We have been doing spatchcocked chicken, turkey, & cornish hens for a few years now. This past year we started getting food from Imperfect Foods as an alternative to going to the store. One of the times, I missed that there was a butternut squash on the order and it came. I am NOT normally a squash fan. In my family, it was always sweetened and served as squash mush. However, one day, we had the extra squash, some monster shallots and a whole chicken (and it was decidedly not nice for grilling outside). So I decided to try using the squash as the star under the bird rather than the more traditional stuffing, potatoes or root vegetables. It was a hit!. We have repeated this a few more times to make sure it is a repeatable recipe and we still love it.
1 butternut squash, peeled, seeded and cubed into 3/4″ pieces. COmpost the stringy stuff but add the seeds back in with the cubes
2 large shallots coarsely chopped or 1 medium onion and 2-4 garlic cloves chopped
1 lemon (optional) cut into 1/8ths with the skin (you may also want to add a few pinches of oregano in this case as well)
In a bowl place the above items and sprinkle liberally with granulated garlic, Penzey’s Seasoned salt and fresh ground black pepper. Mix well and dump into a oiled (or non stick sprayed) 13×9 or better larger glass baking dish. Keep the pieces towards the center as you want it all under the chicken.
Take a 3-4 lb chicken, clean out the cavity, remove excess fat and slice down the backbone and also the keel bone. Flip over and press flat (it should make some vicious crunching noises in the process) .
Place the chicken on the veggies. Spinke with more granulated garlic, Penzey’s seasoned salt, pepper and some rosemary sprigs
Bake at 400 F convect for about 45 minutes, until a probe in the thick part of the thigh reads 170F. Pull from the (now well splattered) oven and let rest for 15 min. Beat off the crispy skin thieves with a thick spatula (BTW the skin off an entire leg does not count as a schnibble). Slice and serve.
The table was modeled completely in Sketchup. This helps to not just plan the work but then I can export individual components such as the legs as input for the CNC router.
The leg component was exported and then imported into vCarve Pro which I use to model items for the CNC router. I was able to fit 2 legs per plank, working around any defects in the wood I had. During the preparation of the profiles to cut tabs are added and adjusted. These will hold the leg in place during the final pass at full depth. I also set it to take the last pass slightly smaller to do a full hight clean up of the leg profile so there are no lines visible form the individual passes. I used a 1/2″ diameter 3″ cutting length solid carbide 2 flute endmill to do the cutting (same as I used on the crib project). Below you can see highlights of the CNC cutting process.
After the legs were cut out, the tabs were trimmed off with a small saw and filed flush. Now it was time to make the blocks that join the legs and stretchers. The blocks are angled at 40 degrees to provide the splay of the legs. These joints are a high stress area in the design. So I did not want ot just have a butt joint and decided to use floating tenons. The joint is 5″ tall, and given that this is sort of a long grain to side grain joint, I decided to use 2 tenons per joint. I happened to have some stock pre-made for the floating tenons (left over slats from the crib project) .
The corner blocks are 2.5″ wide , 1.25″ thick and 5″ tall. The first step after squaring up the stock, is to cut the 40 degree bevels on the table saw using a hold down push stick.
The position of the mortises is laid out on the ends for the legs.
Now the router is set up in the router table with a 1/2″ carbide end mill. The fence is adjusted to match the router bit teeth to the marks on the end of the stock.
Now a test cut is made in the block to show if the cut is in the right position. Note that the blocks are still extra long at this point.
The fence is marked with the leading and trailing edges of the router bit using a squared-up block and pencil. This is much more convenient than using a square.
Once the test cuts show the bit & fence are positioned correctly the blocks are now cut to length and the positions of the mortises are marked for length. The piece is moved to the start and stop points for the cuts and the marks are made on the fence to register the ends of the piece for the start and stop positions. Each mortise should be cut in 2 passes to make it easier to tilt / plunge the piece onto the router bit and minimize burning. Note that the top corners of the feather board are knocked back to enable easier plunging. After all of the leg mortises bare cut, The fence is readjusted for the position to center the mortises for the cross bar.
Next is cutting the mortises in the legs. Basically it is the same process. The marks for the bit position are placed on the table top instead of the fence.
Now the stretcher is cut to size. The tenon thicknesses are cut most easily on the radial arm saw.
Next the tenon lengths are transferred to the tenon and the cuts are made with a Japanese style hand saw or dovetail saw.
The waste in between the tenons is then chopped out with a chisel, similar to doing dovetail pins. The corners are then rounded off with a chisel and touched up with a file for final fit.
The slip tenons are cut to length and everything is test fitted into the corner blocks.
The legs now get the edge profiles added after a light pass on the drum sander to remove the facets on the curves. . FOr the outside / front edges a 3/4″ round over bit is used. This took 3 passes to avoid tear-out and burning and sneak up on the final profile. The back edges and feet were rounded over with a 3/16″ round over bit.
Now it was time for the first dry fit assembly. A ratchet strap is used as the clamp to hold everything together. I will do the same for the final glue-up later as well.
Next will come the table top supports and slide mechanism.
The wood was broken down into usable lengths, run through the jointer with the power feeder and then planed. It became evident that there was an almost even split of “light” and “dark” cherry pieces. So the next step was sorting them out. Teal had asked for a lighter top and at one point we were considering hard maple. So, the light pieces were selected and then further sorted and matched for the top panels.
As you might expect some were rejected for use in the top, due to bad imperfections (tear out, knots in the middle, etc). The rest were matched up and the edges jointed, as well as being ripped to maximum usable width per piece. This machining / matching / culling continued for a few rounds until I had what looked like enough boards for the 3 major panels for the top. The 2 end panels are about 36″ wide and the center about 24″ wide. The pieces were then cut to rough length – 44.5″ to allow for trimming and removal of the hard to eliminate jointer snipe at the end of the cut. Note that some sources say that adjusting the outfeed table of the jointer can eliminate snipe, but this is wrong. The goal of the outfeed table height vs cutter blade height adjustment is to have a straight edge. Matching the outfeed table to the highest point of the arc of the cutter knife is key to straight edges and the first phase of a good glue up. However, it does not always eliminate snipe whether on my previous 6″ jointer or my current 135 year old 16″ Colladay monster of a jointer.
After jointing, ripping and frequently jointing the ripped edges, lay the boards out for best grain and color match as well as trying to alternate the ring direction if possible. Mark across all of the faces, across the grain, in pencil with a large V or Triangle. Th is is to keep the pieces in order and reduce the chance of an unplanned flip of a board during planing or glue up. The the next step is to fine tune the edges with a hand plane. Yes, you will find claims of “glue line ” rip blades but this does not work reliably for much more than 20-24 inches. Wider pieces benefit immensely from (optionally) running on the jointer and hand planing for a perfect fit and invisible glue line. When hand planing for final fit you want a #4 to #6 plane. My favorite is a Veritas low angle jack rabbet plane set for a very fine .002-.006″ translucent shaving across the width of the blade. Yes, this is 1/2 to 2 sheets of paper thickness. The regular low angle jack would work as well, but I can only afford one of these, and this one is also great for truing up tenons as well.
Next best choice for a plane is typically a old Stanley or Record #5 or #6 with the frog moved up and the mouth set very tight. Having an even shaving across the width of the blade is very important. The goal is to remove the “hills” from the joint without adding any new slopes and gaps across the width of the joint or inducing tear-out. A firm, steady stroke, evenly applied across the width of the board is required. It usually takes a few tries to get a perfect fit. I do prefer to sneak up on the fit rather than risking a bad gouge and tearout. Being able to have finely set planes for this sort of joinery and others for regular “rough” planing is reason enough to have 2 of each size for your favorites (typ. $35-55 each at flea market, garage sale or ebay).
Once the joints are good, now is the time for the glue up. I prefer Titebond III as my go to glue. Not only is it water proof, it offers a few minutes more working time than Titebond II or ordinary PVA such as Elmer’s. For the bottom clamps, Bessey K body or the Jorgensen equivalent are ideal (but 3/4″pipe clamps will work too). You will need a large flat work surface and the best I have is the tablesaw for pieces this size (~36×44″). Apply the glue evenly to each side of each joint and then just barely tighten the boards in the clamps. Next place clamps vertically across each joint – not super tight yet. This will hold the ends in alignment. Working across the glue up, even up the middle of the boards, pounding form the top or pressing up from the bottom and then tightening the clamps a bit more. Once the board edges are even it is time to place the alternating top clamps. Now work your way around and progressively tighten the clamps and keep checking the joint alignment and tapping or pounding the high boards into submission. If you wonder why there are top and bottom clamps you have probably not yet seen a “board fountain” when clamping from only one side, and then while tightening, the whole thing sort of explodes upwards due to the clamp bars curving under pressure. Ideally, you want to be able to run your finger across a joint and not feel it. A bit of care now will save a lot of planing and scraping exercise later. Flattening a table top by hand can count as aerobic exercise!
Next, will be cleaning up the top panel joint lines as well as starting on the base / leg joinery.
With the back and doors assembled, now we can start on the “dresser drawers”. The drawer fronts are solid stock. I chose maple as it can be finished clear and provides a pleasant contrast to the planned light grey / off white color of the paint.
Check your actual dimensions vs the drawings. The 3 drawers when stacked should have 1/16-1/8″ gaps on each side and between the drawers. Rip the drawer fronts to width. Then cross cut them apart for height. The goal is to have the grain flowing across all 3 drawers if possible. In my case I had some long narrow pieces of maple left over form the cabinet project and used them. Cut the rabbet for the bottom panel in the lower edge of each drawer front. Then cut the side panel rabbets. These rabbets will cover the ends of the plywood when the drawers are closed and provide ample glue surface as well as a self-squaring reference edge during assembly.
Rip the sides from 1/4″ plywood and the back form 1/2″ plywood or solid stock. The sides are 1/2″ shorter than the drawer front minus the 1/4″ bottom panel. This is to allow space for the drawer runners/kickers which will be glued to the cabinet sides.
Lay out the parts for the glue up. I use thick super glue / cyanoacrylate. Having an assistant spray the accelerator once you have the pieces in position is a huge help. This avoids all but finger pressure clamping.. I attach the sides to the front as the first step (use accelerator). Then the back, and finally the bottom and once the back and bottom are in place and the drawer is square, then hit it with accelerator.
The next step is to place the runners (1/4″x 1/2″) in the case. This is a little bit fiddly to get the spacing just right. Since I was doing mulitples, I made spacers to aid in gluing in the runners (more super glue). For a single cabinet, marking the lines with a square would be sufficient.
Prior to painting, break all of the corners with sandpaper or preferably a 1/16″ radius router bit. Softening the edges gives better dent / ding resistance and just feels nice. Any gaps or edge voids should be filled with Bondo. Make small batches as it cures fast. Sand everything to 120 grit.
Cut the 1/2″ dowel for the closet rod to length. Make the closet rod support blocks from scraps (~1.75″ tall x 1″ wide x 1/2″ thick ). For the hangers we have, a 1″ space above the rod seems ideal. Drill the hold for the rod in the support blocks a bit oversize (e.g. 9/16) to ease assembly after painting.
Paint should be a water based enamel. We used Benjamin Moore INSL-X Cabinet Kote which is very durable. I do NOT recommend the BEHR enamel / trim paint as it is too soft.
Once painted the finishing touches can be applied. The closet rod is installed with CA or hot melt glue. You want the blocks to slide easily so that you can get the glue on both blocks and everything slid into place prior to the glue setting up. Similarly, the mirror is held on with hot melt glue as well. It is centered in the opening.
Now it is time to install the hardware. Start with the outside corners. Next add the hinges and latches. Then the handle and the bottom feet. All screw holes should be pre-drilled. The tiny screws easily strip or snap off.
The feet were 3D printed using TPU filament. The Sketchup and STL source files are available at: https://www.thingiverse.com/thing:4723078. Note that for TPU you need a direct drive extruder.
Materials and suppliers
(4) 25mm x 9mm BP Filigree Corners
Pr 1-1/8″ x 1-3/16″ Snap Clasps
Pr Brass Case Handles
1/2″ diameter knob for top and middle drawers #37663
My grand-daughters were getting 18″ – “American Girl ” style dolls for Christmas. My daughters asked that I build doll cases for them. They had some suggestions on how they should look based on a discontinued one from Pottery Barn Kids and another on Ana White’s website. One of the goals is that these would be able to be passed down to future generations much like the Pencil Post doll bed I had made for Elyse, as well as other accessories that had been saved. So, they need to be rugged as well as pretty.
The ruggedness drove the use of 1/2″ baltic birch plywood for the sides. This will provide a strong base for the hinges, latches and handle. The “fronts” and “back” are 1/4″ plywood to save weight and these really don’t have any hardware fastened into them . The vertical “closet” divider is also 1/4″. The top of the “dresser” is 1/2″ but you could probably substitute 1/4″ as well. The sides could easily be solid wood, but I had the baltic birch left over from making kitchen drawers. The girls wanted these to be painted rather than stained. So this also opened up some more flexibility in the joinery choices.
The joinery is simple with all rabbets and dadoes. The rabbets make it relatively self-aligning and ease single person glue ups. The dadoes are just for the inset dividers, so that there is no need for additional brackets / gussets.
After ripping the sides, tops, bottoms to the same width of 5.5″ , then cut the pieces to length. The sides are 21″ long, the long / back top and bottom are 17 1/2″ long and the door top and bottom are 8.5″ long.
Next the rabbets for the fronts and back are made. These are 1/4″ wide (across the thickness of the plywood) but the depth will depend on exactly what thickness plywood you have for the fronts and back. It could range from 5mm all the way up to 1/4″ (6.3mm). Cut these rabbets for the fronts and backs first on all of the sides, tops and bottoms.
Ideally the rabbets are all done with a dado blade. Bury the blade in a sacrificial fence for the back panel rabbets and then push back the fence for the top and bottom (approx 1/2″ wide) rabbets.
The last cuts are for the dadoes. These will likely be a different width. With a dado head this means removing, adjusting and and then resetting the height. The key thing to watch for is that the dadoes must be equal to or slightly shallower than the depth of the back panel dadoes to ensure that they will not show on the exterior, once it is assembled. The test cuts below illustrate this.
The dadoes for the ~1/4 ” panels are likely need to be narrower than a dado set can go. So you will need to make the first cuts, move the fence make more test cuts and proceed.
With all of the case pieces cut our you can now start to assemble. What is critical is that the 3 case pieces are square once assembled.
The assembly for the “front doors” is similar. However be sure to dry assemble them and make sure the pair is not wider than the back. A small gap is OK and even desirable i.e. both doors together 1/16-1/8″ narrower than the back.
When designing these cases, I was doing it the “old fashioned” way with my gridded notebook, pencil and ruler (can’t forget the eraser as well). The sketches below are photos as my scanner apparently died recently.
Well I am committed now. We have discussed making a new dining table for years. However it had not risen to the top of the project heap. Plus, it involves not just making the table but also a matching set of chairs.
The goals include:
Seat 6 when collapsed
Seat 10 when expanded
More width as the current table leaves little room for side place settings and a center piece
One or 2 leaves, preferably butterfly style – self hiding / storing
NO corner legs. This is one of the biggest complaints with the current Scandinavian / Dutch style with pull out end leaves
Reasonably resistant to pen / pencil denting, when used by the grandkids or errant uncles
So over the last week I researched ideas and ran a number of concepts past Teal. She did not like many of the trestle styles as being “too heavy / too much wood”. I did not want a single post style pedestal and anything with 6-8 legs is definitely out as going counter to the “no corner leg” goal.
Finally, I homed in on one that we both liked. However it has a number of construction challenges:
The angled and swooping legs will be difficult to cut and join securely. Shaping will be done on the CNC router
Desire for an even more graceful sweep to the leg (rather than straight “feet” which the original had. You can see the swept feet shown in the attached photos.
An alternative needed for the commercial equalizer slides. These allow you to pull one end and the other slides out as well. Plus I want heavy duty slides that will not sag over time.
Better fit needed for the the butterfly leaf than shown in the photos and videos
So I started modeling it in sketchup while continuing researching the problems. The third leg design I tried looks promising. The joint of the legs and stretchers is a potential weak spot. So, I beefed this up vs. the commercial example. This means the legs will not directly meet when viewed from the end. There were some good clues on LumberJocks to handling the slides and I will probably go with Accuride 600lb heavy duty slides (9301E) and aircraft cable for the equalizing mechanism. The GT2 timing belt method is too expensive and the strength specs.
There is still more final drawing / modeling to do for the internal works, but the over-all approach seems doable. However, I am now committed with several hundred pounds / couple of hundred board feet of Cherry wood (and several hundreds of dollars spent) now sitting in my shop. There are still more design decisions:
Do I arc the table top edges or leave them straight?
Should I undercut / bevel the top?
What final profile for the top edge?
How much of a bull nose radius on the outside of the legs? (I am having a hard time modeling this in Sketchup).
Should I soften the inside edge of the leg or leave it square?
The key to this recipe is to use both Sous Vide and grilling. Sous Vide properly cooks the chicken keeping it juicy and tender. Grilling adds the smoke and crispiness.
We started with a big economy family / restaurant 10 pack of legs and thighs. This was split into 3 batches. Trim off any large chunks of fat and tail if present. These can cause flare ups while grilling. I leave the legs and thighs attached, but you could separate them for easier portioning for kids.
Season the chicken generously with granulated garlic, Penzey’s seasoned salt and Italian Seasoning. Place 3-4 in each bag. We kept one out for cooking today and froze the other 2 bags for later.
Sous Vide at 158F for 4 hours. Add 20 min if frozen.
Chill the chicken while still in the bag in ice water or throw in the snow bank for an hour. You could also do this a day or 2 ahead as well and refrigerate.
Heat grill to hot (~350-400 as indicated on the BGE). Too hot, and you just burn the skin off.
Remove the chicken from the bag, pulling off and saving the juice / gelatin. Make sure it is all off or when grilling, you will have an immediate flare-up.
Place on grill meat side down and grill for 3-5 min. Flip to skin side down and grill until crispy – approx an other 3-5 min.
In the winter, don’t forget to re-insert the sheet metal ring when closing the BGE. This reduces the occurrence of the gaskets from freezing the lid shut. If it does freeze shut it helps in melting it open. See: Cold Smoking and Frozen Smoker Entry from more info.
The first of the 3 Lonzino pieces from Lonzino batch 4, that I started at the beginning of November is ready. For the first 5-6 weeks the drying chamber (Rubbermaid bin with fan for humidity control) was in the garage. Once it dropped below freezing outside, the basement was now cool enough (55-60F) to bring it to the basement. Initial weight was 1122 g and final was 585g. My target was 45% weight loss.
It is delicious. This is the one that I spiced with Penzey’s 33rd and Galena (by eye). The spices were applied at the same time as the curing mix. I think the high humidity of the last couple of weeks, which was due to adding the Sopressata and Pepperoni to the chamber helped avoid any case hardening.
The next piece should be ready in a week or so, it is just over 40% loss. The last one is being slow so it may be another 4 weeks.
I started this batch with “recalibrating” / adjusting the target setpoint the Inkbird humidity controller and adding monitoring using an Ecowitt temperature and humidity sensor and gateway to log to the ecowitt.net website. Now I can see how the temp and humidity tracked over time. You can see the temperature shift as I brought the bin ito the basement and then later the humidity shift as I added the sausages just before Christmas. I have more of the Ecowitt sensors in the garage fridge and basement freezer (which is why this is Channel 2).
Note commissions hopefully earned on the Amazon links
I have been wanting to get into making fermented sausages such as Sopressata, Pepperoni, Salami and more. My sausage making has been rather limited as Teal is not a big fan of “tubular food” and each time I made sausage (summer, brats, Italians, etc.) it in the past it was painfully slow , so the gear would get set aside for another year or 2. I have both the meat grinder attachment for the Kitchenaid and a Kitchener #10 hand crank grinder. Both are slow to use.
Recently, I was making Sopressata with help from my grandson. However grinding in the hand crank grinder was too much for him this time. However he is a fan of the sausage samples we fried up and commercially made Soppressata.
Even for me it was difficult with the grinder constantly slipping on the counter and the blade and disc/plate getting gunked up. After grinding, I gave up on the hand grinder for stuffing and used the Kitchenaid (slow going). Stewing on this for few days, I was thinking whether I had to get a new grinder and stuffer or does the existing gear just need a tune up? The clue I needed was one of the grinder vendors recommending to send the blades in to them for periodic sharpening. I thought, this would be simple to do myself and then the light bulb came on: I had not done anything with either of the meat grinders! Basically, I was just using them “out of the box”. It appears that this is the primary problem!
I tune up and sharpen all of my knives and hand tools (chisels, planes, jointer, planer, etc). on a regular basis.When I get a new (or new to me- used) plane or chisel I spend the time to flatten and tune up the new tool. On closer examination of the Kitchener meat grinder there is a list of problems:
The discs / plates are not flat. Actually both are concave on both sides (how this would happen I do not know). They are not worn by any means, just poorly manufactured. You can see the coarse factory grind marks and no real sign of wear from use.
The knife is far from sharp
The plastic spacer they provide to go between the feet and the counter is some sort of polyethylene – slippery. We need something that will grab.
The feet are far from flat and have a very limited contact patch near the inner corners. This can be seen in the photos with the 2 small wear areas.
On the plus side the rear plastic bearing has a reasonable fit and enough thickness to push the blade against the plate.
The first step was flattening of the plates. You can use diamond stones 180 grit and then 300 or silicon carbide (wet or dry) sandpaper on the countertop (or glass plate or table saw or jointer) as shown below. Just wet the sandpaper on both sides and it will stick sufficiently for the flattening. If doing this on the counter, stay away from the front edge which may have some dish to it. It may take 20-30 min to do the initial flattening if they are poorly made (as mine were). However, subsequent sharpenings take but a few minutes.
With the plates flattened, now it is time to sharpen the blade. For this you need something that can reach all the way into the corners. The diamond plates work well or you can use inexpensive small diamond sharpeners (often sold for sharpening fishing lures) . Aggressively sharpen the bevels, then do the flats the same way as the discs. It now actually feels sharp. So now this is getting promising.
Next was to address why does the grinder slip so badly on the counter? I threw out the stupid polyethylene spacer (which is slippery) that came with the grinder and used a rubber pad of the type for opening stuck jar lids, but this still did not help a lot. Looking more closely I noticed that there are 2 very small wear / contact points on the feet. They are FAR from level.
The feet need flattening. Given that this is cast iron, a file will do. Use a 10 or 12″ mill bastard file and carefully file the 2 feet flat. You must make nice parallel strokes or you can easily end up rounding the feet rather than flattening. Check that they are flat with a straight edge such as the blade of a square.
When researching other grinders, I saw the LEM #10 had a plastic/ rubber boot over the feet. This makes a lot of sense. So I drew up the foot and clamp pads in Sketchup and then 3D printed them in TPU (Thermoplastic Urethane) which is a flexible filament. I am using EOLAS brand TPU+, one of the softer ones for the foot and clamp pad, which they also claim is food safe (also great for gaskets).
With these changes in place, the grinder works far, far better. It grinds much more easily, does not clog as often, and no longer slides on the edge of the counter at all. I am still thinking about a proper sausage stuffer however.
The sketchup and STL files for the pads and a stuffing spider are at: https://www.thingiverse.com/thing:4695351
Now that the lights have all of the pieces soldered together and cleaned up, it is time to cut the stems and get everything powder coated.
You need to be sure you have calibrated your oven and are sure of the actual temperature vs what is set. Additionally the parts should be placed to the sides, and not in front of the fan. The problem is that the powder coating cures at temperatures very close to (if not exceeding) the melting point of the solder. So it is very possible to have your assembly fall apart or significantly change shape while curing the powder coat. Below is a chart I pulled from the Kester web site: https://www.kester.com/Portals/0/Documents/Knowledge%20Base/Alloy%20Temperature%20Chart.pdf As you look at the chart, the solders we have commonly used for electronics, stained glass work and plumbing in the past are referred to as the “mid temp solders”. Note that the lower end of their melting range is 361F. The more modern lead free plumbing solders that I have run across are the Sn97Ag3 , and SN96.5, Ag3, Cu0.5. These have melting points above 420F. This gives us a TINY bit of wiggle room above the curing temperatures required for the powder coat.
So far I have used powders from Eastwood and Prismatic Powders. The Eastwood powders typically recommend 460F until the powder starts to fuse and then 400 for 20 min. This does not work with solder as I demonstrated with my first test piece. The solder joint (60/40) opened up during the curing. This lead to some more research and there were a number of posts that said a powder that cures at 400F / 200C for 10 min (as the Prismatic powders I was contemplating) could also fully cure at 355F/180C for 15 min. Note that this is not from Prismatic (I should really contact their tech support for confirmation). I ordered the candidate Prismatic powders and ran some samples. Sure enough, the powder seemed to cure very nicely at the reduced temperature (nice gloss and the metallic flakes glistened). My oven was calibrated against a Thermoworks Chefalarm and an IR thermometer. It runs +5 to -20F vs the set temperature once warmed up and allowed to soak for 15 min. I also decided to avoid placing any parts directly in front of the fan but rather off to the sides. I assume that the hot air coming from the fan is likely hotter than the “overall” oven temp. I had switched to using “lead free plumbing solder” to assemble the parts. The chosen powder coat was Prismatic’s “Soft Misty Copper UMB-1352“. So now was the time for the test of a complete lamp. This was the prototype path light. Everything held together and the finish was great.
So this proved that you CAN powder coat soldered parts. However, you MUST be careful in you choice of powder, solder, and oven calibration. There is little room for error and oven temps often swing +-20 to +-30F with greater swings during warm-up or when re-heating after opening and inserting the parts. It may be a good idea to line the bottom with fire bricks or tiles for more thermal mass, but I have not tried it. I did get all 8 lights powder coated without incident.
Prior to powder coating, the parts need to be bead (sand) blasted. I am using 80 mesh glass beads at 45-50PSI. The reason the pressure is so low is that the beads will fracture – turning to dust if the pressure is too high (increasing your materials cost) and the brass will deform under the pressure of the blast. This is most notable with the brass U which should be blasted at an oblique angle. Blasting straight on, can completely deform and close the channel. Once the parts are bead blasted they are then blown off with an air gun and hung on stainless wires from the oven rack.
The process of bead blasting and powder coating can be seen below:
While my powder coating setup is a bit low tech, I am pleased with the results .
The lights get a bipin socket for the LEDs hotmelt glued into the base plate . Extension wires are soldered on. I used 5 watt LED “corncob” style lamps. These run off of standard 12 VAC low voltage wiring (and are darn bright).
The posts / standards are 21″ long 1/2″ copper pipe. This was the longest length that would fit horizontally in my oven for powder coating. They are rather tall, but this is needed as my old lights would repeatedly get buried in snowdrifts in the winter. The “anchor post” was made from 1/2″ galvanized iron pipe. I used 24″ sections, cut in half, ends split and then pounded to some semblance of a point. The same 1/2″ copper to 1/2″ threaded adapters that were used under the base of the light were also used on top of the stake. It is not necessary to solder the posts together. The friction fit with a bit of powder coat overspray is more than tight enough.
I found some new screw together vampire tap connectors that worked out very well to connect into the existing 14 gauge cable that was left from the old lights. These are far better than others I have used in the past.
I think the finished lights look great and are a definite improvement over the several previous light sets.
Please note that I may get some revenue from the links.