Welcome to our resource for MG Car Information.
MG MGA - dyno test carb stacks
|A couple of years ago I inquired about carb stacks from a local super sales man mg specialist who has a dyno . Don't bother he said unless you are going racing. ? ? ? Sean 1500 A|
As I have not done any dyno tests before and after I cannot comment with authority. Nevertheless, as they say over here in one of our money saving supermarkets, 'every little helps'. If the thin stacks, like the ones I have made for my use, smooth the air entry flow into the carb, then that has to be an efficient modification with gains and no losses. It might not be much for the racing boys, but for us road goers looking for maximum efficiency from our precious gallon of fuel (£1.36 a litre or US$2.20 a litre)) it surely has to be a consideration.
|Where do you get it for only £1.36 Steve?|
|It's £1.32 here in the Highlands|
|I'll come up for my next fill Iain ...|
|Point taken Steve. It has occured to me that the smoother air flow improves throttle response ie., the engine picks up with less throttle opening whilst not necessarly increasing H.P.|
|Interesting point SS, I can see that in terms of max power with full throttle open you may be getting slightly more air into the engine and therefore a responding small increase in power(if jets are adjusted to suit) but throttle positions at other times are whatever they need to be to obtain the lesser power required. The throttle opening simply responds to the power (and therefore the air flow) required and a smoother airflow means it does not have to open as much.|
|J H Cole|
|I seem to recall that part of the tuning on Weber carbs is via fitting different length stacks but I may be mistaken|
|Some strange thinking above,I believe,. . The engine acts as a suction pump and draws air in and the objective is to suck as many air molecules as possible into the cylinders so when efficiently mixed with fuel the max energy is released for a given engine size. With a supercharger or turbo the suction effect is massively supplemented by an external compressor and the increased press compresses more molecules in. Without a supercharger etc the focus is to reduce the suction side pressure loss as the air enters the filter and progresses through the system and passages to the cylinders. This is why the passages are often ground/ contoured/ polished and why radiusing the sharp edge entry at the bottom of the pancake helps. At all partial throttle positions the butterfly valve destroys press. It is the max HP that is increased...fuel economy should not be affected. I did some very rough calculations to estimate the potential scale of HP benefit ....|
For a 1600cc engine running at say 2000rpm..
Vol of air per sec....0.47 cuft. Per sec
velocity through entry....21 ft per sec
Sharp entry loss from filter....1.23psi
Rounded edge loss from filter...0.31psi
Therefore potential reduction in loss....0.92 psi
therefore potential extra molecules entering carbs and cylinders etc ....6%
Above calc. Is optimistic and is only to estimate scale and given the real world, scavenging and valve inefficiencies ,combustion efficiencies etc etc ( and trusting my quick calcs) I would believe the order of a few percent increase in max HP and a quicker throttle response is a practical possibility.
|On a parallel subject the importance of keeping the carb air intake temp as low as possible is highlighted by similar calcs......every 10 deg F rise means about 2% fewer molecules reaching the cylinders with consequent max power reductions. .....so get the air intake hose in place if you don't have one.|
|Great calculation results! |
Note the air inlet grill and ducting on the Sebring cars.
See Frank Grahm's heroic restoration #43 MGA Coupe:
Fuel temp affects density of charge as well. Back in the old drag racing days ('57 chevy C Gas) we used a coil of copper line in a large coffee can filled with ice, it did make a difference in our ETs.
By now all of our heat shields should be back in place.
Jet coat the exhaust manifold...
It all adds up!
BTW, we are all whining about Regular gas being $3.99 per US gallon in California. Few people realize the price of fuel in the rest of the world.
|Neil, with respect, there seems to be something wrong with your calculations. First, taking your air velocity of 0.47 ft3/s as correct at 2000 rpm, with a density of .076 lb/ft3, and a filter entry diameter of 1.5", the velocity is 19.2 ft/s. At this velocity, the kinetic energy of air is 0.5*.076*(19.2)^2 ft poundals/ft2, or .0030 psi pressure equivalent. Entry loss for a sharp entry is about 0.6 of the kinetic energy. I'm not sure where you got 0.47 ft3/s. I make it 0.95, based on the swept volume of 1622 cc every 2 revolutions. I am assuming that the theoretical volume is sucked in, but maybe that is too optimistic and maybe you have some figures for that efficiency. Anyhow, with my assumptions and at 5500 rpm, I only make it an entry loss of .057 psi. It will be much lower at your lower volumetric efficiency. If you are correct, then at 5500 rpm, scaling by flow squared, you would have an intake loss of not 1.23 but 9.3 psi!!|
So my conclusion was that it is not worth the effort based on pressure loss, but there does seem to be an improvement in driveability based on Steve's experience.
|We have to assume that the factory designers included the stub stacks as standard in the MGB air filter canisters, there must have been a sound reason behind this decision.|
|Art..Thanks for comments..I did the calcs on the back of a ferry ticket whilst coming across the adriatic...Patras to Brindisi...but hope I am not that far out ( though I was surprised at the result being order of 6%).I am usually in the right ball park with quick calcs.. Still travelling but will have calculator and brain in gear on a flat surface in a few days .....|
|I am in the process of making a couple more of my own design stub stacks with the remaining material I have (other thread or my website for details). The thought occurred to me that it would be useful to have a second and perhaps more professional analysis of their effect rather than my 'seat of the pants' opinion.|
So, if anyone has suitable analysis equipment handy and wants to have a go measuring the before and after effect, then I am happy to donate this second set to the worthy cause of our hobby. Argue it out amongst yourselves and I will supply. Just remember that they will be drilled for the Volkes canister and that the welds on the internal posts will need to be dressed for the stubs to fit.
I do not pretend to understand all the maths on this one; that got left behind at school in 1965. However, the only observation I would make is that the filter entry is 1.39 inches diameter, not 1.5 inches that you used. The 1.5 inch hole is between the carb and the manifold.
|OK Steve, seems that surprisingly no-one is jumping in so I will offer to dyno-test the stub stacks. After making sure the stacks fit correctly I will remove and take the car to the dyno at JDM engine tuning and get it set up to optimum, then re-fit the stub stacks for comparison. Then we can all see the comparative data!|
|Ok Neil, sounds good to me. Email me your address and I will post them when finished. Might be a week or 2 as I have a busy schedule - become an OAP next week!|
|Thanks for the correction Steve. Small difference compared to the final result. Regarding airflow, does anybody know the relationship between theoretical flow (displacement x rpm/2) and actual ?|
To ensure the test is a comparison of with and without you cannot change anything on your car except to install the parts. Looking forward to the results.
15% difference in surface area of a hole would seem significant to me.
As I understand it from what you and Neil are trying to say is that a piston on its suction stroke is trying to suck the volume of one swept cylinder through a 1.39" diameter filter hole in a time determined by the engine speed. This will lead to a reduction in pressure through the carb and manifold as the air tries to negotiate the right angle turn through the filter box (molecules colliding etc). If I have got this right then it seems obvious to me that smoothing off that right angle turn is bound to assist the air flow and get more air into the chamber in that same period of time.
|Just a thought on accuracy of the dyno-test. If it is accurate to say plus or minus 1% then when measuring 80 bhp there would be a tolerance of 1.6 bhp that may be in the same order as the change being measured by the carb stacks. There is also the issue of whether the second test with carb stacks can 100% fully replicate the first test in terms of engine performance. One test to give confidence in the results would be to do the first test 2 or 3 times and see if there is any variation. This is pure summize on my part since I've never done a dyno-test but it might be an idea to ask what accuracy can be expected. |
|J H Cole|
|John and David, I plan to get my car set up without the stacks and then fit them there and then to compare. If I can I might test the standard MGB stacks and anything else I can think of or find (like no filters). It will depend very much on how things go on the day. I will of course confirm test accuracy and repeatability and publish all the nice colour charts.|
|Neil I'm particularly interested in the results of no air cleaner. I've always thought that these were restrictive mainly because the air has to make a sharp 90% turn. Maybe you may be able to test with just short ram tubes and no filters ? ? Sean|
|Steve, yes 15% is a significant change in flow area, but my point is that the friction loss from the sharp entry into the carb is so low to start with, there is really nothing significant to gain. I have assumed what is I think the maximum possible air induction. It will be less I'm sure due to factors such as the cylinders are hot and draw in less air mass, plus not all the gases get scavenged. With the revised diameter 1.39" I get .077 psi loss at 5500 rpm. |
My bet is the dyno test results will be zero, within experimenpheric error. things like atmospheric conditions will alter the result far more than the entry effect.
It will be interesting to see what results Neil gets with his tests.
Whilst accepting that the overall BHP may not improve much, if any, I have certainly felt a much improved throttle response. i.e. better acceleration for the same throttle movement I am used to making. Does the dyno test show this or is it all done at stabilised RPMs?
|Don't know Steve, but I would guess a dyno test is flat out maximum. The throttle response thing seems to be the benefit, which is certainly a nice thing.|
|Art..now on a flat surface so re-did the calcs. on a larger piece of paper. .|
Some comments first..
The volume I quoted at 2000 rpm was correct.It is for each filter . the veloc was for a reduced diameter throat to the carb. You quote 0.95 for engine I think.
I dug in a little deeper and also attempted to estimate the filter loss to get closer to the pressure at the entry.....difficult but using some actual test data on the web it would seem to be somewhere between 0.25 to 0.62 psi based on a google literature search.I would believe the small mga units would be to the high end so I have used such.
Based on this and using a truly excellent online gas flow calculator I discovered ( www.pipeflowcalculations.com ) and at an engine speed of 5500rpm I got following results..
Flow per filter........ 1.3 ft3/sec
Velocity from filter into carb entry ....141 ft/sec
Loss due to sharp entry.... 0.08psi.. ie about 0.56 % fewer molecules entering cylinders
My gross error in the original cals was in the air density ...no more quick calcs. for me. Thanks for the cross check!!
One point though and that is in ref to the very high velocity ..it is 13% of the speed of sound and is outside the normal range of flow calculations for flow in pipes. The fluid momentum at the centre of the filter is very large and the fuid has to turn through 90 deg in to the entry. I believe there is a large margin for error in the loss coefficient which will be based on experiments at much lower velocities. It will be good to see the results of the tests.
|Neil, send me your email address|
Mine is firstname.lastname@example.org
I'll send you my spreadsheet. I think we are boring the rest of the members!
|Art.. To be on this site often plus slaving over a hot mga one has to be an Aspie or similar and fairly obsessive..but so what! My wife has been diagnosed with the syndrome ( only 3 out of 10 thank goodness)and claims they seek each other out. On a more serious note the calcs are very limited in this situation and I also believe we are looking for a benefit in the range 0 to 2% at max engine umph so good chance the dyno test will be inconclusive.Probably better relative test would be to put a water gauge connected to a tapping (?) downstream of filter,run up the engine to and get the relative water head loss over filter etc with and without Steves creation. 1% extra loss would be about 3in WC . This is the technique used to judge the relative losses for different filter nmaterials and assemblies and is cheap. I am away from car until mid Sept and will probably do when I can get my hands on it..... I will reshape the 'horn' I have in the housing before doing.|
|OOOPs.... 1% is about 6in WC....total loss over housing and filter plus entry likely to be in a range up to 12in WC..so any significant change will be detectable.|
|Neil and Art,|
I'm not going to pretend to fully comprehend what you two are on about. But as I'm trying hard to understand it, it is not boring me, so please keep this up on the board.
|Willem vd Veer|
|As mentioned I intend to pursue and see if I can get a reading of head loss in water inches at max HP revs for filter assembley with and without the stub/rounded entry. I can't lay the sweaty hands on the car for another month to inspet for best way to do it and the main difficulty I know is going to be getting a decent tapping downstream of the filter assembley and upstream of the venturi in the carb. To go downstream of the carbs introduces a load more losses and complications . Last resorts ...I may make an insert to go between the carb and the filter assembley with a tapping in the insert or altrnatively feed a small dia tube through the filter and down past the rounded entry . Both will slightly change the dynamics ( epecially the tube ) but the relative comparison ( with and without ) will be still about correct. ..other option is to drill into the carb body ( ugh!!). Anybody got any ideas from looking at their own cars.? |
Your maths are giving Victorians a bad name, maybe you should change your location to NSW. (lol)
|Split my time between the two states normally..maybe contamination has hit me as I used to be an ace at maths and flow calcs..earned my living at it many many moons ago !|
|Your forgiven, if you overlook my spelling mistakes!|
Both stacks machined. I think I need to broaden the throat a bit on one of them and then drill them out. I should be mailing them to you later this week for the dyno tests. I am quite excited to see the results or at least a second opinion. I having been taking the car out on numerous occasions and remain thrilled with the extra feel/performance.
Not sure my old 1950s lathe will stand up to mass production. We may have to settle for a Far Eastern backstreet pedal lathe!
|Neil, how about using a pair of the thick heat blocking gaskets just behind the filter? You could drill a pressure tap through these.|
|Art.....back in Oz and will be laying hands on the car next week. Thanks for suggestion ..will check.|
|Looking good Steve!|
|They look great Steve. I ordered a pair at my engineer friend, and he tells me they should be ready for me next week. Can't be sure though as he is extremely busy. I will post a picture for you to see the result. I'm keeping my fingers crossed that they turn out identical to yours.|
Finished the pair. Putting them in the post this afternoon. You will have to fettle to get them flush mounted on the filter box face plate. I have countersunk the reverse sides of the holes but you may need to enlarge and/or dress the welds on the tubes.
Frank; the smaller holes should be drilled to 7.5mm not 6mm as I stated previously. That said, I doubt whether this is critical. They are just air pressure equalising holes.
thanks pal, I will pass on the info to the engineer. I don't think he has stared them yet.
|Actually, I have just had a thought about those air pressure equalising holes for the carb pistons.|
Now that their orifice is raised away from the filter box face plate they now have more uniform cross-airflow over their surface. This probably alters the pressure differential significantly from the normal and could perhaps alter the rate at which the piston slides up during acceleration?
|Stacks received (last Thursday) and dyno booked!|
They are real works of art Steve and much thinner than I had imagined (despite the dimensions clearly shown above!)
|Stevie G, -- I took my best shot at making an engineering drawing of your new stub stack. I didn't have an air cleaner or gasket handy, so I was scaling some of your pictures to get location of the drilled holes. Could you check my drawing to see if I have it close enough to fit? See here: http://mgaguru.com/mgtech/power/pdf/air_plate.pdf|
What a terrific drawing. Without a gasket and cleaner you have done an ace job.
I make the centre between the 2 posts 2.55 inches.
The centre between the 2 air holes 1.85 inches.
I agree with the 0.44" measurement.
|Barney, nice to see you are an AutoCAD guru also, no end to your talents!|
|Steve, -- Okay, fixed up. See here: |
I also posted it in a web page so the whole thing should be perfectly clear:
Mike, -- I'm a career machine design engineer, and I've been using AutoCAD since the late 80's. There are quite a few CAD drawings on my web site for body sill panels, engine rear seal, flywheel lightening, trailer hitch, and more.
|Good stuff Barney, nice page. Let's hope the tests warrant the effort put in.|
This thread was discussed between 14/09/2011 and 09/10/2011
MG MGA index
This thread is from the archive. The Live MG MGA BBS is active now.