result as AFR.
The displayed AFR depends on what assumption has been made about the fuel while the displayed λ does not. Despite this, and the uncertainties about the fuel, since much of the published dyno data is in terms of AFR for pure gasoline, where stoichiometry is 14.7 and where λ=0.82-0.88 corresponds to AFR=12-13, I've plotted my results this way. However, I also show the measured λ, which is what should be used as additional dyno information becomes available about maximizing the h.p. with present and future fuel blends.
One other note is, while it's possible slightly more horsepower could be extracted by tuning on a dyno, the jetting I determined for both carburetor modifications resulted in a Gold Star that behaves very well at idle as well as on the road at all speeds and accelerations.
------------------- to be continued ----------------------
As an example, the next figure shows 10.5 seconds of AFR data (where 14.7 was assumed), where the throttle position in red would be 1.56 V at full throttle and where the rpm reading in black is double the actual rpm.
Modification I: Only Changing Jetting-Related Parts
Making this modification is easiest since it only requires changing four parts. Starting with a stock 2-stroke carburetor the parts needed to alter it for use on a 4-stroke engine are:
622/124 Needle (2 rings on top with 3 grooves below the rings; length 2.676") 622/128 Jet holder 622/122 Needle jet, size .106" 1034//0604 (cutaway #4), slide
Unfortunately, parts for 1000-Series Concentrics are not as common as for the smaller versions, and even then most slides have #2 or #3 cutaways for 2-strokes. If a #4 cutaway slide cannot be located it's possible to modify one having less cutaway by machining a suitable amount from the leading edge as shown in the next photograph where the slide is mounted at 19 degrees as discussed below.
I don't know how much, if at all, it affects performance, but the angle of the leading edge on AMAL 1000-Series slides depends on cutaway as shown in the next figure. Irrespective of what cutaway a slide currently has, in order to mimic AMAL's slides when modifying one to become a #4 it should be mounted at 19 degrees from vertical. AMAL slide cutaways are in steps of 1/16" so a #4 is 4/16" = 1/4". Note that even a half-step in the cutaway has a significant effect on the mixture so careful measuring and cutting is called for.
While a slide with a larger cutaway can be modified by machining the trailing edge doing so also lowers the needle. This would be a problem for the jetting recommended below since it requires the needle to be in the highest position. However, if a slide with a larger cutaway is the only option available for modification, the difference lost by machining the trailing edge can be compensated for by inserting a shim under the needle clip of the same thickness as that removed.
Modification II: Altering the Main Carburetor Body
Modification II requires:
622/124 Needle (2 rings on top with 3 grooves below the rings; length 2.676") 622/128 Jet holder 622/122 Needle jet, size .106" 1034//0603 (cutaway #3), slide 622/074 4-stroke spray tube
The first three parts are the same as for Modification I, but in addition a different slide is required as well as a 4-stroke spray tube. The spray tube can be obtained from a sacrificial Concentric body of any size since they are the same for 600, 900 and 1000-Series Concentrics, and replacements also are available from some suppliers.
For this modification the angled 2-stroke spray tube is pressed out and the replacement flat-top 4-stroke spray tube pressed in using a drift, as shown in the next two photographs.
The OD of the tip of the drift is 0.246" in order to slip into the 1/4" ID of the spray tube, and the OD of the main part of the drift is 0.309" to fit through the 5/16" ID hole in the body.
---------------------- to be continued ----------------------------
In addition to changing the spray tube, the restriction in the air jet passage is removed using a 0.193"-diameter (#10) drill bit as shown in the next photograph.
With these two modifications the body will be identical to the configuration of 4-stroke 600 and 900-Series Concentrics.
Best Settings (General Discussion):
To make a direct comparison of Modification I with Modification II with the least uncertainty I used two AMAL 1036 bodies, swapped the float bowl along with its float that I had previously adjusted for the 15-deg. downdraft angle, and also swapped the jet holder, needle and .106 needle jet. Other than the two modifications to one of the bodies, this means the slide, pilot and main jets were the only components that were different. I then made a series of runs with these two carburetors on my U.S.-only 'Competition'-model DBD Gold Star that has 65-2446 inlet and exhaust cams, 10:1 piston, K&N RC-1250 air filter, and OEM 'twitter' silencer. As mentioned earlier, I used Shell 91 octane (R+M/2) pump gasoline which "may contain up to 10% ethanol" to which I added 4.5 oz./gal. of Race Gas™ concentrate to make it 100 octane.
The exhaust system and air filter have particularly large effects on the main jet for any motorcycle but, with this in mind, and after correcting for local altitude and climate, the settings given below should be a good starting point for jetting a 1036 for use on any "large port" (i.e. not Catalina) DBD Gold Star. Although the settings apply specifically to 1036 Concentrics, I expect those for 1038s would not differ by much.
Best Settings: Modification I:
Using the Innovate data system I determined the following settings:
Altitude: approximately 2500 ft. Air and fuel temperature: 95-98 °F Relative Air Density (RAD): 89-90 Pilot jet: #25 (1 turn out) Needle jet: .106 (measured to be 0.1065"-0.1066") Needle clip: bottom, #3, notch Slide cutaway: 4 Main jet: 200
With the above settings and weather conditions the Air/Fuel Ratio (AFR) as a function of throttle setting is shown in the next figure.
Depending on the conditions (e.g. accelerating up a hill, traveling at fixed speed on a level road, etc.) the AFR at a given throttle setting can be higher or lower than the above curve, which is an average that I drew as "best fit" through over one-hundred data points. Although the AFR of the blue curve is richer than 12:1 for most throttle settings at the test temperature of nearly 100 oF, the bike ran very well so this wasn't an issue (other than its effect on fuel consumption).
As noted on the graph, my AFR measurements were done with the air temperature at nearly 100 oF with the temperature of the fuel in the tank and of the outside of the float bowl the same. Erring on the side of too rich, rather than too lean, at these temperatures seemed prudent. Further, as the temperature drops the air will become denser so the mixture will become leaner. According to the Mikuni "Main Jet Tuning Calculator" at ~75 oF the equivalent main jet would need to be one size larger to give the same result or, if the same jet is kept, the curve will be ~0.5 AFR leaner which would be fine.
The rider can, after adjusting for their local conditions, use this graph to choose the settings they feel most comfortable with. Of course, this recommendation of a single main jet size is for "usual" street riding where the rider typically leaves the same main jet in place irrespective of changes in temperature or riding at different altitudes.
Best Settings: Modification II:
Altitude: approximately 2500 ft. Air and fuel temperature: 101 °F Relative Air Density (RAD): 89 Pilot jet: #30 (1½ turns out) Needle jet: .106 (measured to be 0.1065"-0.1066") Needle clip: middle, #2, notch Slide cutaway: 3 Main jet: 240
With the above settings and weather conditions the Air/Fuel Ratio (AFR) as a function of throttle setting is shown in the next graph. . Tuning for Economy vs. Performance
The spray tube and air jet restriction in a 2-stroke body result in a greater pressure drop at the outlets of the pilot circuit and spray tube than with a 4-stroke body. Unless compensated for with smaller jets and larger cutaway, this would draw more fuel into the air stream at a given throttle setting resulting in richer mixtures across the full range of throttle settings. As shown above, it happens that the jetting can be made to work reasonably well for a 1000-Series Concentric having its original 2-stroke spray tube and restriction in the air jet passage. However, the undulating shape of the AFR curve would present a problem if trying to achieve a leaner mixture for economy, as can be seen from the next graph.
The pilot and main jets are easy to change in order to alter the mixture at the lowest and highest rpms, but intermediate throttle settings involve the slide cutaway, needle position, and needle jet ID. Achieving a leaner mixture below ~1/8 throttle for Modification I would require a greater cutaway but that also would lean the mixture up to ~1/3 throttle. Whether the mixture in the range on either side of ~0.2 throttle then would be too lean would require additional work to determine. Another possibility to achieve a "flatter" response from the 2-stroke body would be to fill in the #2 and #3 slots on the needle and cut a new one halfway between them. Again, though, exploring this possibility would require additional work. In any case, as a very rough estimate, if most of the time is spent between 1/4 and 3/4 throttle Modification II would result in ~5% lower fuel consumption than Modification I when having the same AFR at full throttle.
If I may be the first to jump in here to congratulate MMan on this exhaustive (and for him, exhausting) exploration of the jetting of these Concentric carbs.
It is a quite brilliant explanation of the jetting process, with illustrations to boot.
Has such an exhaustive process ever been published before? Probably not even close. Presumably the engineers at AMAL know this stuff. They had to, to be able to design the carbs. But I doubt if this information has ever been in private hands.
We're all the better for having this information. Not that it does me any good. My ZB34 bitsa (not a GS) has some ancient carb on it that I dare not touch. And my yet-to-be built Catalina has a brand new Monobloc sitting on the shelf, awaiting its call.
1949 BSA ZB34 'Bitsa' 1959 BSA DBD34 Catalina 1973 Norton Commando 850 R 1974 Norton Commando 850 R (I know, one too many) 1975 Honda TL250 Trials, a new addition to the family 1998 Montesa Cota 315 HRC 2004 Ducati M1000ie
Well done MM. You obviously planned the above description, with the photos at each step of the way!
Othe overlay of the resultant 2 stroke and 4 stroke fuel curves is the kicker for me - it shows that the whole 2 stroke setup is an effort to offer a fueling option that party addresses piston port 2 stroke 'over fueling' when the engine is 'off the pipe', as intake air can pass through the carb 3 time, collecting fuel as it goes. the fact that a wholly acceptable jetting setup can be achieved on a 2 stroke Concentric, fitted to a 4 stroke is valuable (and will probably drive the price of 2 stroke 1000-series carbs up!).
No generalisation is wholly true, not even this one. Oliver Wendell Holmes
Thanks to all of you for the thanks. I certainly didn't expect I'd end up as deep in the weeds as I did when I started out on this three months ago. A sticky should make it easier to find and help at least a few people in the future avoid jetting frustrations, so thanks for that.
Originally Posted by Kerry W
the fact that a wholly acceptable jetting setup can be achieved on a 2 stroke Concentric, fitted to a 4 stroke is valuable (and will probably drive the price of 2 stroke 1000-series carbs up!)
Lucky for me I have as backups the '4-stroke' 1036 that's on the bike plus the now-jetted 2-stroke 1036 I started out with and a 2-stroke 1038, so I'm immunized against possible Concentric bidding wars.
I have a wide-band sensor mounted in a modified Gold Star pipe and two other Gold Stars so this probably isn't the last you'll hear from me about jetting. Although, most likely, additional jetting work will be on hold until temperatures drop in the fall. But, that gives me time to upgrade my "instrumentation package" to capture additional relevant parameters...
In my defense, this 1036 that arrived yesterday was priced too low to resist. Plus, it's a right-hand version whereas the "4-strokerized" 1036 I now have on the bike is left-hand. This forces me to waste an extra 5 sec. to reach the tickler each time I need to start the bike when it's cold. Clearly, the time savings alone justifies the purchase. It will now go on the shelf until I take the time to convert it to 4-stroke configuration and swap it for the left-hand version currently on the bike. In principle this could happen before hell freezes over, but I'm not making any promises.