Synchronization of Rotax 912 Carburetors

Gauge and color-coded valves for synchronization of carburetors
The Rotax 912 engine of our Zenith 601HDS, N314LB, ran smoothly from 3,800 rpm to maximum 5,800 rpm since overhaul of the muffler. But at lower rpm, there was some vibration. 

Since ignition of the Rotax 912 engine is virtually trouble-free and compression numbers for the four cylinders were almost identical, the cause might be uneven performance of the two carburetors. 

For other causes of rough running engine, see posts on gearbox, setting of propeller blades, and ignition cables

Carburetor Problems

Three things can go wrong. 

First, fuel-air mixtures in the two manifolds during idle may not be identical. 

Second, the mechanical operation of the throttles may have play so that throttle positions can change at random. 

Third, the position of one throttle relative to the other may be wrong so that different manifold pressures are produced. The latter effect is most pronounced at low- to mid-range rpm.

Fuel-Air Mixtures for Idle

The first item is easily checked and corrected. For each carburetor, the idle mixture screw is gently turned until it reaches its stop, and then it is opened 1 1/2 turns. 

This value may have to be changed later, but it suffices for investigation of items two and three.

Linkage System

During construction of the plane 20 years ago, friend Mel and I discarded the cable system supplied in the Zenith 601HDS kit for operation of the throttles.

Instead, we designed a linkage system that does not suffer from the friction and subtle length changes that cable systems are prone to have.
Linkage system for carburetors
Over 20 years of use, a bearing of the torque tube of the linkage system had developed some play, as had the rod end spherical bearings of the rods connecting the torque tube to the carburetors.

We added a Delrin bearing to the torque tube and replaced all rod end bearings.
Delrin bearing added to torque tube
At the same time, we converted one of the four rod ends from right-hand to left-hand thread with the help of friend Les, thus making the length of one rod adjustable without disconnecting a rod end. 

With these changes, the mechanical operation of the throttles was without play, and length adjustment of one rod with running engine had become possible.

On to the synchronization task, which is covered in Chapter 12-20-00 of the Rotax manual for the 912 engine.

Mechanical Synchronization

The first step is mechanical synchronization, where rod lengths are adjusted so that both throttles close fully at the same time. 

To achieve this, the rods are disconnected at the throttle links. On each carburetor the screw controlling idle speed is backed out until the throttle can be fully closed and there is a 0.004 inch gap between the idle screw and its stop. 

The gap is measured while light pressure is exerted on the throttle link. We used a rubber band for this.

Once the appropriate gap is achieved, the rod ends are reconnected to the throttle links, and the length of the adjustable rod is changed until the same 0.004 inch gap results when the rods are gently pushed forward to close the throttles.

Then each idle screw is turned 1 1/2 turns to create the correct throttle opening for idle. 

Finally, it is verified that the two throttles go simultaneously to their full-open stops. That check worked out perfectly, which was no surprise since the linkage system had no play.

When Mel and I built the plane, mechanical synchronization was considered sufficient for smooth engine operation. 

But in intervening years, Rotax has added a second step where manifold pressures are synchronized.

Synchronization of Manifold Pressures

For this, the compensation tube connecting the two manifolds is removed, thus allowing measurement of the difference between the two manifold pressures.

There are several ways to do this. For example, using a special differential gauge to directly measure the pressure differential, or two gauges to measure the pressures separately, or a single gauge that is operated via two valves connected to the two manifolds.

We opted for the third approach. Indeed, friend Jack gave us a WWII manifold pressure gauge for that purpose. 

Testing proved that, after use during WWII and then 70 years in storage, the gauge still was quite precise.

He also loaned tools so we could easily make a handy aluminum display stand for the gauge. See photo at the top of this post. 

A rubber mat was glued to the bottom of the stand so that it doesn't scratch any surface. 

Note that there is a hole in the display stand for a second gauge, made when initially we decided to use two gauges. 

But then we switched to just one gauge to avoid synchronization errors due to gauge inaccuracy.

The gauge is connected via a tee to two valves each of which is connected to one manifold. Thus, each manifold pressure can be read separately by opening one valve and closing the other one. 

Red and green color coding readily tells which valve controls which manifold.

The average manifold pressure can also be displayed, by opening both valves. The latter setting has the same effect as the compensation tube and thus can be used when idle speed and mixture are fine-tuned. More on this later.

For the synchronization test, the airplane must be firmly tied down. We attached the tail of the plane with a steel cable to our car to guarantee this.
Steel cable holding plane
We ran up the engine to operating temperature, then checked the two manifold pressures from idle up to 4,000 rpm. 

It turned out that over that range, the difference between the pressures was very small except near idle rpm. Changing the length of the adjustable rod by a small amount eliminated that difference over the entire rpm range.

The length change of the adjustable rod affected the idle speed setting of the carburetor operated by that rod. 

Thus, we adjusted the idle screw so that it hits the stop precisely when the other carburetor reaches its idle stop.

We also checked that both throttles can go to the full open position. This was the case after the mechanical synchronization. 

But after the small change of the adjustable rod, this was no longer so, but only by a very small amount.

The carburetors do not allow adjustment of the full-open stops, so we simply ignored that small difference. 

Indeed, when a throttle is very close to the full open position, a small throttle change has virtually no effect on carburetor performance.

Initially, we had set the idle mixture using 1 1/2 turns of the adjustment screw. But this need not be the correct mixture. 

We do not have equipment for an idle-mixture test, but did the following process, learned decades ago when cars had carburetors.

The muffler has a single exhaust pipe on the left hand side. Running the engine at idle rpm, we held a hand behind the exhaust pipe to sense the smoothness of the exiting combustion gases.

Stuttering pulses of the gases are a symptom of improper mixture, while a smooth flow indicates that the mixture is at or near optimum. 

So while holding one hand behind the exhaust pipe, we slowly turned with the other hand the mixture screw of the left carburetor toward a more rich and then to a more lean mixture, all the time sensing the smoothness of  the exiting exhaust gases. 

We stopped at the point of maximum smoothness. This coincided with minimum shaking of the engine, another confirmation that the mixture was correct. 

It so happened that this setting amounted to 2 turns of the idle mixture screw from the stop position, thus resulting in a slightly richer mixture. 

Upon turning off the engine, we reset the idle mixture of the second carburetor using the same 2 turns.

To complete the synchronization process, we removed the gauge connections and reinstalled the compensation tube. 

The test flight turned out be most pleasing: From idle to max rpm, the engine ran smoothly and without vibration during all phases of flight.

CAUTION

A running propeller is almost invisible, and there is a natural tendency to ignore the propeller while approaching or walking around a running engine.

Here are three rules promoting safety:

Rule 1: Wear a hearing protector while working on the running engine

It eliminates the loud wind noise produced by the propeller blast in your ears. With the hearing protector, you think more clearly, feel less pressured, are more aware of the running propeller, and focus better on the steps of the job in progress.  

Rule 2: Approach the engine from a wing tip, moving along the leading edge of the wing

Rule 3: Do not walk around the engine

Instead, depart from one side of the engine along the leading edge of the wing until you reach the wing tip. 

Walk around the back of the plane until you come to the other wing tip. Then move along the leading edge of the wing until you reach the engine again.

Yes, these rules are tedious. But in hindsight they will look like a time saver if you ever come in contact with a running propeller.

Have any questions or feedback about the synchronization of carburetors? Please share your thoughts in the comments.

Comments

Popular posts from this blog

Testing Rotax 912/914 Generator and Voltage Regulator/Rectifier

Rotax 912 Engine: 2,400 Hours in 28 Years (achieved in 2023)

uAvionix TailBeacon: Installation and Testing