What to do about radio tray rivets

[bradp]

So I have an IA on the field who is willing to supervise me doing a 330ES for an ads-B upgrade.  Now comes the planning phase... 

The old radio tray for the 76A - looks like has  rivets of some sort.  The tray seems to be fastened to the next tray. 

Anyone have tips about how best to drill these out - I have an old dremel tool with a flexible neck that probably won't fit a properly sized bit.  I've seen flexible necks elsewhere.  

Should I replace with same type of squeeze rivet? A rivet nut into the next tray seems like a good idea (nothing seems to be fastened to anything structural in the panel in the vertical plane) followed by a countersunk flat screw. 

I think there a couple of folks on here that have changed trays. Most of the guidance I've found on the internet seems to be for the experimental guys building from scratch.  

[jetdriven]

And angle drill with a #30 will remove them. Squeeze some AN426-4-2A rivets to replace.  Those are pretty soft 

[N201MKTurbo]

You will have to remove that whole rack. To remove the whole rack you will need to remove all the radios from the trays. then remove the connectors from the back of the trays. The dimmer is on the top of that rack and it unplugs. Remove the three screws at the top of the rack and the three countersunk screws at the bottom. the whole thing will come out.

Once you have the rack out you can drill out the rivets from the outside. The rivets need to be flush on the outside. There are a couple of things you can do on the inside. You can do a double flush rivet which is hard to do unless you have a reverse countersink which is rare. You can take the bit out of your micro countersink and spin it with your fingers to put as much countersink as you have patience for on the inside or just drive the shop head as flat as you can get it. Then test fit the radio and file the shop heads down until the radio fits. I found a 45 deg rotary file on a Dremel works best for trimming down the shop heads.

[jaylw314]

If it's just a small project, I'd recommend using CherryMax rivets.  These require only a standard pull rivet gun, and are structural equivalents to the AN426 rivets.  They're more expensive, but the convenience is worth the extra $10-15. The only downside is the smallest diameter is 1/8" (AN426-4 equivalent), so no 3/32" rivets.

To drill rivets, your A&P should have a right angle extended drill or a drill extension like this

http://www.yardstore.com/90-deg-drill-attachment.html

Edited September 13, 2017 by jaylw314

[N201MKTurbo]

That won't work. There are no short cuts here. While you may get away with Cherry rivets on the right side, on the left you will be drilling into the radios on the left rack. And if you do it you will make it almost impossible to remove them properly in the future.

Removing the radio racks properly isn't that hard.

[jaylw314]

Ah, didn't realize the rivets need to be flush on the outside too

[RobertGary1]

Also CherryMax are almost impossible to remove if you plan to change later. You have to drill through the steal core.

 

-Robert

[jaylw314]

1 hour ago, RobertGary1 said:

Also CherryMax are almost impossible to remove if you plan to change later. You have to drill through the steal core.

 

-Robert

Harder, yes, but I find them strangely satisfying to drill out

[thinwing]

The job is a bitch..even with a bench mounted rivet squeezer..both rivet heads need to be flush from both sides..yep the entire tray has to come out...good luck..k

[LANCECASPER]

1 hour ago, thinwing said:

The job is a bitch..even with a bench mounted rivet squeezer..both rivet heads need to be flush from both sides..yep the entire tray has to come out...good luck..k

Which is another really good reason to plan out your avionics upgrades well so you don't have to remove the whole stack too many times - a lot of labor involved.

[bradp]

So in a cascade of previous crappy installations I noted the following:

The right radio stack would need to come out to do this properly as Turbo stated. You need access to both sides of the rack and the surrounding "box". 

Unfortunately, in a series of previous installations, previous installers have riveted or bolted things in a domino pattern to whatever was next. To take down the entire panel would require me somehow figuring out how or why in the world the installer who did my engine monitor put the hex nut that releases the unit in a pretty much totally inaccessible orientation.  My favorite was the screw for an air gizmos dock that was literally screwed into the side of an old KX-170B and it's associated rack.

Mind boggling and lazy, but that's beside the point.

The offending air gizmos dock was removed and the circuit breaker panel was pulled down, giving access to both sides of the rack.  From there it was easy to drill the flush rivets with an extension.

I hate drilling cherrymax. It's like a bad joke.  They wont work because both sides of the rivet need to be relatively flush because they but up against the next thing in the stack.  Rivnuts would result in the same.

Best practice is probably to hand squeeze a standard countersunk rivet - issue is now that the stack holding the tray has been previous worked and drilled poorly, so I have two little figure eights to deal with.  It's either going to require squeezing a bigger rivet than I want or drilling and countersinking new holes in the tray and the rack.  Egads.

Now I see why these things take so darn much time and where the "while were in there" mantra comes from.

Also boo to Mooney for making this design not very user friendly. 

 

 

 

[kerry]

My dental drill would easily fit in there.  Ask your dentist if you could borrow his electric lab handpiece with a E-attachment contra angle latch with a #4 0r #6 round bur.  Or you could buy a cheap unit off of ebay for under $100.  You might be surprised how often you'll use the drill.  Much better than a dremel.

http://www.ebay.com/itm/New-Dental-Lab-MARATHON-35k-Rpm-Handpiece-Electric-Micro-motor-10-Drills-Burs-/201527992673?epid=1780369240&hash=item2eec012161:g:eUEAAOSwwKtZjWua

[thinwing]

Drilling out the rivets while in place is not the problem...it's getting the rivet squeezer in place ...I had to use a bench mounted hydraulic squeezer...to get flattened out enough for a radio to slide past...took multiple attempts...

[N201MKTurbo]

Just use a rivet gun and a bucking bar. You can hammer the shop heads down to nothing.

[bradp]

I figured it would be easiest to do the wiring while I had the tray out to get more access for my paws.   Unfortunately once I got in there this is what I found:

 

Literally a rats nest of crossed wires.  From what I can tell most of the jumble was introduced by a single west coast avionics shop that the previous owner used.  Harnesses crossing though the backs another etc.  I'm at a bit of a crossroads to figure out how much digging and reorganizing I want to do.  For the most part I'm thinking that if it's functional and in can keep my work separate and better organized than leave it alone.  I don't want to render myself AOG because I get into something that I don't understand. 

Here are some splices I made to go with each of the KT-76As leads.  There's one hanging down vertically in the picture above as a test to make sure it is feasible / practical to use solder splices.  I needed some helping hands and a very small butane torch to get in the space, but the splices are quite neat and sturdy. 

Below are the rest of them to go on:

 

And here's the labeled  harness I made to go between the GTX330 and and 430W.  

 

One of the beechtalk avionics gurus says that he makes full wiring diagrams for avionics installations.  I don't have solidoworks electrical to make graphical diagrams, but I have and excel.  So I made a map of every connector on every main piece of avionics and I'll print that out and put it with the logs for the next go around.

One thing I discovered is that there is only on AIRINC GPS out on the 430W.  From what I can gather in the installation notes in the 330, this is required for the transponder installation.  However, the 330 can serve as a data concentrator as send multiple AIRINC signals back out.  I'm going to do the G5 HSI / autopilot over the winter, but the GAD29 will also rely on the same AIRINC that the transponder needs.  So I'm coming to terms with the possibility that the HSI may be dependent on an output from the transponder - I don't like interdependence very much.  

Anyway- I'll make a harness for the GAD29 and dig around the back of the panel once instead of twice in the course of a few months. 

-Brad.