Doesn't maintain altitude in turbulence

[SebComm]

We have a Sorcerer system installed in the prototype Compair 9 that will hold altitude in smooth air, but any turbulence and it starts climbing? The pilots report this, I haven't experienced it myself. One report was that after disconnecting the AP after it started climbing (uncommanded) that the aircraft was trimmed to climb - there is autotrim installed.
We had issues with the static connection at first, and now have dedicated static ports. Connected to the primary static system, the autopilot would not hold altitude at all. We opened the a AP to cabin pressure and it worked much better, so we installed dedicated ports and the short test flight we got seemed to work ok - but after the latest trip, no so much.
Could we have a problem with computer? The servo's are heavy duty models and pitch has good authority at less than have torque / activity settings. Roll is maxed out...

We have 2 more of these aircraft under construction and we to get this system worked out very soon!

[TrutrakTech]

Carl? ;)

Do you have an autotrim bypass plug? If not, I can send you one. The plug eliminates the autotrim from the system so we can see which one is actually causing the problem.

I would also keep the vertical torque set to at least 11. Even if it seems to fly well at 6, that may not be enough to overcome slight out of trim conditions and/or control it in turbulence. Let me know. Thanks!

[SebComm]

Yes, please send me a bypass plug.

Thanks

[SebComm]

We have the aircraft back at our shop and I did some testing. I cannot get the Auto Pitch trim to run in both directions. I have checked all wiring, tried all combinations of switch positions on the Autotrim module and even tried a different pitch servo. I can change the direction of autotrim operation, and the manual electric trim operates in both directions.
We are sending you the computer (Sorcerer) and the pitch servo (DSP HB24V).

[TrutrakTech]

I'll keep an eye out for them as I assume they've already shipped. Please remember to get an RMA number for any returns to us here, otherwise they will sit on our receiving shelf indefinitely. :) Thanks!

[SebComm]

I read the post from GCannon about his autotrim issue and it appeared to be the same scenario. I added a jumper from pin 4 to pin 1 on the Autotrim plug, but no joy. I checked the voltage on pin 4 during operation and it varies between 0 and 5 vdc, about 2.5 when neutral.

Don't know if it makes any difference, but this aircraft has a linear actuator for electric pitch trim instead of a RAC servo. I don't believe it draws a lot of current, but I'm sure its more than the little RAC units.

Carl

[TrutrakTech]

Carl,

That could be the problem if you have one of the older style autotrim units. They weren't set up to handle a whole lot of curent. The new models can handle up to 5A @ 12v. We'll check it out when it gets here. Thanks!

[SebComm]

We received the new style autotrim unit today, installed it and it works fine. But it seems I had more than one problem, we have a divergent pitch oscillation that I cannot adjust out. Turning the static lag up to 2 reduced the oscillation. I had to use the autotrim bypass to get anywhere - having the autotrim chasing the oscillation made for an interesting flight.
We originally had the A/P connected to the primary static system. It would oscillate, especially in turns. We split the static system into a primary and secondary (for copilot instruments) and connected the A/P only to the secondary static system - no better. We opened the A/P static system to the cockpit and it worked pretty good (not perfect) for a while. We added dedicated static ports for the A/P on the aircrafts last visit.
The autotrim problem then manifested itself so we didn't know if the new static ports helped or hurt, until now.
Why are we having so much issue with a static system connection to this system? I've installed quite a few of these systems and never had this much trouble, could it be a problem with this computer?

[SebComm]

Lucas, we would like to send the this Sorcerer processor in to upgrade to latest software and give it check up. Its currently v2.25
We are about to do a test flight with one of the newer computers we just received, I have already noticed a few differences. One is the GPSV button now says EXT? Will there be an update to the Operating Handbook?

[TrutrakTech]

As soon as I can get the handbook finished up (I work on it when time allows), I'll get it online. The only difference is we've now put GPSS and GPSV on the same button. One push of the GPSS button will engage both functions. The EXT button now controls the External DG mode.

RMA number 27288.

[SebComm]

Lucas, I want to send you another heavy duty servo to upgrade clutch to 100 in/lbs. Servo p/n DSP-HB, s/n 206787.
I think we should do the 2 new servo's for the CA-9's currently under construction: p/n 8100-050, s/n 100716-8100-050-001 and s/n 100720-8100-050-001.
RMA please...

Thanks
Carl

[PhantomPholly]

We originally had the A/P connected to the primary static system. It would oscillate, especially in turns. We split the static system into a primary and secondary (for copilot instruments) and connected the A/P only to the secondary static system - no better. We opened the A/P static system to the cockpit and it worked pretty good (not perfect) for a while. We added dedicated static ports for the A/P on the aircrafts last visit.
The autotrim problem then manifested itself so we didn't know if the new static ports helped or hurt, until now.
Why are we having so much issue with a static system connection to this system? I've installed quite a few of these systems and never had this much trouble, could it be a problem with this computer?

Sounds like you've tried about everything I've tried - although cabin static worked acceptably for me (I didn't need to go to a dedicated static source - may try that one day).

The very best solution you could have would be to have a combined pitot-static tube (one with static ports on the sides), with a dedicated feed to the autopilot. This does a couple of things for you.
1. The static tubing acts as a "buffer" slowing response time. Dedicating the tube, and even better using a smaller diameter tubing, will give you better response.
2. Most static ports are towards the rear of a plane. Since the tail tends to go down a bit before the airplane actually climbs, some of us have developed a theory that this causes the AP to over correct the pitch, contributing to the oscillation. Having the static source slightly forward of the center of lift and in "clear air" should make the reading optimal.
3. Is the airplane very "light" on the pitch axis? My Lancair will fly stable in turbulent air if I don't have too much stuff in the baggage area. Load it up back there (close to neutral stability) and the AP has trouble. For that matter, so do I - but I can fly the plane a lot closer to neutral stability than the AP can. Don't know about the Compair, but in the Lancair adding just a tiny touch of flaps changes the center of lift and thus stability a great deal - you might try that on your plane and see if it helps.
4. When you said it works "part of the time" even in turbulence using cabin air, that makes me wonder if you have a loose connection. Try re-seating all of the pins.

Pretty plane - Good luck!

[SebComm]

I installed the AP static ports just in front of the front doors, very close to the autopilot - 3' of 1/4" static tubing to each side to a tee fitting and 1.5' of tubing to autopilot.
Does increasing the static lag setting increase or decrease the autopilots sensitivity to static pressure?
The aircraft is not particularly light in pitch, but it is lighter that roll. We have the vertical activity set at 16 (verses 36 for lateral). We have the torque at 12 for both.
Pitch stability is not at issue, its very stable. I have on sveral flights got out of the copilot seat, walked to the back, crawled thru the baggage compartment to get to the pitch trim module. Ron says just a touch of trim change (if hand flying).
I think the 'part of the time' issue was a trim amp that stopped working in both directions that fooled us. That has been corrected.
Can you provide me an RMA to return 3 HD servos for upgrade to 100 in/lb torque? I am also returning the servo pushrod components after converting to capstan on 4 units.
Thanks,
Carl

[TrutrakTech]

I installed the AP static ports just in front of the front doors, very close to the autopilot - 3' of 1/4" static tubing to each side to a tee fitting and 1.5' of tubing to autopilot.
Does increasing the static lag setting increase or decrease the autopilots sensitivity to static pressure?
The aircraft is not particularly light in pitch, but it is lighter that roll. We have the vertical activity set at 16 (verses 36 for lateral). We have the torque at 12 for both.
Pitch stability is not at issue, its very stable. I have on sveral flights got out of the copilot seat, walked to the back, crawled thru the baggage compartment to get to the pitch trim module. Ron says just a touch of trim change (if hand flying).
I think the 'part of the time' issue was a trim amp that stopped working in both directions that fooled us. That has been corrected.
Can you provide me an RMA to return 3 HD servos for upgrade to 100 in/lb torque? I am also returning the servo pushrod components after converting to capstan on 4 units.
Thanks,
Carl

Carl, send me the serial numbers of each of the servos and I'll get the RMA number back to you. Thanks!

[PhantomPholly]

Sounds like you're going to be without your AP for a few days while your sevos are checked. When you get them back, here are some thoughts on how I got my AP tamed.

I installed the AP static ports just in front of the front doors, very close to the autopilot - 3' of 1/4" static tubing to each side to a tee fitting and 1.5' of tubing to autopilot.

That should be a short enough run - does your IAS look right as compared with GPS ground speeds? The only thing I wonder about that far forward on the fuselage would be "pulsing" from the prop, which is why I believe the best answer would be a static port on the pitot boom. I have no scientific evidence to back this up, but have long wondered if this might be why some folks have problems (e.g. the pulsing might "confuse" the AP). A way to alleviate this would be to have a dedicated static port run to your wheel well (retractable) or other spot outside of the prop blast zone that is in "calm air." I also suspect that cabin air would work better in this regard if only because "prop blast" runs all the way around, which might somewhat balance / buffer it.

Does increasing the static lag setting increase or decrease the autopilots sensitivity to static pressure?

The TT Engineers can give you the scientific explanation. My user experience is that increasing static lag tells the AP "don't rush to correct." That would be consistent with the idea that there might be a delay (lag) from the time you actually start climbing until the pressure in the static tubes actually decreases (and the reverse for a descent). However, "more" is not always "good." Since you have such a short run of tubing, I would think that your setting should be zero and you should be playing more with the Activity setting rather than static lag. The function of this setting seems a bit more "fuzzy" to me than Activity - if you get really close playing with Activity while Static Lag is zero, try bumping to 1.

The aircraft is not particularly light in pitch, but it is lighter that roll. We have the vertical activity set at 16 (verses 36 for lateral). We have the torque at 12 for both.

Again, my airframe is notorious for pitch instability. I have my Activity at around 4. I might increase that setting in high turbulence to as much as 7 maximum. My subjective experience with that setting is that it changes how "aggressively" the AP corrects to changes. Try reducing that setting to around 10, and if that works ok keep reducing it to the minimum necessary to prevent large excursions then bump it up one.

Final thoughts: The TT is a good AP. 99% of problems relate in some way to installation; however, this seems to be an area of some art as well as science since no one has an "ideal" installation.

[SebComm]

Servo p/n DSP-HB, s/n 206787.
I think we should do the 2 new servo's for the CA-9's currently under construction: p/n 8100-050, s/n 100716-8100-050-001 and s/n 100720-8100-050-001.

[SebComm]

Lucas, do you have a direct line or e-mail? I've posted the serial #'s for the servos and I need an RMA.

Thanks,
Carl

[TrutrakTech]

Oh Duh! I'm sorry. I completely thought I sent that to you. RMA coming right up.

RMA# 28126. Thanks Carl!