OK Art, I just drilled (32) holes in the top of the horn walls for the chamber divider screws. One will need a crow bar to pull this wood off, solid as can be, BUT the thick walls are easy. The 1/4 thick walls are another animal. I need to center punch these critical screws, to assure they are dead nuts on center, any offset, toast.
I had to order new #4-40 x .50 long button head steel screws to make this work, looking good.
Also the tap drill size is absolutely critical, using a .089" diameter drill, has thread bite. Use a smaller drill, and the self tapping wood screws will crack thru the wall, be careful.
For the soundboard, I think glue would be better, but I need to disassemble as needed. Maple will last (50) years, speakers and electronics, not a chance.
Plus if you need to offset 150 lbs. of force on the soundboard, and you also have (32) screws, only 5 lbs. of force on each screw.
Time for REW, stay tuned! Joe
I had to order new #4-40 x .50 long button head steel screws to make this work, looking good.
Also the tap drill size is absolutely critical, using a .089" diameter drill, has thread bite. Use a smaller drill, and the self tapping wood screws will crack thru the wall, be careful.
For the soundboard, I think glue would be better, but I need to disassemble as needed. Maple will last (50) years, speakers and electronics, not a chance.
Plus if you need to offset 150 lbs. of force on the soundboard, and you also have (32) screws, only 5 lbs. of force on each screw.
Time for REW, stay tuned! Joe
REW seems to show very consistent info compared to earlier measurements. The input vs. output via horns matches what I am hearing.
When the horns are not closed (uncovered) I literally need to put my hands on the drivers to verify they are working, hence the very low dB in that stage.
Sealed horns (covered) rock, hence the 25-20 dB gain!
When the horns are not closed (uncovered) I literally need to put my hands on the drivers to verify they are working, hence the very low dB in that stage.
Sealed horns (covered) rock, hence the 25-20 dB gain!
"Consistent info" does not equal useful information, nor a valid frequency response measurement.
I keep hoping you will learn how to measure the frequency response of a speaker enclosure using REW's sweep generator as an input, so we can see the response of yours.
Best wishes for the new year!
Art
Art - my strengths are going well, my weaknesses, not so much, examples:
I finally spent some time researching how to do REW sweeps correctly, boy was I clueless, and it showed.
Correct me if I am wrong on any details please, but I need to connect my guitar/pickups/amp/speaker system to my laptop, so REW can run a sweep signal generated by the REW audio generator through my system, then simultaneously reproduce the signal as an output.
Then I measure this as usual, and setup the correct limits so it is easy to read. I will also setup some of the preferences correctly for sweeps, I see there are some important options, and I believe I need to have REW recognize my device. Then of course I will post the results while I sweat out the response from others. 🤣
I do think my comparisons are helpful and honest data, if only for marketing. The technical specs are more important of course, so working on that. Did I say I was done measuring? OK I am back because I understand it a bit better now, motivator. I am sure I am not there 100% on sweeps, but moving forward.
The woodworking is going very well. I also doubled checked all my measurements with the neck, frets, soundboard, and bridge. I have a nice sturdy steel template with Martin 25.34" long scale, looking good, then add some compensation length to the bridge, and all set. I just need my new screws now, Amazon has been very slow lately.
I have a few knots to fill with a combination of natural wood filler and Titebond glue mixed together, works well and sand it down. Can be stained and sealed. You can still see it, so I am thinking about a combination of epoxy and clear glue mixed together, for any high visible areas.
I saw a YouTube video where a guy literally took a 3" thick by 6 foot long section of wood from a huge tree, gutted out all the rotten wood, and filled it with epoxy, looked pretty good for a table, as the real wood underneath the epoxy shows through. I am only talking 1/8" wide section on my guitar, no wood is 100% perfect if you are going to cut from big sturdy slabs for the body. AAA soundboard wood is thin enough and high enough quality to not have that problem. Stay tuned! Joe
I finally spent some time researching how to do REW sweeps correctly, boy was I clueless, and it showed.
Correct me if I am wrong on any details please, but I need to connect my guitar/pickups/amp/speaker system to my laptop, so REW can run a sweep signal generated by the REW audio generator through my system, then simultaneously reproduce the signal as an output.
Then I measure this as usual, and setup the correct limits so it is easy to read. I will also setup some of the preferences correctly for sweeps, I see there are some important options, and I believe I need to have REW recognize my device. Then of course I will post the results while I sweat out the response from others. 🤣
I do think my comparisons are helpful and honest data, if only for marketing. The technical specs are more important of course, so working on that. Did I say I was done measuring? OK I am back because I understand it a bit better now, motivator. I am sure I am not there 100% on sweeps, but moving forward.
The woodworking is going very well. I also doubled checked all my measurements with the neck, frets, soundboard, and bridge. I have a nice sturdy steel template with Martin 25.34" long scale, looking good, then add some compensation length to the bridge, and all set. I just need my new screws now, Amazon has been very slow lately.
I have a few knots to fill with a combination of natural wood filler and Titebond glue mixed together, works well and sand it down. Can be stained and sealed. You can still see it, so I am thinking about a combination of epoxy and clear glue mixed together, for any high visible areas.
I saw a YouTube video where a guy literally took a 3" thick by 6 foot long section of wood from a huge tree, gutted out all the rotten wood, and filled it with epoxy, looked pretty good for a table, as the real wood underneath the epoxy shows through. I am only talking 1/8" wide section on my guitar, no wood is 100% perfect if you are going to cut from big sturdy slabs for the body. AAA soundboard wood is thin enough and high enough quality to not have that problem. Stay tuned! Joe
A guitar or pickups response are not something you can test with REW.
Your guitar amp frequency response could be tested with REW and adjusted for flat response.
The guitar amp's tone control and effects curcuits may make it difficult to get it's response flat.
To test the speaker system, use a flat frequency response amp and the sweep signal generated by REW.
I see, thank you much Art. For me this is a trial and error learning situation. 🤣
What do you think about using my amp/speakers with all the tone controls on zero, no reverb, no chorus. Master maxed as usual, and volume on about 70% as I usually do, through channel 1. Then of course I can test channel 2 also. Please advise and thanks!
What do you think about using my amp/speakers with all the tone controls on zero, no reverb, no chorus. Master maxed as usual, and volume on about 70% as I usually do, through channel 1. Then of course I can test channel 2 also. Please advise and thanks!
Joe, here's an idea. What if you make your horn guitar all acoustic?
Put a hole in the 1/2" dividing board where the speaker rear chamber is. Cut out the speaker holes. Let the top membrane-o-phone's sound bleed through the dividing board hole into the horns and see what comes out!
I have this Brazilian guitar with really shallow braces. The idea would be to get the sound board as close as possible to the dividing board, so it vibrates but without touching; at the same time provide no space for the sound to go except out through the horns. Of course, it would be a top with no sound hole!
If the top bracing has significant height, you could route out reliefs into the dividing board to really bring that top close, but no touch. Then the sound pressure would have its only escape out through the horns.
Lots of ways to think about how the vibrating top would couple into the horns, with the speaker chamber replaced with some entirely new routed out form, intended to best capture the top's backside radiation and shout it out those two horns.
I know you have a lot to do, just figured I'd add something else to the pile - as long as I can still think about things :')
Put a hole in the 1/2" dividing board where the speaker rear chamber is. Cut out the speaker holes. Let the top membrane-o-phone's sound bleed through the dividing board hole into the horns and see what comes out!
I have this Brazilian guitar with really shallow braces. The idea would be to get the sound board as close as possible to the dividing board, so it vibrates but without touching; at the same time provide no space for the sound to go except out through the horns. Of course, it would be a top with no sound hole!
If the top bracing has significant height, you could route out reliefs into the dividing board to really bring that top close, but no touch. Then the sound pressure would have its only escape out through the horns.
Lots of ways to think about how the vibrating top would couple into the horns, with the speaker chamber replaced with some entirely new routed out form, intended to best capture the top's backside radiation and shout it out those two horns.
I know you have a lot to do, just figured I'd add something else to the pile - as long as I can still think about things :')
Thank you JJ for the idea, I thought about doing this early on, but rejected the idea after experimenting with small amps and speakers.
I don't see how it would be possible to do this acoustically, seeing how much power I need now to make it work now. The horns are long, have a pretty decent cross section (for a guitar) and they need power to work. I must have these horns to get the great efficiency and tone, I have learned plenty about that.
The small amps, I mean 2 to 5 watts will not work, not even close. Same with cheap low watt speakers. There was a time when I thought this might not work at all, until I started using some real power (at least for a system this size). The 15 watt per channel amp is really doing the trick, and less than that would be risky. The 20 watt full range speakers are great, and they fit!
Don't get me wrong, I love being open minded and considering ALL ideas, keep them coming. There is just a time when the rubber meets the road and you have to decide what is feasible for an instrument that can sell, and what goes on the "did not work" pile - I have plenty of those!
Anyway learning from failures has just as much (and maybe even more) value than learning from ideas that make it through. Thanks! Joe
I don't see how it would be possible to do this acoustically, seeing how much power I need now to make it work now. The horns are long, have a pretty decent cross section (for a guitar) and they need power to work. I must have these horns to get the great efficiency and tone, I have learned plenty about that.
The small amps, I mean 2 to 5 watts will not work, not even close. Same with cheap low watt speakers. There was a time when I thought this might not work at all, until I started using some real power (at least for a system this size). The 15 watt per channel amp is really doing the trick, and less than that would be risky. The 20 watt full range speakers are great, and they fit!
Don't get me wrong, I love being open minded and considering ALL ideas, keep them coming. There is just a time when the rubber meets the road and you have to decide what is feasible for an instrument that can sell, and what goes on the "did not work" pile - I have plenty of those!
Anyway learning from failures has just as much (and maybe even more) value than learning from ideas that make it through. Thanks! Joe
You can wire both amp inputs parallel with a "Y" cord if the inputs are not already parallel.
To verify the amp output is reasonably flat, start with the master wide open, pre-amp gain off.
Attach the leads of an AC volt meter to one of the amp outputs + and -.
The speakers need not be connected at this point, and may be annoyingly loud if they are, around 105 dB at one meter.
Using the REW tone generator, put a 50Hz sine wave in at -10dBu (245mV), adjust the amp input gain up until the amp output read 2volts AC (1 watt into a nominal 4 ohm load).
Use the "Next 1/3 Oct." button to advance through frequencies up to 20kHz, if the voltage does not vary, the amp output is "flat", if it does vary, use the tone controls to adjust as close to flat as possible.
Verify both channels are the same.
After verifying the amp is flat, do the sweep tests at the same level for a one watt (per speaker) test.
Also note the difference between 1 watt and 10 watts is 10 dB.
Art
Wow, thank you so much Art! That is the kind of detail that really helps me. Hopefully I can pull this off soon, a learning curve for me. Having this kind of direction really motivates me. Now I just hope the results are impressive, but with good sound by ear and comparisons, I am a bit encouraged.
I had the guitar rocking last night, boy does this thing have potential. It sounds like the acoustic guitar but louder, my goal. I was worried that it would be louder and sound like ****. I have never finger picked so softly, and got such a big sound, really cuts down on the string noise, and can really change your playing style and techniques. Sure you could do this in a recording studio, but with a standard acoustic guitar that you just pickup and play, not a chance.
I think young guitar players will really be into this. Back in the day we used to start jamming on acoustics at a party, and people would love it, singing along with our favorite songs, sometimes there was enough talent to get some great three part harmonies going. We would have so many people singing, we would have to beat the hell out of our guitars just to keep up, not any more!
Happy New Year! Joe
I had the guitar rocking last night, boy does this thing have potential. It sounds like the acoustic guitar but louder, my goal. I was worried that it would be louder and sound like ****. I have never finger picked so softly, and got such a big sound, really cuts down on the string noise, and can really change your playing style and techniques. Sure you could do this in a recording studio, but with a standard acoustic guitar that you just pickup and play, not a chance.
I think young guitar players will really be into this. Back in the day we used to start jamming on acoustics at a party, and people would love it, singing along with our favorite songs, sometimes there was enough talent to get some great three part harmonies going. We would have so many people singing, we would have to beat the hell out of our guitars just to keep up, not any more!
Happy New Year! Joe
Joe, have you finally finished the neck and put strings on the folded horn guitar, or still using it as a guitar amp using another guitar?
Still thru another guitar. I have my holes drilled but am waiting on new screws to arrive before I can assemble, had to go down one size to make sure I don't get any cracked side walls. I have to admit the soundboard may need to change if the CNC soundboard does not work, but I can do a hand made soundboard, know how to build that now thanks to Ian, braced one for my guitar kit. It does not bother me too much, since I know it will be even better when assembled, pretty sure! 🤣
You can, of course optimize for flat frequency response, in which case the horns would be useful for PA duty as well as amplifying the guitar's sound. That said, no acoustic guitar body I've yet measured has a flat frequency response, yet they all sound like an acoustic guitar, with some responses being better sounding than others.
So you have a product design decision to make. Do you go the route where someone could sing, play harmonica, hook up a drum machine (etc) in addition to amplifying the guitar's sound?
Or do you design the horns so you get a definitive F0 (first resonance) at a pitch relative to some concert tuned value (say, "80" Hz) and make a nice whole number harmonic series (2X, 3X, 4X, 5X etc whatever it is a horn can do) so that it sounds like a guitar body, only unique to your invention?
It might take a committee to make that decision. If your design were happening at the Fender of Gibson custom shop, I can see everyone sitting around a conference room table discussing it. Sales, with one appeal or sales, via another completely different one. Which do we put our money in?
Oh yes I hear that JJ. I prefer to go for the acoustic guitar sound first and foremost, but I am interested in having other uses (blues harp, voice). I just don't want it to turn into some one man band joke of a device. 🤣 But your point about bigger companies wanting to do this is well taken. I just hope they get a chance to have that discussion!
Also, I can't find my voltage meter, had to buy another one, on it's way via Amazon!
I am very encouraged by the sound, as far as I am concerned, a win. I am also highly interested in test data and technical specs because I want it to have the highest quality possible, and also because I need this info to prove it to potential buyers. JJ you and Art have been such wonderful contributors, so glad you are staying with me through this very loooong process.
Just yesterday I heard a real story about a friend of a friend that has a new product that is being sold now (not musical), and he got a one year provisional patent. When that ran out he decided NOT to pay a patent attorney big bucks for a 20 year non-provisional patent, which I have now, and have another on the way. I have first inventor to file status on that, also wrote it myself. He did not take the time and long journey to do it himself. Now he is in danger of losing everything, as it looks like he might get sued. Not to mention what patent trolls (scum) can do to people legally without ever inventing anything. I am not saying my path is the best, but it is the best for me, so yes I am willing to go the long route and do everything right (I hope).
Plus the musical instrument business does not change quickly, so that helps me. If it was software my product would be obsolete by now!
Also, I can't find my voltage meter, had to buy another one, on it's way via Amazon!
I am very encouraged by the sound, as far as I am concerned, a win. I am also highly interested in test data and technical specs because I want it to have the highest quality possible, and also because I need this info to prove it to potential buyers. JJ you and Art have been such wonderful contributors, so glad you are staying with me through this very loooong process.
Just yesterday I heard a real story about a friend of a friend that has a new product that is being sold now (not musical), and he got a one year provisional patent. When that ran out he decided NOT to pay a patent attorney big bucks for a 20 year non-provisional patent, which I have now, and have another on the way. I have first inventor to file status on that, also wrote it myself. He did not take the time and long journey to do it himself. Now he is in danger of losing everything, as it looks like he might get sued. Not to mention what patent trolls (scum) can do to people legally without ever inventing anything. I am not saying my path is the best, but it is the best for me, so yes I am willing to go the long route and do everything right (I hope).
Plus the musical instrument business does not change quickly, so that helps me. If it was software my product would be obsolete by now!
I am back on the road again for work, without access to shop or equipment, but some thoughts:
The soundboard screws have plenty of tensile strength, the screw pullout force from the wood is 5X greater than the amount of string force acting upon each screw/washer assembly, so good. The washer will provide greater contact area of course, but it is difficult to calculate how much strain this Baltic Birch soundboard can handle before deformation, in each small screw/washer surface area. Need to test it and find out! The first soundboard off the CNC machine had problems due to holding with two way tape, the outsides machined thin as the board moved up and down on the two way tape, does not look feasible at all (you could actually see another ply machined away in some areas). The good news is after learning a few tricks from this mistake, the second soundboard came off the CNC machine very stable, very consistent thickness, looks doable, hopeful! Plus Baltic Birch is made with opposing 90 degree layers, very strong in this regard. Tone compared to Sitka Spruce, not sure about that. It is lighter, and hopefully stronger, so may get greater vibration.
JJ - I posted pink noise in October, but that was before the horns were sealed, which of course was a huge improvement. After sealing, the harmonics are much stronger in every other measurement. Shorter wavelength highs affected more by leaks?? I am hoping that is true with the pink noise also, because the harmonics dropped off. Good way to measure the horns with an external mic, I will post results.
Art - I will verify the amp is reasonably flat for above using your method, and also for your REW Sweep method, to test the amp and the speakers. Am I correct to assume this is done using the laptop only, no external mic required? I know the REW generated signal comes ffom the laptop speakers, I heard that as I experimented with REW and laptop only (no amp or speakers available at the time). Please advise and thanks!
The soundboard screws have plenty of tensile strength, the screw pullout force from the wood is 5X greater than the amount of string force acting upon each screw/washer assembly, so good. The washer will provide greater contact area of course, but it is difficult to calculate how much strain this Baltic Birch soundboard can handle before deformation, in each small screw/washer surface area. Need to test it and find out! The first soundboard off the CNC machine had problems due to holding with two way tape, the outsides machined thin as the board moved up and down on the two way tape, does not look feasible at all (you could actually see another ply machined away in some areas). The good news is after learning a few tricks from this mistake, the second soundboard came off the CNC machine very stable, very consistent thickness, looks doable, hopeful! Plus Baltic Birch is made with opposing 90 degree layers, very strong in this regard. Tone compared to Sitka Spruce, not sure about that. It is lighter, and hopefully stronger, so may get greater vibration.
JJ - I posted pink noise in October, but that was before the horns were sealed, which of course was a huge improvement. After sealing, the harmonics are much stronger in every other measurement. Shorter wavelength highs affected more by leaks?? I am hoping that is true with the pink noise also, because the harmonics dropped off. Good way to measure the horns with an external mic, I will post results.
Art - I will verify the amp is reasonably flat for above using your method, and also for your REW Sweep method, to test the amp and the speakers. Am I correct to assume this is done using the laptop only, no external mic required? I know the REW generated signal comes ffom the laptop speakers, I heard that as I experimented with REW and laptop only (no amp or speakers available at the time). Please advise and thanks!
You can check the level of an AC (sine wave) signal with a volt meter, but measuring a speaker's sound output requires a microphone. An external microphone is more useful for testing than one built into a laptop.
REW signals are created in the digital domain, then either in the computer's own, or an external sound card the digital signal is converted to the analog output connected to the amplifier.
That will be great to see. Remember, pink isnt as good as a sine sweep. An engineer once told me as the excitation waveform moves toward pure sine - it's like adjusting the focus on a camera or microscope. I still think Pink gets you 90% of the way there; when you need single digit accuracy, like 440 vs 443, sine is the way to go.
Great, thank you much gents! The laptop mic did not make sense to me, will definitely use the external mic.
The soundboard may turn into destructive testing. I hope not, but you never know! Joe
The soundboard may turn into destructive testing. I hope not, but you never know! Joe
OK my #4-40 screws have good bite with the slightly smaller tap drill. They tighten up well, and go right back into the threads after being removed. The washers are very helpful, but are larger than the wall section of the body wall. No problem, I just grind a flat on them in assembly, and they can match the outside wall, just turn as required along the entire guitar shape.
I know I am way to focused on these small screws, but they are critical for this guitar assembly, and more importantly, critical for disassembly. I need to get at the electronics easily.
This is unusual I know, oh wait, having speakers and folded horns inside an acoustic guitar is also unusual! 🤣
I know I am way to focused on these small screws, but they are critical for this guitar assembly, and more importantly, critical for disassembly. I need to get at the electronics easily.
This is unusual I know, oh wait, having speakers and folded horns inside an acoustic guitar is also unusual! 🤣