Let me introduce myself: I’m a long time geek working in the computer industry on open source software. I spend too much time in front of a computer. Way too much time.
I needed to do something to get away from the computer. Something physical – something to get dirt under my fingernails and sore muscles. Something where you can actually touch and see results at the end of the day.
So I decided to restore a car. [Insert scary music here]
Note: while WordPress puts the most recent posts at the top, you can also read the story in order starting with The Car Shows Up.
In the last post we made a Gridfinity bin for a single ratchet. The most flexible approach is to make a separate bin for each ratchet. This takes full advantage of Gridfinity flexibility – you can add more ratchets just by making new bins and re-arranging the bins on the Gridfinity baseplate.
The downside of making a bin per ratchet is packing efficiency – in many cases you can fit more tools into a given space by putting multiple tools in a single larger bin. Basically you arrange the tools in a way that minimizes space and then put a bin around them. In many cases this can reduce the required space by more than 25% by using this approach.
The choice is long term modularity and flexibility vs. fitting more tools into given drawer space in a tool chest. Of course it isn’t an either/or decision – in some cases one approach makes sense, in other cases the other is the best choice. And in some cases it makes sense to just pile the tools in a drawer!
I decided on a divide and conquer approach: spend time packaging my most used tools into efficient Gridfinity bins, lesser used tools into generic Gridfinity bins, and pile miscellaneous tools into a drawer.
As part of this approach I also decided to do selective upgrades of certain tools. For example, I have a random pile of screwdrivers that have accumulated over the last 50 years. These range from good screwdrivers – mostly 1970’s Craftsman – to junk that should really be thrown away. I’ll cover the screwdriver strategy in a future article.
I have, ummm, “several” ratchets. And breaker bars. And accessories like extensions and adapters. Some of them are junk. Over the last five years or so I’ve built up a collection of good Tekton ratchets that are my go-to for anything involving a socket. I have nine of these – eight regular ratchets plus a ratchet with a 24 inch handle for heavy jobs. Plus a three foot 3/4″ breaker bar for really heavy jobs, like the ball joints on a 1963 Imperial. I won’t try to put the six foot pipe, AKA cheater bar, that enhances the three foot breaker bar into the tool chest.
Studying the situation I decided to put the two foot ratchet and three foot long cheater bar across the back of the drawer and not bother putting them into a bin.
The remaining 8 ratchets were a mixture of 1/4″, 3/8″, and 1/2″ drive and included regular handle, long handle, stubby handle, and flex handle. All of these are useful and needed for dealing with various fastener situations.
Grab these ratchets and start trying different arrangements. The goal is to fit them into a minimal space while being able to easily pick up any ratchet.
After playing with various arrangements it looks like I can fit them into an 11 x 4 grid unit bin. My 3D printer is capable of printing a 5 x 5 bin, so I used the Fusion Gridfinity Generator to generate two 4 x 4 and one 4 x 3 solid bins.
I positioned these bins together in Fusion and then started creating the outlinesfor each ratchet. Once I had the outlines I arranged them in the bin with equal spacing between each ratchet. These outlines were extruded and subtracted from the solid bins to create the cutouts. After making the solid cutouts for each ratchet I studied the result for a while and then created a set of finger reliefs to make it easy to pick up the ratchets.
As a last step I filleted (rounded) all the edges. On heavily loaded parts this is critical for strength and to avoid cracks. Sharp inside corners are the most common place for cracks to start. It isn’t really critical for these bins, but it is easy to in Fusion, looks better, and keeps my mechanical engineering spider sense from tingling so strongly.
With the three bins designed, send them to the printer one by one. Total time for this was over 12 hours, but 3D printing is “fire and forget”.
Gridfinity Ratchet Bins
The three bins drop into the Gridfinity baseplate locking them into position. Add ratchets and done.
Ratchet bin with ratchets
The ratchets fit. Once the bins are dropped into the baseplate they don’t move. This fixes the previous problem with the ratchets moving around when the drawer is opened and closed. You can easily grab whichever ratchet you need. And you can tell at a glance if you have returned all ratchets to their proper location.
Despite pounding my head against Fusion (I still have a lot to learn!) this was actually a fun little project. The results were exactly what I was shooting for: creating the CAD model and learning more, 3D printing, and fitting the end result into the tool chest.
By this point you should be familiar with my approach to something new: start with a simple test piece, iterate until I’m happy, and then move through progressively more complex designs until I get what I want.
So, instead of jumping directly into designing a tray for all nine ratchets, grab one ratchet and design a tray for it. One of the goals of prototyping is to minimize the amount of filament you waste, so start with the 3/8″ stubby ratchet instead of the 1/2″ long handle one.
First, how large a bin is needed to hold the ratchet? You can measure and check. Or just lay the ratchet down on a baseplate and eyeball it. In any case, the test ratchet is one bin unit wide and four bin units long.
In Fusion select the Gridfinity Bin generator. Tell it to create a 1×4 solid bin. We start with a solid bin and make a cutout to fit the ratchet. Hit Enter and the bin appears as a new component, ready to edit.
Next, create a Fusion Sketch on the top surface of the bin. Measure the ratchet and create an outline of the ratchet. The easiest way is to create two rectangles, one for the head of the ratchet and one for the handle. Center this outline in the bin.
Make this outline just a little bigger than the ratchet. How much bigger? Big enough to make it easy to remove and replace the ratchet, small enough for a good looking fit. A sixteenth of an inch is probably a good starting point.
Fillet the corners of the outline to make it look better and fit better. While not required this is easy to do in Fusion and makes the end result look more professional.
After finishing the outline extrude it to make a solid body and then extract this body from the bin to create the cutout for the ratchet. The depth of the cutout depends on the tool you are designing the bin to hold. For these ratchets 3/8″ is a good starting point.
Done!
Not quiet. The cutout is a tight fit on the ratchet and roughly half the depth of the ratchet. You can’t grab the ratchet – the only way to remove the ratchet is to pry it out with a screwdriver. Not the greatest user experience…
We need to add a finger relief to the bin so that we can easily grab the ratchet. The finger relief should be located where you naturally grab the ratchet. It should be wide long enough to stick your finger and thumb into it – 3/4″ or 1″ is a good starting point.
The finger relief needs to be wide enough to get your finger and thumb in to grab the ratchet – 1/2″ or 3/4″ is a good starting point. And it needs to be deeper than the ratchet cutout so that you can get your fingers under the ratchet and pull it out. I first tried 1/2″ deep, but this cut through the bottom of the bin. 0.450″ was the deepest I could make it.
I couldn’t make the finger relief as wide as I wanted to, so make it as wide as the bin allowed. Extrude it 0.450″ and subtract from the bin.
Looks good, so send it to the printer. Wait patiently.
Pop the finished bin out of the printer, drop the ratchet in, and start checking how well it works. Some room for optimization of size and spacing, but overall a successful first prototype!
Prototype ratchet binPrototype bin with ratchet
It holds the ratchet but could be a bit longer. We can either tweak the model and make another prototype or note what changes are needed and include them in the actual design.
One of the reasons for getting a 3D printer was to print out various types of organizers for the workshop. There are literally thousands of different designs floating around. A search for organizers quickly leads you to the modular Gridfinitysystem created by Zack Freedman.
Gridfinity is kind of hard to explain – it is in many ways more a community than anything else. Coming from the Open Source software world it seems familiar and comfortable. There are thousands of existing designs available to hold both things you need and things you never thought of. You can also design custom bins for your own unique needs. Gridfinity is used to store both tools and supplies. It comes close to being a universal solution for organization.
At its core Gridfinity is a specification for a baseplate and set of bins that fit into the baseplate, arranged as a 42mm grid. Yes, 42 – the answer to Life, the Universe, and Everything. The baseplate can be scaled to whatever size is needed. The bins can be customized to whatever size it needed – height, width, depth, and internal partitions. Further, you can start with a solid bin and use a CAD system to create custom cutouts to hold pretty much anything. All sizing is done in integer multiples of this 42mm grid size.
Gridfinity base and bins. Source: Hackaday.com
When used in drawers the Gridfinity base locks the bins into position and keeps them from moving around when you open and close the drawers – just what I need!
Gridfinity bins holding a variety of tools
While I have no end of “opportunities for organization” in the workshop I decided to start with my socket drawer. Several reasons for this: sockets and ratchets are the single most heavily used tools I own. Sockets are currently organized, but the nine ratchets, three breaker bars, multiple sets of extensions, and various misc. pieces are just loose.
Sockets and ratchets are prime candidates for Kaizen-type organization where you can immediately find everything. Kaizen organization is even better for putting things up – put each socket and ratchet back in its fitted location. You can instantly tell if anything is missing. I’m already spoiled by the combination of knowing exactly where to grab for a tool and easily seeing if everything has been returned to the toolbox. This makes me more productive and less frustrated – few things annoy me more than searching for a tool I set down a couple of minutes ago!
The good news is Gridfinity is modular. The bad news is Gridfinity is modular. Much like Kaizen Foam, you can fit it to whatever space is available and do custom cutouts for each tool you want to store.
The first challenge is that my 3D printer has a maximum size of 10″ x 10″ x 10″. This means that the largest Gridfinity base or bin I can print out is 5×6 grid units. Anything larger than this will have to be broken down into smaller pieces for printing. As an example, the socket drawer will need nine of the 5×5 baseplates plus two 1×5 baseplates with a 3/8″ spacer down one side and three 2×5 baseplates with a 1/4″ spacer down one side.
Lev Mishin has created a Gridfinity Generator for Autodesk Fusion that is available from the Autodesk App Store. This Gridfinity Generator can generate arbitrary sizes of baseplates and bins with a variety of options. If you are using Gridfinity and Fusion this is a must have.
The Imperial has been running hot ever since the engine rebuild. It isn’t overheating, but running on the Interstate it goes to the top of the normal range – sometimes just a bit above. Running on back roads at slower speeds it runs cooler. In cold weather it runs cooler.
Potentially related, the last time I had the AC worked on, the shop told me that the fan clutch wasn’t working – the fan was just freewheeling. While this shouldn’t have any impact on the Interstate it was another data point.
OK, check parts. Hmm, there is an option of a heavy duty fan clutch that applies more torque to the fan. Looks like I should try one of those.
While I’m looking, I’ve been a bit suspicious of the water pump. Interesting – they have standard volume water pumps as well as high volume/pressure water pumps. Might as well do that while replacing the fan clutch, so add a high volume/pressure water pump to the order.
While I’ve already replaced the thermostat once, might as well see what else is available. Would you look at that – there are high flow thermostats available! Add one of those to the order.
Replacing the water pump and fan clutch isn’t that bad of a job, but it still took a couple of hours. The thermostat is a different story – it is buried under the AC compressor and tightly protected by the 5 braces supporting the compressor. One of the bolts can be reached with a swivel socket, but the only access to the other bolt is with a wrench that can only make 1/8th of a turn at a time. Both removing and installing the bolt…
By exercising the full capabilities of my patience the high flow thermostat was finally installed with no collateral damage. Top off the coolant and time for a test drive!
Initial results weren’t encouraging. Once I got on the highway temperature headed up toward the hot line. Crud! Another dead end!
But after 10 or 15 minutes of driving the temperature started dropping, ending up on the normal temperature line. And it stayed there! Odd, what is going on here?
After a few minutes of thought, it looks like this behavior could happen as air was purged out of the system. The cooling system had to be drained to replace the water pump and thermostat, so introducing air into the system was a concern. Air locks in the cooling system are a common cause of overheating.
Did I manage to get lucky? While encouraging, this was a 60 degree day, so still not definitive. Need to try it on a hot day to be sure.
Today was in the mid-80’s so fire up the Imperial and head out. It quickly warmed up to normal operating temperature – so far, so good.
Head to the Interstate and start cruising with traffic at 70-75 mph. The temp gauge moved about a needle width above the normal line. And stayed there! I put in close to 50 miles of high speed driving with the temp gauge holding steady. Even better, the air conditioner was running the whole time, putting additional stress on the cooling system.
Getting off the Interstate the temp dropped slightly to right on top of the normal mark. Perfect! This is exactly what you would expect if the cooling system is working correctly. I need to test it on a 95 degree day, but it should still be OK.
Like most car collectors I have a cover for the car. Even in a garage dust collects on cars, so you throw a car cover over them for protection. The cover I’m using for the Imperial was originally for a full size extended cab pickup. Meaning bulky and heavy. And difficult to store when it isn’t on the car.
I usually just throw it in the trunk. Where it looks bad.
As I was standing behind the Imperial admiring the finished trunk and nodding in satisfaction I noticed the rolled up car cover. Right next to the remnants of carpet from the trunk.
Whoa – what happens if I make a bag out of trunk carpet, stuff the car cover in it, and then shove the bag onto the trunk ledge behind the back seat? This would look fine and would conceal all of the unfinished areas around the back seat that were bothering me.
Since it is the same carpet as the floor of the trunk, it would look like a finished continuation of the trunk floor.
Right! Grab the one remaining large piece of carpet and measure it. It was long enough to hold the cover and fill the gap in the trunk. But it wasn’t wide enough to wrap around the rolled up car cover. Krud! So close, so close…
Just a minute – there are still some smaller carpet scraps… Start laying pieces of carpet on the work table. Just enough to get to the needed size. Well, if I’m willing to sew smaller pieces together I can make this work!
So I started sewing smaller pieces. And it worked! The car cover fits inside, it fits in the trunk, and it nicely finishes the trunk.
Car cover in carpet bag, stuffed under package shelf.
And there was much rejoicing!
I remain very happy with the Juki sewing machine – it takes whatever thick material I’m trying to stick together and just calmly munches through it with even stitches and no sign of strain. In this case I was sewing up to four layers of carpet together – the Juki just said “whatever” and took it in stride.
A few of years ago I picked up a couple of yards of cheap indoor/outdoor carpet and (somewhat roughly) cut and fitted it to the trunk. The plan was to replace it with something better when I actually finished the trunk.
But after (semi)careful consideration, it didn’t look too bad. It was already (somewhat roughly) cut and fitted. The new side panels hide the edges of the carpet. And there were things I still needed to learn about working with carpet.
OK, change of plan – finish off the existing carpet and then decide what to do.
Something needed to be done about the rough edges of the carpet. These edges are typically bound with bias tape – a fabric tape that wraps around the edge of the carpet and is sewn in place.
Carpet binding. Including sewing error at top…
There are sewing machines specifically designed to bind the edges of carpet. Just feed the carpet through the machine and you have a perfectly bound edge. Of course these machines cost several thousand dollars.
They make gadgets that fit regular sewing machines – including Juki – and fold the bias tape over the top and bottom of the carpet and hold it in place while sewing. There are commercial versions for several hundred dollars. There are also a bunch of different models in the $10-$20 price range. Several hundred dollars is too much. But I’m willing to risk $10 and some of my time on one of the cheap ones.
Tape folder – carpet binding tool for sewing machine
You may find this hard to believe, but the cheap tape folders suck. And I suck at using them. It is difficult to load the fabric tape into them, difficult to keep the tape properly feeding through them as you sew, and difficult to keep the tape on the carpet when sewing. I kept sewing off the edge of the carpet.
It could be that this binding attachment is intended for regular fabric – I had to open up the throat to feed this thin indoor/outdoor carpet through it. There is no way that it could be used with regular carpet.
After a fair amount of practice I got somewhat less bad. But it was still a painful experience and there was usually at least one place on each edge where I sewed off of the carpet.
After finishing edging the entire carpet for the trunk I learned a distressing truth: After all that work, I don’t really like the way it looks. A big part of this is the tape I used. Part of it is probably the cheap, thin carpet. But I just don’t like it. OK, something to keep in mind for the future!
As part of the fitting process there were several places where I needed to add additional pieces to the carpet. The initial plan was to sew these pieces in place. Which was shaping up to be a difficult and painful process. With questionable results.
Inspiration finally hit on how to do this – cut and fit cover pieces and then hold them in place with velcro. This would GREATLY simplify the job and would allow fitting after the carpet was in place. And allow adjusting and tweaking until I was happy.
Further inspiration – the sticky backed velcro tape could be used on the trunk floor to hold the carpet in place and keep it from sliding around. I’m seeing more velcro in my future!
The first attempt using this approach was over the gas filler. The gas filler extends into the trunk before entering the tank. I cut the carpet to go flat on both sides of the filler; a triangular panel was needed to cover the filler. This needs to be a rather precise fit. So I have been ignoring it.
It looks like I need a triangular panel about 8″ wide and 10″ long. Since this is a removable test piece there is no need overthink it – just cut and hack one together!
Grab a chunk of carpet from the scrap pile and lay out the cutlines on the back. Hmm, need to do something about the raw edges. Don’t really want to try to deal with binding this piece. Why not just hem the edge? OK, add a hem allowance, fold it over, and sew the three edges. Looks fine!
Now to add the velcro tape. Easy to do on this filler panel, but a real job on the huge bottom piece of carpet. Just a minute – I wonder if the hook side of velcro would stick to the carpet? Yes, it does! All I need to do is sew a piece of hook velcro tape onto the bottom of the small filler pieces! This is looking better by the minute.
Since the velcro tape is adhesive backed it sticks to the carpet. Not enough for long term use, but it does a great job of holding the velcro in place for sewing. Sew the velcro to the carpet with one long seam and head over to the car.
Gas filler hump and triangular panel to cover it
A couple of minutes of trial fitting showcase the ease of re-positioning the panel with velcro. The end result is just what I was looking for! The hemmed edges of the carpet blend in – this looks much better than the bound edges. Running the velcro tape down the full length of each edge of the filler panel sticks it securely to the carpet. Success!
The carpet covers the full width of the trunk and extends from the rear of the trunk to the back seat. It also covers the wheel wells. Hmm, with two more carpet filler panels around the trunk hinges I think I can make this look OK…
Whip up two more triangular filler panels – with velcro! – trim and fit the carpet around the trunk hinges, apply and adjust the filler panels, then climb out of the trunk and study it.
“Finished” trunk. Triangular filler pieces can barely be seen under hinge and over gas filler hump.
This doesn’t look too bad!
Dig out the side panels and install them. Install the jack and spare tire. Add the spare tire cover. You know, this actually looks pretty good! There are some unfinished areas visible around the back seat, but the trunk itself isn’t bad.
Passenger side panel
As I stood behind the car nodding in satisfaction something in the corner of my eye caught my attention…
It’s with great relief that the interior is finally done! No more seats, door cards, trim, door switches, lights, or other tasks!
Yeah, about that…
I’ve seen some designs for custom trunks that I rather like. Basically the same kind of panels as the door cards – covered with vinyl, padded, and pattern stitched. I ordered so much material for the interior that I still have a fair amount left. Even after all of the mistakes and re-work. We all see where this is going…
Empty everything out of the trunk, including the carpet I put in a couple of years ago. Hmm, the surface rust is still there. Bummer.
You know the drill from the rest of the car: heat and scrape all of the undercoating off. Wire brush and sand the rust. Clean the entire surface followed by two coats of epoxy primer, sound deadening, and carpet.
Well, this time I got lazy. I just couldn’t work up any enthusiasm for doing all of this in the trunk. Yes, still had to wire brush the surface rust. I had several cans of rubberized undercoat left over, so hit the entire trunk with undercoat. That stuff is durable and sticks well.
The trunk is the last major surface without sound deadening. Order another box of sound deadening material and apply it to the entire trunk including the rear wheel arches. I took the car out for a drive half way through this job and it was already noticeably quieter.
At this point I hadn’t decided whether to go completely around the trunk with covered panels or just panel the sides and cover everything else with carpet.
In either case the starting point is panels that go against the side of the trunk from the wheel wells to the rear of the car. Start by making paper templates for the curved areas – mainly the corners. Once these fit, piece them together, trace them onto a scrap piece of plywood, and shove them into place.
Nope, didn’t fit. Go through several cycles of trimming and test fitting the panel until it fits properly. Quite a bit of fiddling was needed on the driver’s side – the jack sits in two brackets right next to the sidewall of the trunk and extends into the wheel well area. Several cutouts were needed to allow the jack to be stored in the factory location.
Once the template fit properly – on both sides, I’ve learned to be paranoid! – drag out a left over sheet of 3/16″ plywood, trace the template, and cut it out.
From there on it was basically the same as the door panels: mark and cut a piece of vinyl 1″ larger than the panel. Mark and cut a piece of 1/4″ sew foam the size of the panel. Use spray adhesive to bond the vinyl and sew foam together. Decide where to start the decorative grid both horizontally and vertically. Lay out the entire grid with erasable ink using the same 6″ grid that was used on the door panels. Sew the vinyl and sew foam together along the grid. Twice.
Panel cover with grid for stitching
Lay the sewn piece down on the workbench. Lay the panel board on top of it. Wrap the vinyl around the edge of the vinyl and staple it in place. Go around the entire panel, pulling and stretching the cover so that it is tight with no wrinkles. Flip the panel over and admire my handiwork.
Slip the panel into place in the trunk. Hey – it actually fits! As mentioned, I’m paranoid – so the driver’s side was done first. Dig out the jack and slip it into its brackets. Surprise! The jack actually fits! And there was much rejoicing.
Test fit of panel with jack
Lather, rinse, and repeat for the other side of the trunk making a mirror image of the first panel. Of course without the cutouts for the jack. This one also fit.
The last step was to apply velcro tape to the back of the panels and the side of the trunk to hold them in place.
With the panels done, pull them out, store them carefully, and start working on the carpet.
Plans for a future radio upgrade included adding a pair of speakers to the package shelf behind the rear seat. Time to actually install them while the interior was apart.
Turns out that there were already cutouts in the package shelf for standard 6×9 speakers – all that was required was to drill a couple of mounting holes slightly larger for all the screws to line up. One of the few cases where new parts fit easily into this old car!
After installing the first speaker I discovered that the center cone was sticking up just a bit too far. No problem – the speakers come with a cover. Except that I despise the look of aftermarket speaker covers under the rear window! To me it looks like someone took the easy way out on a speaker install. Bad enough on a muscle car and completely out of place on a luxury car.
No big deal. Just make a spacer for the speakers – I’ve done this before. A chunk of 1/4″ plywood, some quick surgery with a jigsaw, and the speakers are spaced.
As usual, the speaker instructions included a template for cutting mounting holes and making spacers. You can guess what happened as I got ready to start slicing up some plywood – you know, a 3D printed part would look cleaner…
To make matters worse, I’ve been taking a online course on Learn Autodesk Fusion 360 in 30 days. Lesson 2 covered how to trace a picture of a part to make a 3D model. I’m clearly required to apply my newly gained knowledge!
In addition to lines, circles, surfaces and solids, Fusion includes an entity type called Canvas that holds a picture. One of the features of a Canvas is that you can select 2 points on it and scale the picture to a known size. Scale the picture to full size and you can accurately trace around it.
With these speakers the template is on the back of the box. Known distance, known distance… Hey, if you include a steel machinists rule in the picture you will have a really accurate known distance!
Template for speaker spacer
Open a new Fusion design and create a Canvas from this picture. Click on the two ends of the machinists rule and specify that these are 6″ apart. Considering the accuracy of the rule and the precision you can achieve selecting the sharp end of the rule this Canvas should be accurate to within 0.010″ – plenty for this case. Using the usual trace and cut approach I would have trouble doing much better than 1/16″.
Now, how to trace this part? It looks a lot like an ellipse – maybe it actually is? One way to find out!
An ellipse is defined by three points: center, long axis, and short axis. If long axis and short axis are the same length you have a circle.
Unfortunately the template didn’t mark the center point. But the four mounting holes look like they are evenly spaced… Create a 1/4″ circle and precisely position it over the mounting holes. Once you zoom in you can do this very accurately.
Diagonally connect the circles with two construction lines. Where these lines cross is the center of the ellipse.
Bring up the ellipse tool, select the center, select the long axis, and drag the ellipse to the short axis.
Huh! Look at that – the ellipse perfectly overlays the template! No need to worry about using line segments, circles, splines, or any other complications. Create the outer ellipse the same way and done. The design process took about 5 minutes – and 3 minutes of that was looking up how to create a Canvas!
Extrude the design 1/8″ to create a solid and send it to the 3D printer. 45 minutes later grab the part off of the printer and use it to mount the speaker. Fits perfectly!
Print out another spacer, mount the second speaker, connect the wiring, and this task is done.
This job was so simple I almost didn’t write about it. But the approach of designing by tracing around pictures is so powerful I had to share it. I expect to use this more in the future!
I have to do a plug here for the YouTube channel Product Design Online. I’ve reached the point with Fusion where I’m comfortable with the basics and ready to learn more. To my surprise I started learning new things I could use in the very first lesson – and continue to learn more in each following lesson! Highly recommended.
The power window switches on old Chryslers have a fatal weakness – the spring clips that hold them in place break easily.
The window switches are inserted into the door cards. Every time the door card is removed the switch must first be removed – the wires aren’t long enough to allow the door card to be removed with the switch in place.
Power Window Switch
The switch is held in place by four steel clips. The design of the clips should be fine – this same design is used in many places with good results. Unfortunately, the Chrysler clips seem to be made of mild steel rather than the expected spring steel. After a few cycles they lose their grip. When you attempt to bend them back to their original shape they break. Most of the time it doesn’t matter – few cars have the door cards removed even once. On a restoration like this, on the other hand, you will start breaking the clips.
Well, not a major disaster! If one breaks you still have three left – works fine. Ummm, OK, after two break you still have two. Make sure they are on opposite sides of the switch and it still works. Yeah, one clip doesn’t hold the switch in place at all.
These switches are hard to find. The clips are complete unobtanium.
You saw this coming: the last time I had this door apart I broke the third clip. Yup, I’m totally up the excrement waterway without a manual propulsion device.
Extensive cogitation didn’t uncover a way to make new steel clips. However… What about 3D printing a switch retainer? Didn’t see that one coming, did you??
The problem is that the hole the switch fits into is a really tight fit. I couldn’t come up with a design that would hold the switch, fit into the hole, and include a retention mechanism.
If only you could 3D print rubber. It looks like it should be possible to design a sleeve for the switch with a couple of bumps that would squish when inserting and hold the switch in place.
You can’t print with rubber, but you can get TPU filament. TPU, or Thermoplastic PolyUrethane, is a flexible material with many of the characteristics of rubber. Time to order a roll of TPU and fire up the design system!
The concept was a sleeve that would go over the switch with bumps on the inside fitting into grooves in the switch body and bumps on the outside to hold it against the edges of the hole it fits in.
As mentioned the hole is a tight fit. OK, do a preliminary design with 0.010″ thick walls and run a test print. Resulting in complete failure – walls completely too thin for the printer. What about 0.020″? It is a bit too large, but maybe the TPU is squishy enough to be forced through the hole? This part printed successfully, so work on a “final” design and load up the TPU filament.
Left: prototype in PLA. Right: final part in TPU
With the TPU part in hand give it a close inspection. The part is firm but flexible and seems to be pretty tough – TPU lives up to its billing.
First question: can I install it on the switch? And will it stay in place?
Switch with sleeve
Yes – it fits snugly over the switch and locks securely in place. OK, easy part done. The big question: will it fit into the hole?
Nope. Side to side seems like it could go, but top to bottom is just too tight – won’t go without peeling off the sleeve. Krud! So close, so close…
If it won’t go in top to bottom, what happens if I cut off the face on the top? Is the remaining part rugged enough to hold together when being stuffed into the hole, or would it just fold back?
Turns out that you can now slip the switch with sleeve into the hole, right up to the point where it hits the retaining bumps. Cross fingers and discover that you can’t apply any pressure with crossed fingers. Uncross fingers and wiggle and stuff the switch into the hole…
It goes in! It doesn’t come back out! And there was much rejoicing!
Once again modern technology comes to the rescue. Leaving only one question: what can I 3D print next?!?
I’m moderately proud of figuring out a way to use modern LED technology to replace the 60 year old electroluminescent technology in the HVAC (Heater Ventilation and Air Conditioning) controls. On the other hand, the mounting for the dimmer control for this was not one of my prouder moments…
I had simply zip tied the control module under the dash. In a visible location. Because there wasn’t an hidden place to mount easily it. As I’ve mentioned, for such a large car there is very little space in many areas. Like under the dash.
With the rest of the interior coming together and looking good something had to be done about the HVAC LED dimmer.
When starting the HVAC lighting project I actually ordered two dimmers: a packaged, ready to use unit, and a more compact modular unit that consisted of a 1″x1″ circuit board, connecting cables, and rotary control knob. While I wanted to use the modular unit it was easier to use the packaged unit for prototyping.
The biggest problem, of course, was finding the modular unit. After extensive searching I found it just where it should be – in the box with the rest of the LED components…
You know that old saying if the only tool you have is a hammer, every problem looks like a nail? And my personal favorite: a big enough hammer can drive any screw.
Well, with a 3D printer suddenly any problem looks like something that can be solved with a custom part!
Studying the underside of the dash it looked like the perfect solution would be to extend the custom bracket holding the updated fusebox. I had been wanting to do this anyway – this was the first fusebox bracket I did and the cutout for the fusebox itself had been gouged to “close enough” size with hand tools. Later brackets used laser cut parts from SendCutSend which are a much better fit. And I had an extra blank.
OK, let’s get this party started!
First of all, I wanted the LED to be a bit dimmer than the lowest setting on the dimmer control. I noticed that changing the brightness of the dash lights also dimmed the HVAC LED. Putting on the Electrical Engineer hat that I don’t have, this implies that I can make the dimmer dimmer by adding a resistor to the power supply wire. A 2 ohm or 5 ohm resistor should do the trick. Time to head to Amazon!
Although the LEDs don’t draw much power I was concerned that the widely available (and cheap) 1/2 watt resistors wouldn’t be enough. A bit of searching turned up an assortment pack of 5 watt resistors at a semi-reasonable price.
Without being sure what was needed it made sense to test the resistor circuit before wiring it permanently. Easy enough to do – grab a couple of jumper wires with alligator clips. One wire from power to resistor and the second from resistor to dimmer. Hook it up, wait for night, turn off the workshop lights, wait for darkness, and verify that I had the correct value.
Hmm, not much difference in brightness with the 2.2 ohm… OK, the next size up is 3.3 ohms Still not enough change. Well this is why I got the assortment pack! Grab the 10 ohm. Better, but still not quite what I wanted. 22 ohm? 47 ohm? Finally at 100 ohm it was close to the (lack of) brightness I was looking for.
This was much more resistance than I expected – glad I experimented first! Discussing this later with a friend who is a real electrical engineer, he told me that he sometimes has to go as high as 500,000 ohms in similar circuits. Huh, learn something new every day! Guess I should have tried 1,000 ohm and 10,000 ohm resistors while I was at it.
Now to design the case to hold all of the electrical components. Complex designs are basically a bunch of simple designs connected together. In this case I needed a mount for the circuit board, a holder for the resistor, a baseplate for these, a top plate to provide protection, and four columns to tie everything together.
And that is exactly how I designed it: a mounting plate for the circuit board that picked up the screw holes in the circuit board and was extruded to provide space for the components sticking out of the circuit board. A pair of rectangular pillars with holes for the tubular resistor. These were split in half forming a saddle so that the resistor could be inserted and then firmly clamped in place. Next was the baseplate with both of the mounts integrated into it and round columns at each corner. Finally, a top plate that screwed to the columns and which included the top half of the resistor mount.
Print it out, test fit everything together…. And as usual discover a few things that needed to be tweaked. Absolutely normal. The amazing thing about CAD and 3D printing is that you can do this quickly, easily, and cheaply. For example, these parts used about $0.25 of material.
Dimmer Module Box
With the housing done, turn to the bracket. Remove the original bracket from under the dash and use it as a reference. Decide how much it needs to be extended to support the dimmer box and dimmer knob. Mark this extension as well as the bends for bolting the bracket to the dash on the fuse box bracket panel.
Drilling the mounting holes for the dimmer box was easy – take the top plate and use a transfer punch to mark the center of each mounting hole and then drill them. Determine where the control knob should go and drill a hole for it. Drill a hole for the mounting bolt that will secure the bracket to the bottom of the dash.
Bend the bracket so it will mount properly and weld a nut to it. With an earlier version of the bracket I used a separate nut – trying to hold the bracket and loose nut in position while starting a bolt proved to be a nightmare. Test fit the bracket. It fits, so hit it with a couple of coats of paint.
Really should clean up the dimmer box before final mounting. There are four wires on the dimmer – two for power in and two going to the LEDs. Add these wires, sleeve them, and add a connector to the end. Add a matching connector to the wires in the dash. This will simplify installation and any future maintenance on the dimmer circuit.
OK, to be perfectly honest, I’m turning into an electrical snob as I learn more about wiring.
Dimmer Module ready to install
Bolt the dimmer box to the bracket, mount the control knob in its hole, and we’re finally ready for installation. I had an extra fusebox from another project, so I temporarily mounted it in the bracket to make sure everything fit.
Dimmer Module mounted on bracket with reference fusebox
Moment of truth time: slip the bracket under the dash, position the fusebox in its cutout, and bolt the assembly to the dash. With the mounting bolt secure, run in the four machine screws that fasten the fusebox to the bracket. Plug in the connector for the dimmer module.
Fortunately it was late enough that it was dark. Turn off the workshop lights, turn on the Imperial headlights, and adjust the dimmer knob. Everything worked!
And there was much rejoicing!
The dimmer module is now securely mounted, completely hidden under the dash, yet the control knob can be easily reached to adjust brightness. I’m declaring this a success and moving on to the next project.
Introducing the Imperial Deathstar, a black 1963 Chrysler Imperial. This is one of the largest production sedans ever built, and arguably the best luxury car of its day.
Join me what will probably be a never-ending saga of grease, aching muscles, and an empty wallet as I work to restore this over 50 year old survivor to a reliable cruiser.
