Next up: Back panel installation with bass reflex port, input cup, and crossover board. Once this is done, I should be able to sand all surfaces smooth and then move out to the garage to start applying finish. Hopefully we will get a few cool, dry days in the mid 50's or 60's next week.
Here's another option for finishing - dissolve steel wool in vinegar and then apply. With knotty pine it should give a similar appearance to hitting it with a torch.
Here's another option for finishing - dissolve steel wool in vinegar and then apply. With knotty pine it should give a similar appearance to hitting it with a torch.
Thanks. Steel wool & vinegar? Are you pulling my leg?
Here's another option for finishing - dissolve steel wool in vinegar and then apply. With knotty pine it should give a similar appearance to hitting it with a torch.
Thanks. Steel wool & vinegar? Are you pulling my leg?
No, not pulling anyone's leg. One of the managers at work finished part of a guitar he's building that way.
From the link in my previous post if you add extra pads to the mix it gets darker, same with letting it sit longer. I think my boss used their recipe and let it sit for a full day to get a nice dark shade.
Here's another option for finishing - dissolve steel wool in vinegar and then apply. With knotty pine it should give a similar appearance to hitting it with a torch.
Thanks. Steel wool & vinegar? Are you pulling my leg?
No, not pulling anyone's leg. One of the managers at work finished part of a guitar he's building that way.
I broke the back panel into 3 sections: 1) A permanent 11 inch lower section; 2) a permanent 8.5 inch upper section, and 3) a removable 24.5 inch middle section.
The removable middle section will contain the input cup, the crossover, and the 3x6" BR port. I mounted it with sixteen 2-1/2 inch long pocket screws to force it flat. I will be using a 1/16" thick foam rubber gasket as a seal, so I will be adding a 1/32" deep recess around the inside perimeter to keep it flush with the permanent back panels.
The pine stair treads had knots all over the place. Because of this, I made a photo copy transparency showing the port and input cup locations and used it to hand select two back panels that avoided knots in critical flange or hole locations. But the knots travel throughout the wood at odd angles and sometimes show up in unexpected places. When recessing one of the two input cups, I hit a knot with the router. The knot broke up along the edge, leaving a big crater along the gasket area! (see pic).
As you can see in the pic, this knot traveled at an angle from the side to the top edge. I knew this knot was there, but my hope was that it would fall mainly into the scrap area and not cause a problem. I was wrong.
So, I nibbled out the area with my jigsaw and then smoothed out and squared up the resulting slot area using my flat file. I then cut a small 10/32" x 1 1/4" x 1" slug from some longer stock and glued it into the cutout as a patch. This repair patch area is now almost as good as the 3 other edge areas and should provide a very good seal to the input cup gasket.
From the link in my previous post if you add extra pads to the mix it gets darker, same with letting it sit longer. I think my boss used their recipe and let it sit for a full day to get a nice dark shade.
I mixed up a batch about a week ago and did some initial testing. I used one 0000 pad in a roughly 1.5 pint container. Let it sit for 3 days, then filtered it with a coffee filter. I tested it, giving it 3 coats on one side and 1 coat on the other side. Seemed a little gritty when I applied it. Then I let the mixture sit for another day and it developed a scale type sludge on the top, which I removed. I then gave it a second filtering with a new coffee filter. There was still quite a bit of steel wool left in the mixture that I could see in the filter.
I diluted this mixture 50-50 with white distilled vinegar and let it sit for a couple days. All the steel wool seems to be gone now, and the mixture is getting darker each day. Will test this in the next day or so to see what difference this makes. Bill
I cut the port hole about 1/32" oversize so that I could glue the port into a thin "bath" of silicone as I pushed it into position. Also, I added eight small 1 inch triangular "trusses" for additional support. Each one is held firmly in place with a dab of silicone rubber cement. When the glue sets, this port and flare system should be very rigid.
I cut a piece of 1/4" thick HDF material to match the exact size, mounting holes, and gasket shape of PE's input terminal cup. I then mounted three dual banana jack terminals on it to create an FRD/ZMA testing "pass thru." I may also use this "temporary pass thru" plate as the final input plate, as I would like to start moving away from plastic terminal plates.
The port and input cup are part of a removable back, mounted with sixteen 2 1/2" #8 pocket screws. The crossover will be mounted in the empty space between the BR port and input cup.
As promised, here are my knotty pine test samples. Thanks again for all the great suggestions.
To my eyes, the best test samples are the ones with 2 coats of "Sealcoat" sanding sealer followed up by several coats of semi-gloss clear lacquer. I was somewhat disappointed by the look and feel of the wipe on poly test. My vinegar samples did not turn out well, probably due to my inexperience at mixing and filtering. My burning sample turned out kind of "blotchy" looking (probably need to refine my technique). I also tested shellac (not shown), which looked very good, but the lacquer samples were even better.
We finally got a warm day , so I opened up all the doors in my garage to create a good cross breeze and parked my vehicles outside. In one day, I was able to shoot 3 coats of "Sealcoat" sanding sealer on both cabinets, lightly sanding with 220 between coats. I sprayed both cabinets inside and out. Used up an entire quart of Sealcoat.
Next up: multiple coats of semi-gloss clear lacquer. But the sealcoat will probably have to sit and dry for about a week, as it is going to be in the 30's & 40's next week. They are even predicting snow this weekend!!
The area in the rectangle is what you should be trying for. The blotches are because the torch was in one place for too long. It depends on the grain structure of the pieced-pine that you are using whether or not they will take well to torching.
I agree. What happened is that I was moving the torch slowly back and forth but nothing was happening. The wood was getting dry but it was not turning brown. So I became impatient by the slow progress and paused a little bit too long in some spots. And the wood burned.
I will give this another try next week. Also, I need to use a larger test piece with the grain going in the other direction.
I decided to make use of the current "mid-April-snow-delay" by installing the back, port, and drivers onto one of my partially finished cabinets and running up a complete set of FRD's and ZMA's:
Here is the 3 pane XSim model generated using NF+FF blended minimum phase FRDs. For testing, I used an 80uF NPE protection cap in series with the midrange and a 50uF NPE protection cap in series with the tweeter. For the Neo8 & ND-140's, minimum phase was initially created in the blender and then fine tuned in XSim. "Get file" matches seemed to line up very well. I used a "mod delay" of -0.29 inches for the NEO3W and a "mod delay" of 3.59 inches for the ND140's.
And here is the 3 pane XSim model generated using the FF unblended raw phase FRDs for the NEO8 and ND140's. NF data files were not blended or used in any way. Again, the same 80uF & 50uF NPE protection caps were used in series with the NEO8 and NEO3W. "Get file" matches, again, were very good. The "mod delay" distances came out slightly different using raw phase data:
Neo3W: -0.09 inches. ND140's: 3.99 inches.
Overall, the topology I used is identical to Scott's; total parts count is still 11. I changed the two dropping resistors to account for sensitivity differences and 4 other parts to smooth out differences that were probably due to my baffle faceting. I also changed the overall baffle width from 7.0 to 7.4 inches to accommodate my 61/64th thick side walls. And I increased the total box volume from 1.5 to about 1.7 cu. ft. and changed the port tuning from approx 38Hz to approx 35Hz. Some of these differences are minor, but I have found that almost anything can impact FRD measurements. Bump the microphone stand just a little and I have to start all over again!!
I ran up a complete set of polar FRDs so that I could experiment with the new XSim 3D program. The procedure, as I understand it, is to create and then save your individual driver polar FRDs using the angle number as part of the filename. Then you store the files in the same directory as the on-axis FRDs. The program then automatically grabs and uses the polar data to create an interactive 3D simulation.
As you can see in the attached screen shots, I had to fudge the baffle board a little bit because XSim 3D only appears capable of modeling circular drivers on a rectangular or sloping baffle. Later on, I will be taking actual measurements to see how they compare. But for now, the predicted horizontal polar response appears excellent. The verticals are not as good, but the model seems to indicate that there will be no difference in sound between a seated or standing listening position.
Thanks, Nicholas! It's going to get up to 50+ degrees tomorrow, maybe 60 if I am lucky (SE Wisconsin). Also, it should be fairly dry with a light cross breeze. Decided to go with gloss instead of semi-gloss lacquer. If this makes them look too shiny, I can give them a final coat of semi-gloss to erase the sheen.
Comments
I drilled clearance holes in the Neo8 and Neo3 recess areas for the pop rivets:
I chamfered the rear of the ND140's driver holes at roughly 45 degrees with my dremel tool that had a 60 grit sanding drum type bit installed:
And, finally, I made another fixture to facet the top edge of the baffle board:
Here's another option for finishing - dissolve steel wool in vinegar and then apply. With knotty pine it should give a similar appearance to hitting it with a torch.
Here's a reference I found online: https://www.diypete.com/how-to-make-steel-wool-and-vinegar-stain/
OK, thanks for the reference & tip. Will give this a try. Should be fairly cheap to mix up a batch.
A little progress:
I broke the back panel into 3 sections: 1) A permanent 11 inch lower section; 2) a permanent 8.5 inch upper section, and 3) a removable 24.5 inch middle section.
The removable middle section will contain the input cup, the crossover, and the 3x6" BR port. I mounted it with sixteen 2-1/2 inch long pocket screws to force it flat. I will be using a 1/16" thick foam rubber gasket as a seal, so I will be adding a 1/32" deep recess around the inside perimeter to keep it flush with the permanent back panels.
Bill
Input cup recess repair:
The pine stair treads had knots all over the place. Because of this, I made a photo copy transparency showing the port and input cup locations and used it to hand select two back panels that avoided knots in critical flange or hole locations. But the knots travel throughout the wood at odd angles and sometimes show up in unexpected places. When recessing one of the two input cups, I hit a knot with the router. The knot broke up along the edge, leaving a big crater along the gasket area! (see pic).
As you can see in the pic, this knot traveled at an angle from the side to the top edge. I knew this knot was there, but my hope was that it would fall mainly into the scrap area and not cause a problem. I was wrong.
So, I nibbled out the area with my jigsaw and then smoothed out and squared up the resulting slot area using my flat file. I then cut a small 10/32" x 1 1/4" x 1" slug from some longer stock and glued it into the cutout as a patch. This repair patch area is now almost as good as the 3 other edge areas and should provide a very good seal to the input cup gasket.
Bill
I mixed up a batch about a week ago and did some initial testing. I used one 0000 pad in a roughly 1.5 pint container. Let it sit for 3 days, then filtered it with a coffee filter. I tested it, giving it 3 coats on one side and 1 coat on the other side. Seemed a little gritty when I applied it. Then I let the mixture sit for another day and it developed a scale type sludge on the top, which I removed. I then gave it a second filtering with a new coffee filter. There was still quite a bit of steel wool left in the mixture that I could see in the filter.
I diluted this mixture 50-50 with white distilled vinegar and let it sit for a couple days. All the steel wool seems to be gone now, and the mixture is getting darker each day. Will test this in the next day or so to see what difference this makes. Bill
Another small update: I silicone glued the ports in place and gave them 1/2" round-overs on both ends:
I cut the port hole about 1/32" oversize so that I could glue the port into a thin "bath" of silicone as I pushed it into position. Also, I added eight small 1 inch triangular "trusses" for additional support. Each one is held firmly in place with a dab of silicone rubber cement. When the glue sets, this port and flare system should be very rigid.
I cut a piece of 1/4" thick HDF material to match the exact size, mounting holes, and gasket shape of PE's input terminal cup. I then mounted three dual banana jack terminals on it to create an FRD/ZMA testing "pass thru." I may also use this "temporary pass thru" plate as the final input plate, as I would like to start moving away from plastic terminal plates.
The port and input cup are part of a removable back, mounted with sixteen 2 1/2" #8 pocket screws. The crossover will be mounted in the empty space between the BR port and input cup.
As promised, here are my knotty pine test samples. Thanks again for all the great suggestions.
To my eyes, the best test samples are the ones with 2 coats of "Sealcoat" sanding sealer followed up by several coats of semi-gloss clear lacquer. I was somewhat disappointed by the look and feel of the wipe on poly test. My vinegar samples did not turn out well, probably due to my inexperience at mixing and filtering. My burning sample turned out kind of "blotchy" looking (probably need to refine my technique). I also tested shellac (not shown), which looked very good, but the lacquer samples were even better.
Next up: multiple coats of semi-gloss clear lacquer. But the sealcoat will probably have to sit and dry for about a week, as it is going to be in the 30's & 40's next week. They are even predicting snow this weekend!!
The area in the rectangle is what you should be trying for. The blotches are because the torch was in one place for too long. It depends on the grain structure of the pieced-pine that you are using whether or not they will take well to torching.
InDIYana Event Website
I agree. What happened is that I was moving the torch slowly back and forth but nothing was happening. The wood was getting dry but it was not turning brown. So I became impatient by the slow progress and paused a little bit too long in some spots. And the wood burned.
I will give this another try next week. Also, I need to use a larger test piece with the grain going in the other direction.
A little progress:
I decided to make use of the current "mid-April-snow-delay" by installing the back, port, and drivers onto one of my partially finished cabinets and running up a complete set of FRD's and ZMA's:
Here is the 3 pane XSim model generated using NF+FF blended minimum phase FRDs. For testing, I used an 80uF NPE protection cap in series with the midrange and a 50uF NPE protection cap in series with the tweeter. For the Neo8 & ND-140's, minimum phase was initially created in the blender and then fine tuned in XSim. "Get file" matches seemed to line up very well. I used a "mod delay" of -0.29 inches for the NEO3W and a "mod delay" of 3.59 inches for the ND140's.
And here is the 3 pane XSim model generated using the FF unblended raw phase FRDs for the NEO8 and ND140's. NF data files were not blended or used in any way. Again, the same 80uF & 50uF NPE protection caps were used in series with the NEO8 and NEO3W. "Get file" matches, again, were very good. The "mod delay" distances came out slightly different using raw phase data:
Neo3W: -0.09 inches. ND140's: 3.99 inches.
Overall, the topology I used is identical to Scott's; total parts count is still 11. I changed the two dropping resistors to account for sensitivity differences and 4 other parts to smooth out differences that were probably due to my baffle faceting. I also changed the overall baffle width from 7.0 to 7.4 inches to accommodate my 61/64th thick side walls. And I increased the total box volume from 1.5 to about 1.7 cu. ft. and changed the port tuning from approx 38Hz to approx 35Hz. Some of these differences are minor, but I have found that almost anything can impact FRD measurements. Bump the microphone stand just a little and I have to start all over again!!
I ran up a complete set of polar FRDs so that I could experiment with the new XSim 3D program. The procedure, as I understand it, is to create and then save your individual driver polar FRDs using the angle number as part of the filename. Then you store the files in the same directory as the on-axis FRDs. The program then automatically grabs and uses the polar data to create an interactive 3D simulation.
As you can see in the attached screen shots, I had to fudge the baffle board a little bit because XSim 3D only appears capable of modeling circular drivers on a rectangular or sloping baffle. Later on, I will be taking actual measurements to see how they compare. But for now, the predicted horizontal polar response appears excellent. The verticals are not as good, but the model seems to indicate that there will be no difference in sound between a seated or standing listening position.
Thanks, Nicholas! It's going to get up to 50+ degrees tomorrow, maybe 60 if I am lucky (SE Wisconsin). Also, it should be fairly dry with a light cross breeze. Decided to go with gloss instead of semi-gloss lacquer. If this makes them look too shiny, I can give them a final coat of semi-gloss to erase the sheen.