Honeycomb Alpha Yoke Problem

Crashnburn

Pattern Altitude
Joined
Aug 8, 2018
Messages
1,776
Location
Erie, CO
Display Name

Display name:
Crashnburn
My yoke suddenly started getting sensitive in pitch around the neutral position. Also, the button that allows me to change viewpoint stopped working. (I know it's the yoke because my Eclipse yoke doesn't have that problem.

I have an email in to Honeycomb-Aeronautical, but so far no response, and I can't even leave a voice mail!

I'm wondering if any of you have run into this problem, and if so, how you fixed it.

Unless I get some support from Honeycomb, I won't be buying anything else from them.

Thanks
 
I might be wrong but I thought they went out of business.
 
Thanks for the reply. As far as I can te3ll, you're right. I found this article about them.


It's a shame, the Alpha Yoke was great while it lasted.
 
The company kind of imploded. The founder is trying to save it, but it doesn't look good. He's posted a couple times on FB, and there's some more info on reddit. The short answer is you're not likely to get any support in the short term, and likely in the long term. On reddit it appears that many, if not most, of the alpha yokes are beginning to fail, so even if the company survives, I doubt they'll be in a position to support them. I think the best hope is another peripheral manufacturer picking up the IP, but again I'm guessing the older hardware will be SOL.
 
Thanks. I didn't realize I was part of the Beta testing group. I think I pretty much have landing in a real airplane down; I just have to work on countering drift as soon as it starts. Rod Machado says a flight simulator is good for developing that skill.
 
Thanks. I didn't realize I was part of the Beta testing group. I think I pretty much have landing in a real airplane down; I just have to work on countering drift as soon as it starts. Rod Machado says a flight simulator is good for developing that skill.
Yeah, me neither. I don't flight sim much, so my alpha is still okay as far as I know; haven't used it in months. I find simming to be incredibly frustrating. I usually crash or ground loop when I land, in spite of having 1000+ successful landings in real airplanes. It seems like flying a sim is almost a completely different skill set.
 
Yeah, me neither. I don't flight sim much, so my alpha is still okay as far as I know; haven't used it in months. I find simming to be incredibly frustrating. I usually crash or ground loop when I land, in spite of having 1000+ successful landings in real airplanes. It seems like flying a sim is almost a completely different skill set.
I have been using a flight sim for a number of years. I have a G1000 mock up, and honey comb controls. I use Xplane 12 now. For landings and take offs it is absolutely useless. I have a mock up of my T206H. I have "landed" on the island where the Statue of Liberty is, on the George Washington Bridge, on the lawn of the White House, on an aircraft carrier, and numerous other places where landing my 206 in real life would be impossible. I find the sim useful for practicing IFR approaches, and practicing flying by instruments. Otherwise not so much. I have utilized it to reconstruct some accidents from published flight data as well.
 
I have been using a flight sim for a number of years. I have a G1000 mock up, and honey comb controls. I use Xplane 12 now. For landings and take offs it is absolutely useless. I have a mock up of my T206H. I have "landed" on the island where the Statue of Liberty is, on the George Washington Bridge, on the lawn of the White House, on an aircraft carrier, and numerous other places where landing my 206 in real life would be impossible. I find the sim useful for practicing IFR approaches, and practicing flying by instruments. Otherwise not so much. I have utilized it to reconstruct some accidents from published flight data as well.
I "flew" an earlier version of X-Plane and I was able to keep a Skyhawk in the air well below stall speed. That is probably why you were able to land in all of those places. FSX and P3D stall about the real Skyhawk stall speed, so I doubt you'd be able to land in all those places using either of those simulators.

All I know is I hadn't been inside a Skyhawk for about 2 months, just flying my simulator. I had three landings, each with a different instructor, and did a pretty good job on each. The first CFI was surprised how good my landing was for still being in stage one. The only criticism the second CFI had was I floated because I only used 10 degrees of flaps. Except for letting the plane drift off center line, the third CFI liked the landing. Draw your own conclusions.
 
Last edited:
I "flew" an earlier version of X-Plane and I was able to keep a Skyhawk in the air well below stall speed. That is probably why you were able to land in all of those places. FSX and P3D stall about the real Skyhawk stall speed, so I doubt you'd be able to land in all those places using either of those simulators.

All I know is I hadn't been inside a Skyhawk for about 2 months, just flying my simulator. I had three landings, each with a different instructor, and did a pretty good job on each. The first CFI was surprised how good my landing was for still being in stage one. The only criticism the second CFI had was I floated because I only used 10 degrees of flaps. Except for letting the plane drift off center line, the third CFI liked the landing. Draw your own conclusions.
It is not stall speed. It is handling. The flight characteristics are not even close. I can do all sorts of things in the sim that would tear my 206 apart. I would suggest your landings had a lot more to do with your flying skills than with your use of FSX.
 
lol. I just bought one for my son in December because that was the good yoke. We'll see how it does, I guess.
 
I’ve had an Alpha for several years now and it’s been problem free.
 
@Crashnburn Not related to the honeycomb alpha but to another joystick. Same issue. When trying to fly it was almost like being in very persistent turbulence as it would not trim out. The solution in my case was to adjust the null zone and I think I ended up tweaking the response curve, both of which are possible w/ the alpha.

- on a side note, it is unfortunate what happened to Honeycomb as their products were really nice.
 
I hope you are all sitting down. The new support group for Honeycomb contacted me and helped me verify the yoke is bad, by running the internal calibration procedure.

If I had the original invoice, it looks like I could have gotten it repaired under warranty. Unfortunately, I've already been through all of my late wife's papers (she gave it to me) and didn't find a invoice for it.
I've asked if I can pay for the repairs myself, and if so, give me an estimate.
 
I hope you are all sitting down. The new support group for Honeycomb contacted me and helped me verify the yoke is bad, by running the internal calibration procedure.

If I had the original invoice, it looks like I could have gotten it repaired under warranty. Unfortunately, I've already been through all of my late wife's papers (she gave it to me) and didn't find a invoice for it.
I've asked if I can pay for the repairs myself, and if so, give me an estimate.
Awesome to hear there's still some life in the company.
 
In my years at the flight school I maintained several different flight simulators, from the cheap desktop type to some expensive units certified for IFR training. Only the ancient Frasca electromechanical sim we first had didn't use potentiometers for control inputs; it had other problems instead, but totally manageable. The others all used pots, and they were always troublesome as they got more years and hours on them. Potentiometers use a copper runner on a carbon track to vary a resistance which controls a voltage input to the sim. The carbon track wears and carbon dust forms, and that dust gets under the runner and makes the pot give spurious signals. The center of the travel gets it the worst, as that's where the control is most of the time. When we bought one of the first Redbird sims, I asked the engineer who came to assemble and calibrate it if they had used pots in it. Yup, he said. I asked him why they didn't use an inductive pickup instead; no wear, so no stupid control reactions. An inductive pickup would have two coils on a common tube, with an iron core that slides in and out to vary the induction. One coil has a regulated AC signal on it, and the other receives the induced signal at various strengths depending on the position of the iron core. Millions of arc welders have used this method of varying current for a very long time.

Hall-effect sensors might work, but their travel is really limited and might be too sensitive.

Remember the cars that had problems going to full-throttle suddenly? Some of that was potentiometer problems. Throttle-by-wire. My Hyundai has two potentiometers moving mechanically in parallel, but electrically separated. If the resistance between the two varies by more than 4%, the car goes into a limp mode that limits the power so that the car will only do some slow speed so you can get out of the way, or get to a shop.

Potentiometers belong in antique radios.
 
I'm thinking that if I can get into Tammy's Amazon account that I can find the invoice there. I know her email, but not her password. They sent a one-time code to her email and I'm trying to recover that password.

I've also filed a request to Amazon's bereavement department to get access that way.

I've heard the Turtle Beach yokes use Hall-Effect sensors, so there won't be any wear.
 
Maybe the Honeycomb group can recover fully and finally start churning out the Charley rudder pedals. I really like my Alpha, but there's not a lot of love for my Saitek/Logitech rudder pedals. I wish I had enough to buy the Thrustmaster TPR pedals. Yikes! Nearly $600.
 
Back
Top