FlightAware Discussions

Flight Feeder suddenly issuing overtemp alarms

Hi,

I’ve been a user since April 2019, with the same company-issued Flight Feeder (FlightFeeder 9.0~bpo9+1) in exactly the same location. It’s in a dark steel building; due to the firmness of the coax cable, it typically is hanging in free air. I rarely see the unit, nor does anyone else.

I’m going into the third summer, and suddenly, since the V9 firmware update, the unit is very warm, and regularly reports high temp alarms. Oftentimes internal temps are 167 degrees F. This has never happened before, and the ambient temp around the receiver has been MUCH, MUCH hotter during the past 2 summers. We’ve not even hit 85 degrees yet here.

A request for support languished for a few days, then I received a rather distracted (she thought I had a PiAware) set of common sense ideas to resolve it…keep it out of the sun, etc…and get this: TAKE IT OUT OF THE CASE.

I’ve ordered a baby boxer fan for it…but I don’t think I should need to do that, nor do I think the unit was designed to be operating out of the case. The problems started the very first day over 80 degrees F after I got the v9 firmware update.

I’m thinking this is related to the firmware update…anyone have the same problem? I’m not finding FlightAware’s support really that helpful.

Thanks.

What time of day do you see the over-temperature alarms? What’s the ambient temperature where the FlightFeeder is at that time of day? 85F / 29C is towards the high end for what we’d expect these to run in (& I assume that’s an outside air temperature - indoor temperatures might be higher if there’s not much ventilation where it is?)

I took a quick look right now and there’s nothing obviously wrong, though it’s using more CPU than I would expect (which is likely to be a consequence of some changes in piaware 5 / FF 9 and might explain the heat problem if the environment is marginal). for context, by “more CPU” I mean “about 1/8th of the available CPU” so there should be plenty of headroom, but more CPU use does mean more heat generated.

Hi,

Temp overruns have been consistent with typical day / night cycling…i.e. noon, 1 or 2 PM reports the high temp, and 5 PM to 7 PM it returns to normal.

The environment is a large steel building, 30x90’, 18’ peaked ceiling with a concrete slab throughout. The receiver is about 4’ off the floor, and the temps at that elevation are usually lower than outside ambient due to the thermal mass of the concrete…at least through early August. We had our last frost less than a week ago, and the slab stays very, very cold (e.g. condenses dew on a humid day) due to 6-7 months of 10-20 degrees F ambient.

I’m a broadcast consultant, and often deal with component temps. This unit has been SUPER stable as regards temps for over 2 years…a few days out of what was cold weather it’s showing an alarm it’s never exhibited before. I really think it’s related to the CPU load of the new firmware.

Right now, it’s 68 F in the building, and the receiver has it’s cover off (yes, I took her advice but I certainly don’t plan to leave it that way), and it’s reporting 129 F. That is not a happy component…its truly going to be just a matter of time until it quits…(and RUINS my perfect 785 day run! :slight_smile: )

I dialed back the receive gain one step which may help with the CPU - let’s see how it behaves over the next couple of days.

1 Like

Thanks…I will keep you posted.

Thanks for your help and interest.

FWIW I stopped the receiver entirely for a few minutes (so essentially no CPU load) and the reported temperature only cooled by about 3C (~5F) so I suspect there’s something else going on.

It’s now 67 F ambient in there, and as you can see it’s 124 F (with the cover off and hanging by it’s antenna line).

Do we have a failing receiver…can you arrange a replacement?

Could be failing hardware. Can you tell whether the heat is mostly coming from the main board (the Pi) or the receiver board (the smaller board that has the antenna connector)?

Another thing you could try is to power it off and carefully detach the touchscreen (there is a single multi-pin connector mounting the touchscreen to the main board) - it should run fine without the touchscreen, though obviously without the display, and the touchscreen backlight adds a surprising amount of heat (and we can’t turn it off! hardware limitation…)

If we can’t work out what’s up and the overheating persists then replacements can be coordinated through adsbsupport@flightaware.com (refer them at this thread)

The radio board is significantly warmer than the processor board…actually, several components are too hot to leave my thumb pressed upon.

Ok, that sounds like the radio board might be failing then. They do run warm, but shouldn’t be extremely hot. Support should be able to organize a replacement of that.

The radio board is mostly a black box and doesn’t get modified by software updates so the timing may just be coincidence.

1 Like

Thanks…will remove the display tomorrow (it’s getting late here) and will fire off an email to support tonight. Thanks for your help…I will let you know what I observe tomorrow.

Hi,

I did the email report last night, and now, about 17 hours later, I received an email stating this:

“Request #396780 “Failing unit” was closed and merged into this request. Last comment in request #396780:

I don’t know what this means; are they sending the part?

I put a fan on the unit, and it’s still out of the enclosure.

thanks

Just as a data point, I have a new-ish FlightFeeder 978 that was apparently upgraded to 9.0 over the past weekend. It is in my garage, about 2-3 feet inside the garage door. outside temps were in the low 80s early this week, and I received two overtemp alarm emails.

On June 1, I received the overheating alert email at 5:01pm local time. At 5:04pm, I received a “temperature normal” email (3 minutes later).

On June 2, the overheating email came in at 3:01pm, and the temperature returned to normal at 5:04pm.

I am away from home the rest of the week, but will set up a fan in the vicinity when I return home to see if that makes a difference.

I have only had the unit for about 3 weeks, and during that time we have had several days with temperatures in the 80s. I didn’t get any notifications until this week, after the update.

Means you sent an email which created a second separate ticket; support merged that ticket into the existing one so they have one place to track it.

Any heat changes are unlikely to be related to the 9.0 upgrade as the 978 receiver software did not change significantly in 9.0.

So I took another look at these numbers and they’re not too far off normal (I mostly work in celsius, so I blame that for the brain fade!). The FlightFeeders typically run 35C-ish above ambient (case on), or about 63F. So 68F ambient vs 129F reported would be fairly normal. I’d expect it to be lower with the case off though.

My test unit here is running at 63C with ambient around 28C.

The temperature alarms are at 70C / 158F, so - all things being equal - I’d only expect your caseless FF to reach that if the ambient temperatures exceeded 95F.

As you said, I think it is abnormal for it to run that high…. a especially since it’s never turned in this alarm in 2 summers of operation.

I also don’t think it’s a good idea to run the unit caseless in the environment it is in.

Other than your (thank you!!) excellent help, I’m not getting a lot of constructive response from customer service. They seem a little distracted. Suggestions?