CONSOLE
There are three inter-dependent aspects to the console; the instrument pod, the instrument panel and the center line switch support.
INSTRUMENT POD
INSTRUMENT
PANEL
SWITCH
SUPPORT CONSOLE
GENERAL
Before actually fabricating each of these items, I believe the builder should spend a bit of time and consider the ergonomics of their design and layout. Just building a box and punching out the required holes is relatively straight-forward, but it can create subtle safety issues. By doing a little pre-planning one can avoid some of the obvious pitfalls. There have been several articles published on panel layout and there was an ongoing series of articles in Kitplanes magazine. I chose to make some paper copies of the various instruments, avionics and switches that included both their visual areas and also the fully mounted outline. These were then used to lay out the proposed panels on foam board and to make sure that everything would fit.
I believe that one should try to keep the most important information within a relatively small scan window and then work out from there. Indicator lamps should be grouped together and near the scan window to catch the eye rather than just placed whereever space is available. Switches should also be grouped and progress from engine startup, to normal operating items and then to lights. If using a separate breaker area, then it should also be arranged from most to least important. Similar looking/feeling controls, like the mixture and carb heat, should be spaced apart or at least colour coded. Even though I tried to consciously think about these factors, when I actually installed the various components into the physical panels I identified several subtle changes that I then made; unfortunately a couple of changes couldn't be made without re-making the entire panels.
The single most important instrument to my way of thinking is the dual tach and I placed it in the top left position so it is always first in the scan. In cruise flight the altimeter may be the next most important instrument but in an auto it's the airspeed. The ASI is also very important in mountain flying since there is a natural, but subtle, negative tendancy to alter it during some situations. I placed the ASI next to the tach to keep a small scan during autos. The altimeter was also deemed to deserve a top row spot and it was natural to put the VSI below it. The MP gauge which is indicative of power was logically grouped below the tach and the engine instruments made a logical progression along the bottom of the panel. The Dynon EFIS is an optional system for me but I wanted it centrally located if it is installed. I find that I look at the GPS, radio(s) and transponder when required, but they're really not part of a standard scan. Thus they were relegated to the lowest postion on the panel with the GPS at the top of the stack since it has the potential to be the most used.
I know there has been an interesting discussion on ROG in which some of the CFI's were saying that the top row should be: ASI, tach, VSI. This was based on keeping a small scan window during the approach phase and trying to avoid vortex ring state, especially during steep approaches. It's my understanding that the original VSI position in the Rotorway was partially hidden as there is very little room for the larger 3-1/8" gauges. In light of this, I still think my layout is fine and strongly believe that the key to safety is a convenient scan window and a disciplined attitude by the pilot.
When I purchased the kit, I didn't know whether I wanted to use the optional T-panel pod or to build the "boxy" aluminum pod as depicted on the Construction Prints for which enough material is supplied. In order to keep my options open, I actually purchased the T-panel pod.
The pictures I had seen of the T-panel pod showed it made of carbon fiber and I assume epoxy resin; needless to say I was surprised to discover the delivered pod was made of regular gel-coated fiberglass and polyester resin. I still don't have a comfortable feeling after my inquiries about the material switch from conductive CF to non-conductive glass and it's effect on RFI from/to the radio and transponder or the proximity of the compass to the dual tachometer which will be closer than the manufacturer's recommendation.
I decided to try fitting the pod to see how I liked the physical arrangement and to try make a final decision about which kind of pod to actually use. First I sanded the mold seam in the gel-coat and then I opened up the mounting tube holes to the marked indents using a dremel tool and sanding drum. Using some spare tubing, I verified the holes were opened up enough and aligned with the molded-in aluminum angles. So far so good, the marks were slightly off but not enough to worry about and the holes were easily opened up some more.
Before cutting the tabs off the frame mounting uprights or opening up the holes to allow the tabs to pass through, I decided to do a quick alignment check. Wait a minute ... something's not right! After numerous re-measurements and head scratching, I finally discovered that the upright mounting tubes had been constructed according to the plans and had a 9" outside dimension. The inside of the pod however was about 5/16" less than this ... further checking showed that even the outside of the gel-coat was ~1/8" less than the uprights the pod was supposed to slide over! In discussions with the factory, it seems their recommended installation method is to cut the uprights off at the top of the aluminum angles, grind the mounting tabs off, and then BEND the 3/4" 4130 uprights to fit inside the pod before through bolting.
Not very impressive since the same company controls both manufacturing plants. A simple spacer in the mold could eliminate all the headaches and at the very least a forewarning of mounting problems should be given. Then again, there was absolutely no instructions included with the pod, not even a simple picture of the mounting. I realize how this kind of problem with the thickness of the layup can occur when going from a male plug to female mold to final product, but it still doesn't explain the last 1/8" discrepancy in outside dimensions. Perhaps this 1/8" is due to polyester shrinkage during curing. For now, this is on hold but I doubt I'll actually use the T-panel since I refuse to advertise by inference a product that I can't recommend. There were also several voids in the glasswork that caused the gelcoat to crack / chip away from the underlying cloth & resin.
I started working on a plug for my own version of the T-panel pod. Some of the key differences I'm making are:
Originally I was planning on using vacuum-bagged moldless layup right on the plug using two layers of carbon fiber sandwiching a layer of glass mat. Instead of laying up directly on the plug, I'm going to try cover the plug with release tape and its possible that it will survive the process relatively intact. In the event it is reuseable, I will arrive at a price in case other builders would like to have another pod pulled from it.
The first version of my console has been pulled from the plug. I'm happy with the shape and barring any real problems I don't think there's any question which one I'll be using. Over the next week or so, I'll be adding the internal flange for the instrument panel mounting and then physically mounting it to the frame. In the photos below it looks pretty rough, but that is because this version was done as a vacuum-bagged moldless layup and the white stuff is a thin coat of sanded micro balloons which is used to obtain a smooth surface. Preliminary weight measurements indicated that the factory console is ~2.5Kg and mine is ~1.0Kg for a weight saving of approximately 3 pounds.
Although we did place the two pods side-by-side at the mounted angle, I didn't take a picture of that. Here I've used a graphic program to rotate the images to what I believe is their relative angles when mounted:
The new console has been mounted and I'm really happy with it. It's definitely less obtrusive than the factory T-panel and has room for all the things that I could think of at the time. Perhaps there are a few changes that I'd do if I was starting from scratch again, but for now its time to move along with the rest of the construction. This took a LOT of time and effort but it was something that was bothering me and I knew that I would always regret not making the change during initial construction.
Perhaps I'm being a bit obsessive about trying to make this pod rigid, but I have no desire to have "buzzing" instruments. Instead of just adding internal tabs for mounting the actual instrument panel shock mounts to the pod, I laminated in a full flange about 1" wide. While I started with 1/8" phenolic, I had to switch to some FR4 board when I ran out of phenolic. Once the instrument panel (with the cutouts) has been mounted, I intend to grind down a lot of this flange for the required clearances. Even leaving a 1/4" ridge will add a lot of stiffness. I also glassed in a piece of ~ 1/4" tubing (actually the sleeve from a multiconductor cable) top to bottom about half way back. This will be used to route the compass lighting wires from the top to the bottom and the flox/glass also adds extra stiffness.
I decided to add some rubber molding to the front edge and
this turned out to be a major pain ... perhaps some kind of vinyl
edging would have worked better and could possibly be heat molded.
Although I'd done some test strips with the molding and adhesive,
I still had some bonding issues ... basically the adhesive didn't
want to fully stick to the molding. I finally resorted to using
the adhesive in a much wetter state than the recommended "tacky"
and had to fully hold the molding in place with tape. Then came
the cleanup of the tape residue, etc. etc. I'm sure glad that
this is done and I hope that I don't have to go back and work on
it again. Since CHR didn't include any instructions with their
pod, I'm not sure what they recommend for edging and adhesive.
Update: Perhaps some self-gripping edge trim such as McMaster's part # 8451A55
would work. I can't say for sure as I haven't worked with this
material. The one question about metal cored versions might be
whether the material is ferrous and how close it is located to
the compass.
Update: After talking to other builders and carefully looking
at them, I believe the CHR-supplied shock mounts are too soft to
effectively do their job. I have a couple of extra shock mounts (10
total) and I'm still worried about their ability to support the
weight of the panel with all the instruments and avionics
installed. I'm trying to locate some 1/16" silicone sheet to
place under the bottom lip ... I've got the ideal material in
electronics instrument case feet, but they're only 1/2"
round. If I can find a suitable material, I'll add a couple of
inches of it such that it will take the majority of the weight
and the shock mounts will only be there to absorb vibration. The
other alternative will be to use a bit of the 1/16" rubber
material that was supplied in the kit to pad the fuel tank
saddles. Also note that the shock mounts I ordered from an
aviation supplier are larger in diameter than the CHR-supplied
units ... the downside is that they have ferrous (versus brass)
studs in them.
Update 2: I noticed on a factory assembled craft that they chose
not to install any shock mounts on the panel. It was directly
mounted to aluminum channel that was bonded to the pod, although
there is one layer of leather between the panel and the mount.
The studs on all the shock mounts that I have are slightly too long for the front side where it goes through the panel and then is held on with a brass acorn nut. It was relatively easy to trim these studs down to 1/4" with a cutoff wheel but I was also careful of heat buildup. I was going to paint the acorn nuts a flat black but was worried that this would easily be rubbed off during installation ... of course they could be painted with a small brush after installation but this still leaves a potential problem if the panel is removed for maintenance. I ended up sending these nuts out to the local powder coater and had them coated with black dry film which is quite tough.
Physical mounting of the pod is via six 3/16" screws that come through the side of the pod and then through the stock tabs on the uprights and it also bolts to the switch panel at the bottom. Phenolic blocks had been inserted into the mold before layup to reinforce the shell and provide solid mount points. Since there was still a bit of twisting that could occur horizontally between the two uprights, I used a piece of tapped 3/8" aluminum rod on the inside between the tabs at both the top and bottom instead of just nuts. This stopped most of the twisting effect and mounting the bottom of the pod to the top of the switch console really locks this whole pod into place and makes an extremely rigid structure.
Also note the trimmed flange inside the pod ... a bit hard to see
as it's been painted black
I'm also mounting a slim fan (Digi-Key part # P9732) in the bottom of the pod. The opening will be covered in copper mesh to act as both an EMI/RFI filter and also as a finger guard. The mounting order will be: pod, copper mesh, RTV gasket and then the fan. Hopefully the RTV will help prevent any amplification of the fan noise and also allow for any irregularities in the pod's surface. Note that I'll also be using a .01u capacitor right at the fan to try filter any electrical noise from entering the radios or power bus. I've also added an airflow dam in the hole at the bottom of the console that leads to the console switch support since the main air exit is via the gap around the perimeter of the instrument panel and the avionics.
Unfortunately the opening to the switch console has become very crowded with wire, coax and tubing which means that my original airflow dam won't work. I still need to investigate this further, but may need to resort to using a separate avionics fan to cool the GPS/COM which is the topmost unit in the stack. The crowded opening is further compounded by the fact that I mounted my radio stack too low in the panel. While everything fits physically, the wiring runs to the panel do not have a clear path to the instruments and avionics. In order to prevent chafing and other nasty problems, I ended up making a composite conduit to route the wires from the opening around to the side of the stack. I also have some fairly large kynar heatshrink that I might put around various wire bundles as they pass through the conduit; alternatively, I also have various sizes of expandable sleeving.
Since the opening into the pod became very crowded, it would have been very difficult to mount a second avionics fan outside the pod and route its tubing into the pod as planned. As a compromise, I installed a bevelled piece of tubing onto the transponder's air inlet and a short piece tubing onto the GPS's air inlet ... both of these terminate directly above the fan in the base of the pod. Since the radio is extremely long and the back about one-third of it sits above the fan, it will also get some direct cooling from the pod's fan.
There is one thing to be very conscious of if
one uses a fan within the instrument pod: the Construction Prints
show the static line as just being left open within the
instrument pod. With the addition of a fan, there may
be a slight pressure increase which will effect any instruments
that use the static line (i.e. altimeter, ASI and VSI). I'll be
looking at moving the static port to either within the switch
console or possibly a dual-port arrangement with one port on
either side of the console or cabin. It should also be noted that
the static line must be accessible when performing an altitude
encoder check/calibration.
Update: My static line runs down through the switch panel and
then through the floorboard. Time will tell if the area below the
floorboard is a good place for it, but it's out of the cooling
fan's effect and should not be directly effected by pressurized
cabin heat airflow.
Once the compass has been mounted, I'm going to look at the possibility of mounting the GPS antenna at the top front of the pod. The bubble won't interfere with it and it will be a very short coax run to the GPS unit, but it will require careful testing to verify that there is no interference between the compass, GPS and radios. The one consideration that I haven't fully investigated is to mount it low enough to not be obtrusive, yet high enough that there is minimal blocking of the sky from the pod and the back of the cabin. I also need to look at the distance from the compass to the tach and determine if I am going to install magnetic shielding between these two. While the "plastic" housing of the stock tach could be wrapped in this material, it may be just as easy to laminate a layer of it into the top of the pod.
Update: I'm going to try this position and will update the comments when it's been fully tested
I had to make a decision whether to make a female mold from my console after it had been cleaned up but before it was mounted. While this would have been relatively straightforward, it does involve time and materials plus it would have delayed my overall construction. There is also the question of how many units would ever be sold and what price one could charge to produce them. In the end, I decided not to make the mold but I still have the original plug and might consider doing this if there was some guaranteed interest in it.
I have seen pictures of several consoles (ie. pods) that builders have made just from sheet metal and some of these appear to be really nicely done. With a little planning and creativety these can be made from aluminum without having to resort to composite construction.
Hindsight:
If I were to rebuild a pod with what I know now, I would change the panel tilt angle. Unless one is specifically using gyro instruments that require an 8° tilt, that angle is too shallow for a clear view of all the instruments and avionics. The Safari seating position is quite high and very upright ... as a result it is very hard to clearly see items near the bottom of the panel unless they're tilted at a considerable angle upwards. As a quick and dirty change I may try changing the various shock mounts in my panel to have longer ones at the bottom and shorter ones near the top. This will slightly change the panel angle but not as much as if I were to start from scratch.
Another change I would make would be to add Mu-Metal into the top of the pod during construction in order to try isolate the compass from any magnetic interference generated by the instruments. I ended up adding a layer of this after final assembly but it would have been much easier to have done it during fabrication.
Tips:
If one were to design and build their own pod and console, they might want to first consider if there is any possibility of ever mounting an attitude or directional gyro in the future. The first problem with this is that these are very deep instruments and the electric versions have a long Canon-style plug on the rear. It is possible to special order some of these type plugs with a 90° bend which shortens the mounted length, but I haven't tried to locate a part number. The second problem is with the panel tilt angle as some of these gyros are orderable with a fixed tilt angle while others have an internal adjustment. In either case, it would appear that about 8° is the normal limit which is a shallower angle than shown on the Construction Prints. I haven't confirmed the T-panel pod angle at this time, but I believe it is also greater than 8°.
When laying out the instrument panel, it is important to check the size for each of the instruments in their final mounting position, especially if one is using some form of back-lighting like Nulites. In addition to checking for interference with surrounding components, this also needs to be done for their depth behind the panel, both for the main body and for the protruding mounting lips. How do I know ... I was trying for minimal edge distance and carefully checked my spacings using a Westach instrument which is about 3/8" thick on the mounting lip. Combined with the 1/2" thick vibration isolators, this allowed me to put a fairly wide panel support flange around the pod without interference. HOWEVER, the lip on Falcon gauges is 1/2" thick and using Nulites adds a further ~1/4". Needless to say, I had some interference issues that needed to be resolved.
If one is designing their own pod and minimizing size, they should be aware of the wiring runs that will be needed later on and allow some space for this. It's amazing how quickly one can get a bundle that's approaching an inch round, especially if it's being held in place with adel clamps. One also needs to consider how the panel will be located while the wiring connections are being terminated.
I remember seeing a reference somewhere about why there's no need to make the front (or a side) of the pod removeable. I agree that it's not necessarily required, but I do know that it would have made my final installation a lot easier. I chose to leave the bubble off while I did my console wiring and it would have helped with some some of the final routing and installation to be able to see it directly rather than in a mirror.
Before making a custom pod (or the instrument panel for the T-Pod), one will need to lay out the location of the various holes for the instruments and avionics. I used printouts of the actual instruments as I was doing this and attempted to keep things relatively compact. While I feel the result is very good ergonomically, the actual installation was quite difficult due to using standard width avionics located in the bottom of the panel ... this created a difficulty in routing the wiring. This would not be a problem with the round-style avionics and if one looks at most Rotorway panels, they'll note that radios and transponders are typically mounted in the center area which allows for easy routing of the wiring below them.
The construction prints show the use of .065" aluminum for the actual instrument panel face on the square pod/panel and I assume that this is the material that the factory implies is to be used for the larger T-Panel Pod since only .063" material is provided. With the larger panels on the T-Pod and my own console, I was somewhat worried about flexing of the panel, especially under vibration, and whether this would create the potential for "blurry instrument syndrome".
I checked with my local suppliers whether they had any .071" or .080" 7075-T6, but unfortunately they're no longer stocking it for cut pieces and I would have had to either order a full sheet ($$$) or mail-order a piece of the right size. I ended up getting some .080" 6061-T6 which I plan to have hard-anodized after all the cuts are made. The anodizing will add a bit more rigidity to the piece and I'm not sure if I'll just leave it as the hard-anodized surface or apply a vinyl facing to it. I wasn't too worried about the weight increase of the thicker material since it will look like swiss cheese after all the cutouts are made.
The actual panel will have about a 1/16" gap around the perimeter when it is mounted. This allows for some movement of the shock absorbers and also serves as the air exit for the cooling fan.
I had originally planned to get a combination 2-1/4" and 3-1/8" instrument hole punch in order to make the instrument holes. These are available from most of the usual aviation suppliers for $120-$150. Not cheap, but I figured this would be an easy way to obtain accurately placed round holes. Unfortunately all the suppliers at Sun-n-Fun were out of them and when I later tried to obtain one, they all seemed to be back-ordered. I had previously tried both a fly-cutter and a 3-1/8" hole saw that I had. The holes from the hole saw were oversize and a 3" version would have needed a lot of clean-up. In general, I have an aversion to fly-cutters as the work needs to be TIGHTLY clamped and any movement of the work or cutter adjustment will ruin the hole and possibly the entire work piece.
I did obtain an HC1
adjustable hole cutter from Aircraft Tool Supply as I had read an
article from an experienced metal worker who was very fond of
them. Well, maybe he can get good results from them, but I found
it was very slow and produced a very rough edge; sort of like
melting its way through the aluminum rather than cleanly cutting
it.
Update: I tried using this tool on some .032" material when
I was cutting holes for for SCAT tube ducting and it actually
worked quite well. I still don't know whether it was the
material, my technique or just a bad day that caused the original
problems.
I eventually came full circle back to the fly-cutter. For me, the trick to using these is to get accurate work placement with the piece being well clamped in place on a drill press with a wood backing. By using a very low speed on a "torquey" drill press, I was able to make reasonably accurate holes.
I used an instrument hole jig to drill the four attach holes around each of the circular instrument cutouts. I did a test with the jig and it seemed like a reasonable (but not perfect) way to do this. So far all the holes are useable without enlarging them, but ... when I went to install the Westach gauges I discovered that they provide 6-32 x 3/8" FLAT headed screws with the instruments whereas CHR had provided pan head screws for the Falcon gauges. I suppose I could have tried to countersink for the Westach screws, but I would have had to be very careful on their location. They also appear to be 82° screws rather than the more normal 100° aviation screws. More importantly, the Westach screws that I received contain ferrous material (i.e. magnetic) and I've been trying to only use aluminum, stainless and brass on my panel (ie. non-ferrous materials). I chose to replace these screws with MS35214-25 brass pan head instrument screws. Ironically, the biggest source of ferrous material in my panel is D-Sub connectors ... I need to see if I can find ones made with nickel or some other non-ferrous material.
If I was going to do this again, there is no question that I would go with laser cutting. This would be more expensive than a fly-cutter, but it would have saved a lot of frustration and testing of various tools. Luckily, I do have access to AutoCad, so it would have been relatively easy to create the layout for the various holes. Even if I opted not to layout the outside dimensions / radiuses and only had the holes cut into a rectangular sheet, it would have made life a lot easier. If I spent the time to do the full layout for laser cutting, including outside edges, then it would make it extremely easy to replicate the panel with subtle changes. The following is a classic example of why one might want to do this.
I just got to experience another problem of not having the the avionics with me at the hangar and double checking them while I was cutting the panel. I knew that there was some discrepency in tray widths with some manufacturers use 6.25" and others using 6.30" (or even 6-5/16" = 6.3125"). Although it's an extra step, this can be easily handled by using shim stock on the narrower trays. What happened is that I got so obsessed thinking about these discrepencies that I overlooked an even bigger factor: all these widths are discussing the tray/rail widths for FRONT insertion of the trays. However, if one is installing the trays from the rear, as in a removeable panel like this, then the cutout width can be ~1/8" narrower (6-3/16").
So what's the big deal? Either width cutout will work, but the difference is cosmetics. By using the narrower width cutout then the edges of the avionics faceplates will actually cover the edge of the cutouts. With the larger width cutout, the edges of the faceplates are inline with the edges of the cutouts and one can just see the cutout edge. Of course I figured this out AFTER I'd finished all the instrument holes, screw holes and the wide version of the tray cutouts. I'd thought that it would be a simple task to fabricate and attach the rails once all the other drilling and cutting had been done so I'd saved that task till last and hadn't noticed the discrepency. At this point it would be a real pain to redo the entire panel, so that leaves the option of leaving it the way it is or trying to narrow the large cutout by ~ 1/16" on each side without making a bigger mess. Since this isn't a show craft, I think the answer is obvious ... somehow I'm sure my eye will always be drawn to this and it will be on the "some day" list to correct it. Also, I'd always assumed that I'd be mounting my avionics racks in the normal method whereby the rack is flush to the back of the panel and the bezel projects about 1/2" ahead of it. In hindsight, it should be possible to mount the racks deeper such that the front of the bezel is flush with the front of the panel, but one would need to be very careful in the layout and positioning.
For the radio stack rails, I used 1/16" aluminum angle on one side and 1/8" angle on the other which were then riveted to that actual panel. Although I had reservations at first, the 1/16" angle should be fine, especially since I'll be using a rear support. The reason I used 1/8" on the other side is because of my layout and the placement of the intercom and ELT remote switch; the rail on that side is severely contoured and there are only a couple of narrow "feet" that provide support against twisting. The actual racks are held in place with 6-32 stainless countersunk screws (MS24693C), washers and elastic stop nuts. I started with 1/2" long -28 screws because I had them, but later switched to 3/8" -26's and -27's.
I made the rear support out of .063" aluminum and although it looks really weird at first, there is a reason. I decided to mount the altitude encoder above the radio stack but physically attached to it. The interconnect wiring is then relatively short and there is no need to remove or install the encoder connector while working on the stack or panel. Additionally, I chose to mount the encoder at an angle above the GPS/COM tray. By just removing the GPS/COM unit from it's tray (a simple procedure), I can then access the altitude adjustments on the encoder from the front without the need to remove any panels.
After talking to my anodizer, I decided not to have the panel hard anodized since they were concerned about the resulting colour with the mix of 6061 aluminum and 2017 rivets ... seems like I forgot that I also wanted it dyed black afterwards ...dumb! Against my better judgement, I chose to alodine the panel and then just paint it with a can of Krylon epoxy flat black over their primer. I'd done a test piece with this paint and it came out extremely well, but one really needs to wait several days (5 according to the can) until it has its full cure and abrasion resistance. Unfortunately, the actual panel didn't seem to fully cure after a full week and I finally had to bake it (200°F for an hour) before I felt it was useable. I'm pretty sure this is related to the use of their primer as the toughest test strip is actually just the paint over bare aluminum and the primer seems to be "soft" for lack of a better description. So far I've only decided to add two more holes in the panel and these are a good way to test the durability of the paint. The primer / paint finally got the full cure but ironically I've now found what I wish I'd used ... the local powder coater has a very thin semi-flat black coating that essentially looks like a fine wrinkle paint but is extremely durable. Since my radio racks and some other items are already final-installed, I'm not about to remove everything plus the paint but it may become another one of those long-term "some day" projects.
I needed to add some labels for both the indicator lamps and the switches that I'd installed on the panel. I ended up using a Brother P-Touch unit to make up the labels and experimented with a couple of different tapes such as white on black and white on clear (which I ended up using). The problem was that these labels have a glossy surface and were then to be placed on a flat black surface. I ended up making the labels and then slightly roughening them with 1000 grit sandpaper before giving them a coat of Varathane Satin Clear (a polyurethane that I had on hand, but I would not recommend it for this application). All of this just reinforced to me how much easier it would have been to have originally done the panel in CAD and then had it laser cut followed by anodizing (or powder coating) and finally sending it out for engraving. Although it would have been more expensive (especially if it had to be redone), it would have been much quicker and more professional looking.
Before changing the CHT gauge ... really hard to get a good picture indoors with a flash
As an alternative to doing a separate CAD layout, laser cutting, anodizing and engraving, there are some specialty service providers (typically oriented to the electronics industry) that will do all of these processes in one order. One such company is Front Panel Express. Although this might appear at first to be an expensive alternative, it should be relatively painless and will save a lot of time dealing with multiple suppliers. There are also some aviation-related panel services that supply some of these functions (including harness) such as ePanel Builder.
Tip: My hangar mates tend to have their panels laser cut and then powder coated. In order to hide the rivets that hold the avionics rails in place, they use countersunk rivets that are SLIGHTLY under-countersunk (ie. the rivets just protrude). After riveting, the entire panel is then sanded using 220 paper in an orbital sander which makes the rivet heads pretty well invisible and they're totally hidden after powder coating. They've also done some of the panels with laser engraving that is then paint filled which gives a very professional look to the finished panel.
When doing my original panel layout, I scanned the various
instruments and then printed these out. The paper "instruments"
were then juggled on a mock panel until I felt I had a reasonable
layout. For those that want to do this, I'm including links to
the pictures I used and hopefully they can be copied and printed
at full size ... make sure to double check this first before you
rely on them.
ALT ASI CHT MP
Quad-1 Quad-2 Tach VSI Push-To-Test_Indicator
Artex_SW ICOM_A200 Garmin_300XL Garmin_327 Dynon
Final installation - bolted in place and ready for testing. Avionics will be installed once power circuits have been tested.
When determining the width of this panel, I chose to fabricate the width of the U-channel a little wider than the plans show such that a standard 3-1/8" instrument could fit in it. I don't plan to need this, but if one ever chose to make a new U-channel and layout including a 3-1/8" instrument, this means the sides & pod attachment will not need to be changed. Since I pre-bent this panel before I'd finished the side panels, I bent a long piece of sheet with the idea that I'd also use it for the front near-vertical piece. As it turned out, I chose to use .063" for the front and I now have enough of the bent U-channel left over that I'd be able to make another top panel of the exact same width if I ever wanted to. However, if I was to redo this panel then I'd probably use something thicker than the .032" that is shown in the plans, perhaps .050". I actually chose to make doubler plates from .063" that connect between the various switch/breaker mounts under the panel as an attempt to add extra stiffness against heavy-handed pilots. Note that you have to be careful about the placement for the nutplates on the sides if placing switches and breakers near the edges or using a 3-1/8" instrument in a minimum-width panel.
The first decision I had to make here was where to mount the rails on the floor board. Although this appears obvious at first, the issue is that seats and collectives are not centered side-to-side in the frame while the uprights for the pod are. By mounting the switch console centered on the seats, it will be off center on the uprights/pod whereas centering it on the uprights/pods means the pilot's foot space is off-center as is the visual reference to the center collective. Of course, there's a third alternative of centering it on both ends which results in a slight rotating of the switch console relative to the frame. After making a foam board mockup and looking at the various alternatives, I finally decided to center it on the front uprights. There is still adequate foot room on the pilot's side, the switches are closer to the pilot's hand and it "looked about right".
I also chose to make the side panels 1/2"
higher at the rear than what's shown on Construction Sheet #44; i.e.
4-1/2" on mine. This puts the switches a little closer to
the hand and adds a little extra room for the wiring underneath.
The downside is that it slightly reduces the amount of space
available for switches and controls on the forward sloped part. I
chose to mount my Hobbs meters on the rear part of this panel and
although they are partially hidden by the copilot's collective,
they really aren't required in flight. It should be noted that
these meters are about 2-1/4" deep and can potentially
interfere with wiring bundles in this area.
Update: I have seen and measured a CHR-installed switch panel
that was 4-1/4" wide and 6-1/2" tall on the rear
horizontal section.
When attaching the side panels, I had to make a decision about whether to make just one or both sides removeable. I chose to make both the top panel and the right side removeable via screws and nutplates. Making the top removeable allows this panel to be replaced if I decide to change the switch layout and/or any embedded instruments at a later time. While it would have been slightly easier from a wire routing standpoint to make the left side removeable, I chose to make the right side removeable. The reasoning being that during the wiring process and any subsequent maintenance in here, it will be easier to access this area from the right side without having to wiggle around the left collective. By fully riveting the left side to the floor sheeting, this should add a bit of extra rigidity with the downside being that the floor can't be removed without drilling out a few rivets ... hopefully this should never be required. If it is, there's a much bigger problem dealing with the wiring.
While fabricating the side and front panels for the switch console, I was concerned about making this assembly very rigid. Since the bottom of the instrument pod uses these pieces for support, any movement in the switch console will be transferred to the instrument pod and in addition to being harmful to the instruments and avionics, it will show up as a visible vibration when trying to read the instruments. Without the front belly rib installed, I noted that it was actually a flexing of the floorboard in the side-to-side plane that was allowing the switch console to move. I don't know if this is a result of my changing the floorboard's material and thickness, but I added a length of angle stock under the floorboard to try eliminate any flexing that may be transferred to the console. While the front belly rib adds a lot of support, any floorboard flexing also causes a flexing along its top flange.
The final step in tying the switch console to the instrument pod was to add some angle stock around what will become the wire routing cutout at the front top. This really locks the switch console and pod into a solid assembly. The side support pieces were made from 1/16" angle stock while the front and rear pieces were bent from .063" sheet stock since they're not at 90°. Of course, once these were all in place I discovered that the angle stock really wasn't a true 90° either. Since I intend to clamp these two parts together and the instrument pod was a one-off custom, I decided to use JB-weld to perfectly mold/contour the aluminum to match the pod. This sounds backwards since the pod is already made from composites, but I still need to sand and paint it separately before final assembly and didn't want ridges along the bottom.
These pieces have been contoured since this photo in order to make a bigger hole for the wiring to pass through.
My side panels are very similar to those shown
on Construction Print #44 and because of the various angles, the
length at the top where it joins the instrument pod is ~ 4-1/2"
which leaves an opening ~ 3" long after allowing for the
angle stock . I seriously considered the use of Canon-style plugs
to be used as a panel disconnect that would allow the actual
instrument panel to be completely removed from the finished craft
with a minimum of hassle. However, the largest common circular
plug is the size 24 one which is ~1-3/4" round/square and
has 61 pins. When I looked at the various wires that I wanted to
run to the panel, this was not enough pins and would require a
second connector. Unfortunately there isn't enough room for a
second circular connector and I would have to use something like
the D-sub connectors. The concept of plugs and an easily removed
panel would have been much easier to implement if I'd made this
opening about 1/2" longer. Since the sides, rails, pod etc.
are all fabricated at this time, I won't be changing the physical
console parts and it makes the idea of a disconnect much more
difficult.
Update: After doing the wiring and then re-thinking this problem,
it's amazing how much clearer the problem and solutions become.
The flaw in the above logic is that I was thinking about mounting
the connectors in the hole at the top of the switch panel.
However, a much more practical solution would be to have the
wires first come through the hole and terminate in connectors
perhaps 4-6" below the hole. The mating connectors could
then be attached to the front of the switch panel for rigidity /
durability and the wires run to their various terminations. In
this way, the instrument panel would be totally removeable.
Another change I'd make if re-doing the switch
console is the size of angle stock I used at the back where the
console joints the seat front panel. I used 3/4" x 3/4"
x 1/16" aluminum angle but it should be noted that the
opening for the wire pass-through is relatively small; using 1/2"
angle would have given another 1/2" of width to the slot. So
why the issue ... Firstly, my Hobbs is right in front of this
area which somewhat limits the useable area. Secondly, rather
than one large bundle of wire passing through here, I prefer a
style used by an avionics technician that I know whereby there is
a bundle for the power wires, another bundle for low-level
signals plus another bundle for "noisy" wires (such as
mags and strobes) while also keeping the antenna coaxes separate
from all other wires. Essentially that creates four bundles, but
in my case it's only three passing through this area since my
antenna coaxes run under the floorboard. Anyone who has tried to
isolate and remove noise from their radio or intercom will
certainly appreciate why I consider taking these extra steps to
separate wires.
Update: The above is a great theory and approach to the wire
bundles. However, due to the relatively small size of the pass-through
it forced me to do some compromising.
If one has a lot of wiring going to the switch panel, they may want to investigate hinging the panel in order to gain access to the various switch and breaker terminals. Since these devices hang down from the panel, it is relatively difficult to access their terminals once the panel has been installed. I chose to pre-wire as much of this panel as possible on the bench before doing the final installation of it. This way I could make sure that all terminals were truly tightened and it was easier to route the wiring. However, I know it is a difficult maintenance area and I hope that I never have to do any significant changes in this area. With hindsight, I do wish that I'd considered how to make this panel easily moveable for maintenance ... note that this would probably require that all the wires leave the panel in a bundle and in only one direction or at least large service loops.
I chose to attach the ground bus to the switch panel side on the left side. I figured this was a relatively central area and it was done without a whole lot of thought as I was in a rush to drill the mounting holes in order to get some painting done. I would not recommend this mounting location as it creates a real tight area under the breakers for maintenance. A much better location for the ground bus would be on the solenoid board and possibly adding a few small remote busses for things like avionics (note that one needs to be very careful about noise and ground loops when adding multiple ground busses).
Ready to install actual switch panel
Layout
Once I knew the physical size of my panel, I then made a posterboard template of it. I also made templates for the various switches and breakers (including rear projections) that I wanted to install in it. It was a lot easier to just move these around and decide on the final layout before actually drilling any holes. For those that have access to it, this would probably be a lot easier to do with a CAD program. My general philosophy was to put the switches on the front sloped part of the panel and breakers on the flat rear part of it. The switches were arranged for a top to bottom sequence during startup and operation while the breakers were also grouped somewhat.
Mixture and Cabin Heat controls are not installed in these pictures
If one is installing standard gauges in this panel from the rear, they're normally designed for thicker material and will project in front if the panel is built from .032" material. I didn't like this look so I fabricated a spacer plate for the clock from .032" material, although in hindsight .025" would have been better.
Tip:
It seems like electrical equipment is normally only added, not removed. I would suggest leaving room in the switch console for another breaker or two as its a lot easier to lay this out and leave room for it during initial fabrication rather than having to try squeeze these in at a later date and drill into a completed panel. An example might be for a governor if it wasn't installed initially. I found some circular plastic cover buttons in the local hardware store that will easily fill these spare holes while still looking very neat and tidy. Also, I've left room for a spare indicator lamp in my instrument panel. While I have a low-priority use for it now, its already installed, symmetrical and available if I ever find the need to have another one. Again by way of example, if LASARŪ mags are installed after the initial construction then they should have an enunciator.
Tip 2: Certain kinds of switches and breakers use a tanged washer to prevent rotation after installation and the proper mounting configuration is the main hole with a small hole (or depression) above / below it that captures the tang. This can be difficult with thin panel material and I've often seen these washers either eliminated or mounted upside down which nullifies the anti-rotation capabilities. One relatively simple solution is to make a doubler plate from .040" - .063" material that sits behind the main panel and captures multiple switches / breakers. This doubler has both the main holes and tang holes in it with the result that one can still use a thin main panel material while having the full anti-rotation capabilities.
The harder anti-rotation situation is found in the P&B breaker switches as supplied in the Safari's instrument kit. The proper way of making these holes is with a punch that forms a hole with two flats along the vertical edges. These are very difficult to form by hand and I assume most builders will do the same as myself and just form a properly sized round hole. One then needs to be very careful when tightening the retaining nuts and perhaps might consider the use of a bit of blue locktite.
Panel pre-wired as much as practical and ready
for installation
Mag wires were added just before installation and there are a few
terminals that still need to be wired post installation
Ready to be tested and closed up. Note that the two radio coaxes (far ones) have been brought to bulkhead connectors so that they may be easily swapped between the belly antenna and the boom antenna. Breakers will be closed one by one during testing as each circuit is individually tested.
<NEXT> Avionics, Instruments and Electrical
<BACK> Cabin and Interior
Last updated: September 18, 2009
