The throttle installed into the simulator is a converted genuine B737-300 throttle. Lately, I have observed inconsistent power thrust issues during the take off roll and climb out.
As I begin the take off roll, engage TO/GA and rotate, but before acceleration altitude or acceleration height is reached, one of the throttles looses or gains power. Moving the throttle handle reinstates throttle power, but the power is dependent on where the actual throttle lever is physically positioned.
When the aircraft is above thrust reduction altitude (1500 radio altitude) the problem rectifies itself. The problem cannot be replicated when flying above 1500 feet. I also noted, and this may also be part of the issue, that the power indicators located on the EICAS display fluctuate (twitch) a little as I moved the throttle levers.
This problem only began to occur after I transferred the avionics software to ProSim737.
Process of Elimination
Problems like this are not uncommon when interfacing real aircraft parts and the challenge is finding the cause of the problem. The only method to determine solutions to problems such as this is to systematically, through the process of elimination, identify the problem area.
My first thought was that one of the potentiometers in the throttle quadrant maybe damaged, although I considered this to be unlikely as the units are still relatively new. The throttle has four potentiometers: throttle 1, throttle 2, flaps and spoilers. Flight testing indicated that the power loss alternated between engine 1 and engine 2; therefore, the likelihood of two potentiometers failing at the same time was minimal.
The next step involved checking the wiring within the throttle quadrant, to ensure there wasn’t damage to the outer coating of the wires. A damaged or loose wire can easily short on the throttle frame and generate a spike. However, if the wiring was loose or damaged, the problem would also occur when flying at altitude, and I had clearly demonstrated that the problem only occurred during the take off roll and climb out to thrust reduction altitude.
The next step was to ensure that calibration of the throttle unit was correct.
Re-Calibration Using FSUPIC
I decided to re-calibrate the throttles using FSUPIC rather than FSX. This process isn’t difficult and FSUPIC allows you to fine tune each throttle with greater accuracy than is possible with FSX.
After re-calibration, the “twitching” of the power indicators ceased, but the initial problem remained.
The Cause of the Problem
The only culprit I could think of to cause this problem was ProSim737.
To check whether ProSim737 was actually the cause of the problem, it is necessary to remove any input from the ProSim737 software. This is straightforward. Either use another avionics software package or use FSX itself. I did twenty trial flights using both Sim Avionics and FSX and the problem did not replicate.
ProSim737 Excellent Support and Advice
I contacted the developers at ProSim737 explaining my problem in detail, and I received a response to my questions within a few hours. Marty was especially helpful and we discussed several potential reasons for this issue and possible workarounds. I must stress that the response I received from ProSim737 was absolutely 100% top notch.
Marty genuinely wanted to help resolve the issue – whether it be with ProSim737 or otherwise.
Real B737 Throttle Operation
Now this where the comment “as real as it gets” does have meaning….
The developers of ProSim737 have designed their software to replicate the logic used by the real B737 auto throttle. The software (ProSim737) is doing exactly what it’s supposed to do in relation to power thrust, and the issue I was experienced is caused by using a real aircraft throttle without automation. Let me explain.
In the real aircraft, when TO/GA is enabled, the auto throttle logic has control of the aircraft. The throttles are off-line and power thrust cannot be manipulated by the pilot. The flight mode annunciator (FMA) illuminates N1.
As 84 knots is passed the FMA changes from N1 to THR HOLD. At this time, the actual throttles come back on-line, meaning that you can manually alter throttle power by moving the levers. After rotation and at 800 radio altitude the auto throttle system is ready to change from take off power to climb power and the FMA changes from THR to ARM. When in ARM mode the throttles are still on-line.
When the aircraft reaches 1500 RA which is the thrust reduction altitude, the throttles go off-line and the AT logic is controlling the power thrust of the throttles. The FMA changes from ARM to N1.
The B737 does not have a manual throttle, but an automated throttle. The software is programmed to move the throttle levers to the correct position mimicking the actual power thrust called for by the auto throttle logic.
If you use a manual throttle (genuine or otherwise) the connection to the automated physical movement of the throttle levers is missing; you must counter this by moving the levers yourself. This issue should not occur with a correctly calibrated automated throttle.
Using an Auto-throttle
If you have an auto throttle, the levers will automatically and physically move to the indicated thrust position as determined by the auto throttle logic (90%N1 at TO/GA). When the FMA illuminates THR HOLD at 84 knots, and the throttles come back on-line for possible pilot intervention, the auto throttle logic will not sense any change in the throttle lever position, and power thrust (90%N1) will be maintained. This is because the automated system placed the throttle levers in the correct position when TO/GA was initiated.
Using as Manual Throttle
However, if you’re using a manual throttle, the throttle levers MUST be physically positioned at the correct location on the throttle quadrant, otherwise the auto throttle logic will sense a change in position of the levers and alter the power thrust accordingly to this new level.
This is what was occurring in my situation. I was resting my hand on the throttle and only advancing the levers 3/4 of the way forward. TO/GA indicated 90%N1, but when the throttles came on-line at 84 knots, the auto throttle logic noted that the position of the throttle levers was not at 90%N1 and subsequently altered the power thrust accordingly.
The reason the issue was inconsistent is that I didn’t always advance the throttle levers to the same position, and if I did the problem did not occur.
LEFT: B737-300 throttle quadrant converted to Flight Simulator use. The TQ is a manual throttle meaning that the thrust levers are not automated and must be moved manually. I have used a pencil to lightly mark the metal adjacent to the most commonly used N1 settings. This ensures the levers are moved the correct location during take off. Lever position is set to 90%N1 and flaps 5.
Solution – Change in Procedures
The solution to this anomaly of using a real “manual” throttle is relatively simple.
You must determine where on the throttle quadrant the various N1 power settings are and then ensure, after engaged TO/GA that you move the throttle levers to the correct position (90%N1). In my situation, the procedure is to advance the throttle to 40%N1, engage TO/GA, and then manually push the throttle levers to 90%N1.
I’d like to thank Marty at ProSim737. Marty worked with me to solve the issue, which ultimately was not really a problem with either ProSim737 or my set-up, but is an anomaly of using a genuine throttle unit without automation.
I may at some stage update the throttle quadrant to allow automation of the thottle levers and speed brake, however, for the time being the TQ will remain manual.
Here is a picture (courtesy of Frazier ProSim737 forum) that provides an ideal visual cue to what has been discussed in the post.