Previously I'd fitted an Icom IC-2800 into my 56 plate Toyota Yaris. This has now been replaced with an Icom IC-E2820 with D-STAR (review of the transceiver). Much of the fitting was the same as for the IC-2800 (obviously I left the microphone extender and antenna in place as the connections are the same).
The IC-E2820 comes with a remote head that can be attached, if required, to the main unit with 2 supplied magnets. For this instal I decided to stick with the layout used for the 2800, the main unit in the spare wheel well with the head in the roof.
This gave me a few problems:
I did not want to "cut and splice" as I did with the 2800 head lead so I looked at the alternative of making my own lead, this has an RJ11 type (6 pin) modular plug at the main unit end and a 4 pin modular plug at the head end. 6 wires are connected in the RJ11 but only the middle 4 make it to the head (what happens with the other 2 doesn't matter, according to the schematic the socket outer pins are not connected to anything in the main unit!). I have plenty of Cat5E cable (8 conductor) along with with a quantity of 8 and 6 pin plugs in my "junk box" and a pair of crimping pliers that handle the 8 and 6 pin modular plugs.
After a discussion at the club with another knowledgeable fellow (he knows who he is) where I mentioned the use of the 4 unused wires in the Cat5E cable to carry the required Data and the audio for the speaker (thus having to run only 1 cable instead of 3) and we talked about the merits of cross-talk between the data for the head, Data for communication and the relatively high level analogue signals for the audio (I thought that the audio for the 2800 was at the amplified level in that head cable), I returned home to check out schematics. Indeed the 2800 had the audio amp in the main unit and fed the amplified audio to the speaker in the head. The 2820 had the ground for the head, Data and audio all tied to the chasis ground so, in theory, I needed: 1 audio, 1 Data in, 1 Data out, 1 head data in, 1 head data out, 1 12V (for the head) and 1 ground for everything = 7 wires!
I decided that I would make a lead up and try it in the shack before I actually fitted it to the car, just to be sure that it all worked. I measured out enough Cat5E cable to go from the main unit, through the car, up the "A" pillar and across to the head (with a bit extra, better 2 foot too long than 1 inch short!). I then marked where the speaker and Data leads would "break out" of the lead (they didn't need to go to the roof, only as far as the dashboard) and removed a portion of the outer sheath at this point. I cut the wires that I was going to use for the speaker and Data and eased them out of the casing.
Back in the shack I measured and cut back the outer sheath for a long enough flying lead to go to the rear panel of the main unit (for the speaker) and decided to use the Brown/White Brown pair for the speaker, with the Brown as ground. 1 3.5mm mono jack plug and a couple of minutes with the soldering iron and that was done.
Next was the turn of the head data cable. I slid a short piece of the outer sheath over the 2 pairs I had earmarked for the head and crimped on an RJ11 plug. At the other end of the cable I prepared the cable and crimped on another RJ11 (I know, ist's not the right one, I'll come to that!) I checked that the connections were good and properly placed with a multimeter. All was good so on with the next step.
I got my Dremel hobby drill out, fitted it with a sanding drum and thinned down the sides of the RJ11 that was to be fitted into the head. I used a flat file to finish it and to slightly thin down the locking tongue, as the 4 pin modular tongue is slightly thinner than the 6. This was done carefully with equal strokes of the file on each side until it slid into the socket on the head. Finally, again with the file, I chamfered the leading edge of the plug as, if you look closely at a modular plug, you will see a small step in the front and where the plug had been thinned there was no step, this stops the plug from locating properly. I plugged in both ends and switched on the radio. All was working fine so I went to the next step.
Those that have carefully counted will note that I have now used 6 of the 8 wires in the cable and have 2 left for the Data in, Data out and Data ground! Well here goes: I took the final pair of wires and trimmed them so that they would be the right length to reach the 2.5mm Data/Clone socket on the front of the radio. I then cut a piece of co-axial audio cable and took out the dilectric and inner conductor. I exposed some of the screen at one end and slid the coax over the 2 wires with the exposed screen at the plug end and connected the 2.5mm jack plug to that assembly. I then exposed some of the screen at the other end and removed some of the insulation from the Brown wire. I soldered the screen to the Brown wire and wrapped the whole connection with electricians tape after checking that all was connected properly with my multimeter.
At the "break out" point I soldered the leads for the Data connection and the speaker to their proper wires (both grounds being common) and checked again with the multimeter. Then it was the turn of the DB9F for the connection to the PC serial port, checking with the "How to make your own OPC-1529R" page and again with the multimeter to ensure no shorts, opens, cross-overs etc. (better to check 5 times than destroy a vital component in the radio, cheaper too!) and wrapped the individual wires and the the whole "break out" point with more electricians tape and fitted the DB9F with a cover.
Back to the radio and I plugged everything into the main unit and the head onto its connector. I switched on the radio and the sound issued from the speaker. Into the back of the PC with the Data lead and I set the radio to CLONE OUT and received the data at the PC. A few changes with the PC software and I sent the data back to the radio. Data in and out were fine, speaker audio fine, control from the head fine. I used the radio with this lead connected for a few days to ensure that all was well before fitting to the car. I'd just resolved all 3 of the problems listed above with just 1 cable!
In the car I checked the head for fit where the 2800 head was. The hole in the roof lining and the inner steel where the map light originally sat was about 5mm wider than the 2800 head but, oh dear! The 2820 head is about 10mm wider than the 2800 head so the hole is too narrow! I was determined to fit the radio head here so out came the Dremel again, this time with a metal cutting disc instead of the sanding drum. A few minutes later I had a hole in the steel that fitted the 2820 head with a little room to manoeuvre, I left the roof lining as it was because it was easy to ease it out of the way and it will look better when I refit the map light when I get rid of the car (I'll worry about how to clip the light to metal that's no longer there when the time comes!).
As for the actual fitting of the IC-E2820 it was simple after the above had been done. I found a screw in my "junk box" that fitted the threaded hole in the head mounting plate and used the bracket that I had used for the 2800 head (which, in turn had come from the same "junk box", it was a bracket that was originally supplied with a CB radio snatch plate, flattened out then bent to the required shape).
The plastic trim panels were removed from the car (see the bottom of the original fitting page for details) and the 2800 head cable removed, replaced with the homebrew 2820 cable, routing the Data and speaker cables across behind the dashboard to the centre console.
In the boot the 2800 main unit was disconnected and removed, the power, antenna and microphone extension cables all being re-used meant that this install went quite quickly. The 2820 unit fitted in where the 2800 came out from, all leads connected, power applied and the radio was up and running. The only thing left to do was connect the GPS antenna and find somewhere (inside the car, the antenna is not waterproof) for it. I decided on the hatch back metalwork at the top of the door (hanging upside down when door closed), the base of the antenna is magnetic and (1 week on, as I write,) is still where I put it. The cable for this was left loose across the opening (for the door to be opened) then tucked behind the rubber door seal and dropped behind the plastic trim when it got to the bottom edge of the door frame, then plugged into the main unit. The GPS function was checked and working.
At the moment the speaker is awaiting a bracket to affix it to the centre console trim, I have not yet decided exactly where it will go (it's the second speaker I've fitted as I didn't like the sound of the first one). I'm still using the HM-98 from the 2800 as the HM-133 is slightly wider and the upper glove box lid clips it on opening and closing when the mic is on the hanger. Also the HM-98 has a cover over most of the (little used in normal use) buttons so, at night, there is less light in the car to distract the driver.
All the trims replaced, radio working and first contact was 2 hours from starting, if you factor in the cable and speaker bracket manufacture as well make it 3½ - 4 hours.
On testing there was no more noise on the 2820 than there was on the 2800 and even at 50watts on VHF and UHF there were no adverse affects to the vehicle (with the exception of the VHF FM broadcast band interference when transmitting on 2M). Data is able to be sent and received by the laptop PC connected to the Data DB9F plug that normally resides in a cubby hole in the dash when not in use. I have not tried packet (in fact I've not used packet at all to date).