Kingtech vs FrSky (Tandum radios)

[Konrad]

This might not be the best place to place  this notice as I see it has been dead here for close to a year. But here goes.

 

Earlier this month I spent 2 weeks in Denver helping my good friend Matt acquire his AMA turbine waiver. I was acting as his spotter and engineer.

 

It went rather easily but for one nagging problem. The new Kingtech turbine a K102 kept having start issues with premature shut down. This  traced to the lack of proper heat rise recorded by the thermo coupler.  We noticed that prior to start up the EGT was bouncing around 0°C to 125°C and this was with the ambient temp at around 22°C.  We changed out the turbine and after a few flight notice that this low time turbine was also exhibiting start problems. Again we say the EGT bouncing wildly prior to any start attempts.  We borrowed a third low time turbine (I think a K85) and again noticed the bouncing EGT. 

 

This lead me to think that Kingtech had a problem with their vendor for the EGT thermocouples. We had sent the first K102 back as soon as we removed it. Matt was able to get in the required time and perform the demonstration/qualification flight with the borrowed turbine. You can see some of it here. (He just had to show off with a touch and go to show mastery of the required energy management).

https://www.youtube.com/watch?v=-dDSsrVQKiU&t=169s

 

Well, Kingtech just got back to us with their findings of no faults found. As a safety guy this always bothers me! But they did suggest where the problem might lie. They asked what radio we were using. I was thinking what does this have to do with anything?  Well, they suggest with the dual band FrSky radio we were using that if we had the 900 mHz power output raised that is was very likely that the thermocouples was picking up RF noise resulting in the ECU thinking this was temp data and putting out erroneous data. 

 

Well Matt and I had set the 900mHz power level at 200 mW and 100mW respectively.  I just did an engine start with the 900 mHz band's power level set at 25 mW and the EGT on the borrowed turbine was stable prior to the start. As a result we have had constant starts with the borrowed turbine.

 

I look forward to seeing if this was in fact the issue with the turbine we sent back. 

 

So if you are having inconsistent start issue with your turbine check the RF power output of your TX. I suspect that there is a weakness in the RF shielding allowing the ECU to see too much RF noice.

Edited October 29 by Konrad

[Martin Harris - Moderator]

Hopefully not something that will bother the majority of our forum members as the majority are UK based where the use of 900 MHz (actually 863 - 870) is limited to 25mW ERP!

[Konrad]

That might be so. But unfortunately I still saw some  instability in the EGT readings even with the 900mHz  power setting as low as 25mW.  At the low 900mhz power setting we got 5 starts  in a row without issues. Setting the radio to being only a 2.4gHz radio did result in a stable EGT reading close to that of the ambient air. 

 

What I wanted to bring up was that the long leads act as antennas and pick up RF noise that the ECU can and does interpret as data. Unfortunately Matt has a very clean installation with wiring looms making nice straight runs.  I'm recommending that we put some twist and loops in the loom and maybe add some ferrite rings. I will be trying to add some shielding all along the run, but this will take a lot of work.

 

The take a way is that what might look like a hardware issue really might be an installation issue. So if someone is seeing some instability in the starting, and maybe even the running (sudden shut downs on the ground). Take a look at the installation, keeping good RF practices in mind.  Just moving the TX away a few feet during start up has helped. Remember that radiation dissipates at a rate of 1/R^2. This might help and keep one from having to send the turbine back for service, loosing the ability to use the turbine for a few weeks and not solving the problem!

Edited October 30 by Konrad

[Konrad]

A bit of follow up. 

 

We haven't as of yet received the original turbine back. But we have spent some time chasing this RF noise.

With the use of some sound RF practices, we now have the signals to and from the ECU matching the same level of bounce (noise) with the power level set at 200mW as we did with the original set up at 10mW. The use of ferrite rings has helped.

 

I'm still not happy with this as the data still is bouncing. I'm not looking forward to running full length wire shielding. 

 

I think I have talked Matt into trying 900mHz @100mW for the next flights. 

 

Turning off the 900mHz band results in clean stable data.

 

The concern with the bouncing data (EGT) is that the ECU has been reading this erroneous data and initiating a premature shut down. This can and has lead to tail pipe fires. Once the turbine is at speed this hasn't been a problem with the 3 turbines we have been using. 

[Konrad]

This is to close the loop.

 

We got back the original K102 turbine and put another 3 flight on it yesterday. The turbine and the MB-339 model performed perfectly. 

I got Matt to cut down on the power level for 900 mHz. He flew on 25mW and 100mW with no start issues. I don't know why Matt did this, but he also did a test at 100mW without the ferrite rings and again no issues with the original turbine during start up. 

 

It was noted that the EGT was still bouncing any time we were transmitting on 900mHz. Just that at the lower power levels this never triggered a shut down. You might recall that at 200mW Matt was seeing problems with the engine start. Again it was noted that with only 2.4gHz transmitting that there was no noticeable bouncing of the EGT prior to engine starts. This is indicating to me that there is some issue with the RF shielding. 

 

For a host of reasons we will not fly single band 2.4gHz other than maybe some low level emotional models like UMX foam. But even then we have concerns about safety that we don't do it if we can switch out the RX to a dual band type. All flight tests with the turbine have been dual band 900mHz/2.4 gHz.