Saturday, 31 January 2009

Lean carb? Enlarge that jet!

Many carburetted RV report running too lean.

From very early in my experience with my RV4, it became clear to me that while I had no problem climbing at a high rate with reduced power, a full power climb for more than about 3000', resulted in the CHT heading rapidly over 400F. Since much of my early power flying, involved towing heavy gliders on hot days, with the throttle wide open for extended periods, without overheating, and at considerably slower speeds than my RV4, I found this less than satisfactory.

I have spent quite a bit of time understanding how the mixture is managed in the MA-4SPA carburettor, and  thought it might be worth capturing what I have learned in one place, since I had  found the relevant information scattered and hard to find. I can give no guarantees to its accuracy, but it is working well for me.

The standard test for the richness of the mixture is as follows.

Climb to 8,000-8,500 feet density altitude. The purpose of this is to get to a height where the engine can not develop more than 75% power and be damaged by leaning.
With wide open throttle (WOT), and mixture full rich, note all EGT's and start to lean very slowly. The EGTs will rise at first. Note the cylinder which is "first to peak" and note that peak EGT temp. This is relatively simple if you have a recording EFIS, but otherwise you will need one person to fly, and another to manage the engine, and take notes. I would continue leaning and gain an understanding of the behaviour of the successive cylinders.

Take the "first to peak". and calculate the delta between full rich, and peak EGT. 150F min difference at 75% power, rich EGT to peak EGT I understand to be ideal. 80-100F is too little.

Flight testing is always easier in theory than practice, at least for me! In the diagram to the right (open in a new tab if you want more detail) after the 8000' climb, where I am at slightly reduced RPM and higher than normal airspeed, to control CHT comfortably below 400F, I start leaning. Without a vernier control, the first time I do this, the fuel flow reduces so much, that the #1(red) ceases to ignite, and the EGT drops. You can see that I then richen it up and we establish a base EGT of about 1400F for #1. I then lean and it peaks at 1443F, for a total rise of only 43F. Far too lean! #3 rises about 130F and the left hand side of the engine does much better. The bottom line is the engine needs more fuel coming through, for longevity. The disparity in the behaviour of the cylinders is I understand a feature of carburetted Lycoming engines.

While there is a big screw on the back of the carb that is used for adjusting the idle mixture, it has no effect on the full power mixture. Don't touch it for this purpose.The full rich mixture is controlled by the pressure differential between the two ends of the main jet, and the diameter of the bore of the jet. This differential causes fuel to be sucked through the jet. The bigger it is, the more that comes through. (At this point I should confirm that yes, you can make it leaner of course with the mixture control, but you cant make it richer. The leaning mechanism on this model works by controlling the flow of fuel, from the fuel bowl, to the base of the jet.)

It turns out that the appropriate sized jet is somewhat installation dependent, resulting in different certified aircraft having different jets on the same engine. In our home built world, no one has done this testing for, in my case, an RV4. The 160HP O-320 I have, was supplied by Aerosport Power with an MA-4SPA carb and a jet of .096" diameter (#41).

As soon as I mentioned this issue to Aerosport Power they indicated they would increase the jet size if I returned the carb, or I could run a drill through myself. They had done this many times before and it was a 'routine' solution to a familiar problem. (If I were ordering the engine again I would raise this issue before hand, and I am sure it would have arrived with a bigger jet. Indeed I wonder why they do not routinely do this?)

Since I am in the UK, and returning the carb. was a little impractical, they in fact sent me a replacement jet of .104" dia. (#37) and the associated lock washers and gaskets etc.. I was a little concerned about the jump in size, a 17% increase, but they said 'This will work. We have done it many times.' 

I will come back to the process of changing the jet, but right now you are probably wondering what was the result. Well, here is the very first flight with the new jet.

The first change I noticed was that on pushing the throttle in for takeoff the historic flow of 50 to 51 lph (13.5USG) had jumped to about 59 lph. It was reassuring to see  an approximately  17% increase in flow, just about the same as the change in area of the jet bore. At the time, as the pilot, it was also reassuring to see that I could climb from the end of the runway to 8000' with the throttle wide open and the prop just short of 2700 rpm with no tendency for anything to start to bump into 'the yellow'. I flew the climb badly. I was trying to talk to ATC, monitor the engine and hunt for what felt like a good climb speed at what was a significantly steeper angle. However, with all that aside, the climb was at least 130fpm faster. As I said before flight testing is easy on the ground , but in the air I find it more difficult to get everything right. I had done the whole climb with carb heat on! (One day I will get some accurate climb figures but that was not today's mission.)

The result of this is that all of the data to the left of the 16:25:25 mark can be ignored for comparison purposes. You can see the second, dark blue line from the bottom drops, and that is where the carb heat is set to off.

I need more data yet for confirmation but if you look carefully at the peaks now, just to the right of the 16:25:25 mark, you will see rises of:

#1 334F red first to peak
#2 343F black third
#3 185F green second
#4 333F blue last

So now I am running rich, though its a much nicer situation than it was. In fact #3, with a rise of 196F, while not the first to peak, is only a bit richer than commonly recommended. I will do nothing in the short term, but it would be nice to try a jet that was drilled .1015" dia (#38). I would like to live with it for a while and collect more data, and also fly in colder air, and repeat the test. Even at 8000' the air on the day of the test was 31F.

It cant be so rich, because I have tried hard to make it rich cut, but its not hesitating. The power comes through smoothly.

Some readers might think that my overal fuel consumption will increase significantly. This is not true, since the mixture is normally controlled by easing back on the red knob, except in high power operations which are a very brief part of my flights.

One thing has improved which I was not expecting. The oil has always run rather cool. For the first time ever the oil temp was sitting just above 175F for a significant portion of the flight. This is certainly a welcome improvement.

So that is the result to date. I will add more if/when I have significant extra data. It does however leave the question of 'How do you change the jet?', which bothered me significantly before I had done it.

First of all you will need some parts. They are the gaskets you see in the picture, note the tiny one at the base of the jet, lock washers to hold the screws that hold the two halves of the carb together, a new lock washer to stop the jet from rotating (part 12 on the diagram), and an important and small split pin (part number 5o). You will then need either a new jet, or be prepared to drill the one you have. If you do it this way I suggest you go in small increments. I was lucky to be supplied a spare by my engine supplier, though I think it is perhaps one size too large.

I was tempted to split the carb on the aircraft, though was talked out of it. I am glad I was, because it would have been a mistake.

I removed the 4 bolts that hold the two halves together, but they were unwilling to separate at the gasket. If you have to take a knife, as I did, to ease the gasket away, try to leave it initially with the upper half of the carb., since it will be trapped behind the float assembly otherwise. 

As it separates, there are four items you have to be careful of:

1) the piston of the accelerator pump, almost out of sight in this first picture.

2)the brass fitting that responds to pulling the 'red knob'. It sits on a spring and has to be gently coaxed out.

3) the little clip on the centre of the float shaft and the needle of the needle valve which it grips and controls. If you double click on the picture, you can just about see these two part, just below centre.

4) the float assembly itself.

The carb is a precision piece of equipment so take it slow and gentle. Because the gasket stayed stuck to the lower half of the carb., with the blue float trapped by it, I had to gently clip the split pin on the float shaft. 

Either way, you will probably have to remove that pin sooner or later in order to install a new gasket, though it may be possible to do it without doing that.

With the carb in two halves you finally have access to the main jet, but make sure you spot the minute gasket on the lower end of the main jet when you remove it.

In this picture it is worth noting the copper tabbed washer that locks the jet in place. Just to the right is the jet that operates when you function the accelerator pump.

Once you have the jet out you can drill it or replace it etc., and then very carefully reverse the process and put it all back in place, remembering to do all the lock washers as you go. It is the end in this picture that is out of sight that has to be altered.

Make sure you hook the throttle cable through the same hole as before. 

On a certified aircraft I once found the carb coming loose despite the lock washers, so I have taken this rather unconventional step of putting a wire lock across the threads of the pairs of bolts as you see in this picture. I defy anyone to get a nut off without first clipping the lock wire.

With it all back together get it checked over and signed off, before you flight test it.

Information for this came from many sources and conversations. A particularly useful post I found on the Vansairforce web site was this one. I would like to thank all the others who have helped me. They know who they are.

Postscript 8th October '10 - I subsequently reduced the jet very slightly to a #39. This reduced the fuel flow very slightly at full power but still considerably richer than it was. Flow now peaks at 55 lph (14.5USG).


  1. Thanks for taking the time to post this. I hope I don't have to re-jet my O-320 when I test it, but it is great to see that it is not that difficult to do. What do you think caused your oil temperature to go up?

  2. Hey Retro - This is a most excellent posting. You're right...there is so very little about these ubiquitous Carbs online. I'm currently chasing an accelerator pump issue on a Cessna and this post has made me understand working with the carb much better, even though your issue was different. Thanks for the pics. I've just started looking at your Blog and love it.

  3. Brilliant Steve! Thanks :-)