Europa #435 G-RODO Build Journal - 2021 03

index

sitemap

advanced

search engine by freefind

go to list of milestones

Navigation & Acknowledgements

day	notes	hours
1	Abrade the bonding area under the edge of the windscreen surround for the TA50 GPS antenna mount, and the mount itself. Degrease both bonding surfaces with acetone. Mix a batch of Evo-Stik Control 2-hour epoxy and thicken it with s small amount of flox to non-slump state. Apply to TA50 mount and press into position. Clamp in place, with a piece of polyethylene sheet over the clamp to stop it sticking. Return to the comm radio antenna tuning and just out of interest, check how far apart on each side of the resonant frequency (119MHz, SWR 1.11) are the SWRs of 1.5 (116MHz & 128MHz; 3MHz below, 9MHz above, 12MHz wide), 1.8 (110.5MHz & 132.5MHz; 8.5MHz below, 13.5MHz above, 22MHz wide) & 2.0 (108MHz & 134.5MHz; 11MHz below, 15.5MHz above, 26.5MHz wide). As the aeronautical band is under 20MHz wide, it should be possible to get a SWR of under 1.8 over the whole band, but the shape of the curve suggests that the optimum frequency for the resonance should be lower than the commonly-advised midband of 127.35MHz. We'll see if things look any different when the antenna is out in "free space". I also note that the ATR833 manual calls for a SWR of lower than 3, so we're already compliant with that even before any antenna tuning!	3503.0
3	Power up the panel and turn on the ATR833 radio. Plug in a headset on P2 side. Intercom squelch has a slight delay before sidetone is heard, but also has a delay before cutting off. I can hear carrier wave and brief speech transmissions from the handheld radio. Plug in speaker to socket on F14 instrument module and that works OK too. Press P2 PTT and it transmits a loud low-frequency tone, also audible in headphones. Take the connectors out of the P1 stick base in case a short there is causing a problem. Connect those flying leads to the stick-top switches and try again. Same tone still there on transmit, from either PTT switch. Referring to the Funkwerk manual, go through the configuration steps and all seems fine there. Notice that the tone on transmit is not audible on the speaker, only on the headset (which makes sense, as having the speaker live during transmit could cause feedback). Pausing for thought, check the fuse overlay transparency and note that the C fuseblock is slightly offset from the central position shown on the drawing. Edit the CADintosh file to correct that, add a "no access during flight" legend on the white layer, and print out on transparency film. Disconnect and open up the hood of the ATR833 D-sub connector. Check with magnifying glass for solder-splash shorts and bad terminations. All appears OK, and all pins that should be NC are so. In the connectors drawer, find a male-female pair of 25-way D-sub connectors joined on all ways with a short length of ribbon cable. Cut the ribbon cable in the middle.	3505.3
4	Find a stock of test-lead wire in black and red that is heavier than the cable now in use for connecting the bench PSU to the aircraft battery leads. Make up new leads and connect them. Try the ATR833 transmit again just in case the problem was voltage sag, but still the tone is generated on transmit. Strip the ribbon cable wires on positions 13 & 25 (earth) of the D-sub socket I found yesterday. Solder those wires together. Similarly connect positions 12 & 11 (+12V). Also strip the wires on positions 15 (PTT L) and 17 (PTT R). Fit the connector to the ATR833 and connect the croc clips from the PSU to it. Turn on the ATR833 and touch wire 15 to wire 13 & 25 while monitoring with the handheld. This time, just plain carrier wave transmitted, so the fault is not in the radio, but somewhere in my wiring. Strip the corresponding wires on the ribbon cable to the D-sub plug and connect them to the ones on the D-sub socket. Fit the assembly between the ATR833 and the socket on the aircraft harness. Power up the panel and now both P1 and P2 PTTs give a plain CW transmission with no problem. Connect position 3 wires (headphone ground) and re-test - OK. Connect position 2 wires (headphone live) and re-test - OK, with headphones working. Connect position 6 wires (mic L ground) and re-test - OK. Connect position 19 wires (mic L std) and re-test - OK, with P1 mic working but no intercom yet. Connect position 7 wires (intercom) and re-test - OK, with intercom working. Connect position 14 wires (mic R ground) and re-test - OK. Connect position 18 wires (mic R std) and re-test - OK, with P2 mic working. Connect position 8 wires (mic L dyn) and re-test - the fault is now back! Check the Funkwerk manual again. It says that if the L dynamic mic is not being used, pin 8 should be connected to pin 6 (mic L ground). I have connected pin 8 to the common earth point, which is effectively the same thing, but via a long (unshielded) length of wire. I will have to check whether shorting pin 8 to pin 6 within the connector hood solves the problem.	3507.3
5	Cut wires 423 and 424 just outside the rubber boot of the ATR833 D-sub connector. Remove the other ends of the wires from the common earth point (which I can do by just pulling hard on the wires, showing the strength and integrity of the crimps on the blade connectors!). Power up and test again without the test harness interposed - all OK. Solder the shortened wires 423 & 424 onto the earth pad on the PCB between the D-sub connector solder buckets. Test again and there is now a much higher-pitched (but not so loud) tone on transmit. As there is no problem with the dynamic mic inputs unconnected, I think that is how I should leave it, so remove the wires from positions 5 & 8 on the connector. Test again and all is well again now. Power down.	3508.1
6	Connect Tascam digital recorder to line-out socket and power-up panel and radio. Nothing heard on recorder from intercom - but of course I now know that the speaker is muted except for incoming radio signals, and as the line-out circuit is fed from the speaker circuit it will behave the same. A test transmission from the handheld records OK, after turning down the line level to about a quarter. Next turn on the SmartASS and hope to hear its opening announcement in the headphones, but there is nothing. After some fruitless further checking including applying test tones to the radio's NF input, check the external input volume setting. It is 0 so that explains the silence. I thought I had previously checked that it was set to something sensible, but apparently not - should have double-checked rather than relying on memory! Set to 10 and try again - SmartASS startup announcement comes through very loud! Change setting to 7 and with the SmartASS volume set to about a quarter the announcement level seems to match the intercom level.	3509.6
7	Climb in to port seat, power up the panel, radio & TRX-2000. Going into the settings and adjusting the audio out level on the TRX-2000 gives a very short but quite loud ping in the headphones. Good to know that that aspect of the audio input is also working. For full recording of all radio traffic, I will see if I can fit a 3.5mm jack into the headset box, in parallel with the headphones. Also set the ICAO address of the aircraft in the TRX-2000.
8	Wrap the wire bundle within the ATR833 D-sub connector with self-amalgamating tape to fatten it enough for the cable clamp to get a good grip. The clamp was barely gripping before, and now that 2 wires have been removed from the bundle it needs the extra diameter. Re-assemble the clamp & hood. Attach the locking ears and fit the connector to the ATR833. Tidy up various wire bundles with cable ties.	3510.7
9	Move car, remove wiring paperwork from top of fuselage, tidy garage floor a bit to clear tailwheel route and roll fuselage out onto drive. Park with nose towards the garage. Make a camera hood for the NanoVNA from card and black velvet. Set the NanoVNA sweep to 115MHz-140MHz and do full calibration with a rigid SMA lead in place on CH0. Measure SWR with the antenna as-is. Take pictures of all significant readings. SWR 1.37 at 118MHz, 1.06 at 121MHz, 2.84 at 138MHz. Start trimming a few mm off both ends of the antenna, checking bottom of band, top of band & lowest SWR after each pair of cuts. Lowest SWR frequency moves in sequence to 121.75MHz, 122.75MHz, 123.5MHz, 124.25MHz and 125MHz. The latest trim takes the length of the dipoles to 488mm each, with SWR 2.10 at 118MHz, 1.00 at 125.5MHz and 2.34 at 138MHz. Decide to do one final trim of about 2mm, taking the dipoles to 486mm each and giving SWR of 2.23 at 118MHz, 1.01 at 126.25MHz and 2.26 at 138MHz - very satisfactory. Connect the antenna to the ATR833, lift the panel up to the vertical and power up. Can't hear anything on the Gloster ATIS frequency, but on the approach frequency I can clearly hear aircraft transmissions but (unsurprisingly) not ground transmissions. Leave the radio on with the speaker connected. Airbox, Dynon & TRX-2000 all now have a GPS fix. Dynon shows 276ft and Airbox 205ft - h'mmm. Also, Airbox is occasionally saying "no GPS". Lift the tail and set the tailwheel on a small stand to get the fuselage fore-and-aft level (to within 0.3 degree - also level transverse to the same tolerance). Turn on SmartASS and press button until 3 beeps heard, to set orientation. Verbal confirmation heard. From an aircraft transmission I hear that Gloster Approach is offering QNH 1018. Set the altimeter subscale to that and it shows 260ft. Dynon now also showing 260ft, but Airbox is showing 203ft and shortly also shows "no GPS" and stays like that. I think it will have to go back to the manufacturer for attention. Tidy up a lot of the wire offcuts from the garage floor, and put away most of the tools. Roll fuselage back into garage.	3514.4
10	Send e-mail to Airbox re "NO GPS" warning.
11	Trace outline of aluminium fuse cover plate onto the protective film of one of the 2mm polycarbonate sheets. Cut around the marked outline with a fine-tooth hacksaw. Trim down to the lines with a coarse rasp and smooth with a Perma-Grit block. Similarly mark, cut and file the second sheet. Reply received from Airbox; the Aware 5 is no longer supported. They suggest re-positioning it for a better view of the sky. Failing that, they will take it back in exchange for a Runway HD subscription (which needs an iPhone or iPad to run on). Need to think about the choices.	3515.3
12	Thinking about the fuel system, and filling it, order and collect a couple of jerry cans from Halfords, and an air-pressure jerry can pump online. Start to ease the radio panel into position on the F14 instrument module, but the temporary support shelf is obstructing it at the bottom. Remove the shelf and the support line on the radio panel. Move the radio panel into place, checking as it goes that there are no wires trapped or strained. Fit the 5-off button-head stainless fixing screws (1/8" hex key). Clamp the shaped polycarbonate sheets together with the aluminium cover plate and drill through the fixing holes 3/16". Unclamp and open up the holes in the plastic sheets to 7/32" for screw clearance. Line up the fuse identification overlay transparency with the aperture and mark for trimming. Trim roughly and re-check. Align between the polycarbonate sheets and trim to the edges with a razor blade. Interpose an A4 sheet of the drafting film and trim it to shape. Try punching the holes in the transparency but not very successful and the leather punch isn't really sharp enough for the thin film. Hold everything in alignment and drill instead.	3516.6
13	Power up the panel and start to configure the MGL instruments. The E-3 does not respond to any presses of buttons or knob, nor to turning the knob. Leave that for the moment and work on the TP-3. Configure for oil pressure on channel 1, oil temperature on channel 2, and water temperature on channel 3. On exit from configuration the values shown are not sensible; realise I've got the pressure & temperature sensors the wrong way round. Change to oil temperature on channel 1 and oil pressure on channel 2, which gives a more sensible display. Oil temperature shows 9C and water temperature 10C which seems close enough as both sensors should be at ambient. Oil pressure shows as 4.5 bar which is surprising as it should be zero. Will need to check if I've got the sensor type wrong. Returning to the E-3, it is still not responding to any front panel interaction, although just once I got the contrast adjustment menu item, but could not adjust it. Powering off and and again gives no change in behaviour. Make a video of the startup screen as there is a strange number (13108) appearing along with the alternating flight time & volts displays. The firmware is v1.09 which according the the MGL website documentation is the latest for this device.	3517.0
14	Send e-mail to MGL about the E-3 problem. Try powering up while holding down both buttons on the MGL E-3, which does initiate the rest to factory defaults. After that I can enter the menu system and do configuration. After exiting the menu, the CHT (after settling) shows 12C, which matches the TP-3 display of oil & water temperature. However, after turning the power on & off again the E-3 is again unresponsive to button presses. Send a follow-up e-mail to MGL reporting all that behaviour.
18	Nothing heard from MGL so send a follow-up e-mail. Reply received suggesting that I try re-setting it to factory defaults again and then see if it continues to respond to button presses after powering down and up again. Power up the panel and check if the E-3 responds to buttons or knob - it does not. Power down, then power up with both E-3 buttons pressed. The "loading default settings" display shows. After it boots up, it responds as normal to presses of any button or knob, and to knob turning. Try cycling the power again several times, and it continues to behave normally. Report results to MGL, asking if I could have somehow disabled the buttons by my configuration settings. Turn attention to unexpected oil pressure shown on TP-3 display. Although engine not running and not even any oil in the system, it shows 4.6 bar. Disconnect lead from sensor and display shows 99.9 bar. Short lead to engine case and display shows 0 bar. Check the sensor with an ohm-meter and it registers about 100 . Refer to the Rotax installation manual and the graph in section 21.3 shows 100 corresponds pretty closely to 4.6 bar, so the TP-3 is working correctly. Consult the Rotax documentation. The resistance-type sensor on my engine is no longer fitted or supplied by Rotax - the new one is a current loop type with a different thread. Send an e-mail to Kevin Dilks of Special Aviation Services asking for advice on the oil pressure sensor. He replies saying that he stocks my type of sensor and suggesting that I take the sensor off and try tapping it on the threaded end to see if that frees it up.	3517.5
19	Unscrew oil pressure sensor (17mm AF spanner). Check resistance - still about 100 . Tap the NPT threaded end gently with the smooth side of a spanner and resistance goes down to about 60 . Tap again a few times and it gets down to about 10 , but that seems to be about as far as it will go. Start the air compressor and fit a rubber nosepiece on the blow gun. Connect the ohm-meter to the sensor with croc-clips and apply the blow gun nozzle to the small orifice on the threaded end of the sensor. With the regulator as normally set (about 6.5 bar) the resistance goes to about 129 , and on releasing the pressure it drops immediately to 10 again. After adjusting the regulator to max (about 8 bar) and applying the air gun again the resistance goes to about 148 and again drops to 10 on release. Those figures are in line with the graph in the Rotax manual. With no pressure applied the sensor is now consistently dropping to the 10 minimum resistance shown on the graph, so I think I will be happy to re-install it.	3517.8
20	Re-fit oil pressure sensor with Loctite 577 on threads, ensuring no Loctite on the working end. Re-connect wire 337. Power up panel and TP-3 now shows 0 bar for oil pressure. Change TP-3 ch 3 warning to 120 C. Check that E-3 responds OK to buttons and knob - no problem. Power down, then up again with E-3 buttons held down to ensure factory defaults restored. Go through the configuration menus, setting as follows. Flight log: auto detect. Hobbs RPM: 2000, Maintenance time: 25 hours, Hour fraction: minutes. EGT Channels: 0, Alarm: off. CHT Channels: 2, Mode: Highest, Span: 150 C, Topscale: on, Setpoint: 130, Alarm: on, Probe: NTC, Unit: C. RPM Scale: 6000, Alarm: on, Alarm: 5810, Pulse/rev: 1.0, Pulse: time, Filter: scale. Pressure Display: off. Temperature Display: off. Voltage: no change to defaults. Exit menu and check button & knob behaviour - all OK. Cycle power off and on again. Buttons and knob now seem to be unresponsive, except that occasionally the menu system comes up, but apparently not directly as a result of pressing the knob. When it does show, turning the knob has no immediate effect; the menu changes some time after, but again not predictably. Cycle power a few times with no improvement. Had turned on several other items including the comm radio while doing the E-3 setup and pleased to note that even inside the garage it was picking up airborne transmissions from 10 miles north. Send e-mail to MGL reporting on today's operations.	3518.9
23	Reply from MGL suggesting diagnostic approach - try setting one item at a time and checking if that causes the problem, to help them figure out where in the code the problem might lie.
24	Power up the panel while holding down the buttons on the MGL E-3 to reset to factory defaults. Enter the configuration menu and do one selection at a time, then exit the menu system, cycle the power and check if it's still responding to the knob and buttons. Flight log: auto detect; 2000 RPM. All OK. Hobbs: RPM 2000. All OK. EGT: Channels 2; Alarm off. All OK. Tried to set up all the CHT stuff but after doing that I got an "invalid engine selection" message; have to select pressure & temperature to OFF before I can do the CHT stuff. Start again. RPM: Alarm 5810; pulse/rev 1.0; Pulse time. All OK. Pressure display: OFF. All OK. Temperature display: OFF. All OK. CHT: Span 150; Setpoint 130. All OK. CHT: Probe NTC. Knob and buttons stopped working after power cycled. Send message to MGL with results.	3519.4
25	Rest the MGL E-3 to factory defaults. Set the CHT probe to NTC. On exit from menu, shows "invalid engine selection". Set temperature and pressure displays to OFF and now there is no complaint on menu exit. Buttons & knob still working OK. Power off and on again; buttons & knob still working OK. Power down. Send message to MGL with this latest data point.	3519.6
26	Reply received from MGL saying they will sort it out.
30	Investigate MGL E-3 behaviour a bit more. Reset to factory defaults and then set temp & press OFF, and CHT 2 channels, sensor NTC. Exiting menu. All buttons working. Cycle power off and on, buttons still OK. Now change EGT from default 2 channels to 1, exit menu, cycle power, buttons still OK. Now change EGT from 1 channel to 0, exit menu, cycle power, buttons now unresponsive. Try it all again from default settings and and behaviour repeatable. Send e-mail to MGL with latest diagnostics, adding info about inconsistencies within the E-3 manual. I have been setting things up based on Section 3.2 (Special Rotax 912/914 display mode) which says "The number of EGT probes can be selected from 0 to 4." However the table in Section 5 (Engine Configurations) shows there must be a minimum of 1 EGT probes selected; it offers 1 or 2 CHT channels combined with 1, 2, 3 or 4 EGT channels but not 0. That makes no sense if no EGT sensor is fitted!	3520.1
31	Using the usual toothed cutter on the Dremel tool, cut slots, adjacent to the inner faces of the layups, around the edges of the compass mounting arm. Remove the remaining central ridge with a Stanley knife and clean it up with a Perma-Grit spar slotter, leaving a full-width channel all the way around ready for filling with flox.	3520.5

go to previous page of journal	go to list of narrative pages	go to next page of journal
	Return to Rowland's home page

This page last updated 2022-10-10 08:56. I try to make this page as accessible as possible, by adhering to HTML 4 standards.
I welcome comments on this website. However, because of the amount of spam it attracts, I no longer post a direct e-mail address on any page. Instead, please click here to contact me. You will have to confirm that you are human before the message will be sent on to me.