As you now know, Jim and Matt from Grafted Coatings have been giving me help in parallel via email. It's getting a little confusing doing this in two paths, one of which is private, so I thought I'd get everyone on the same boat with the same information by copying from my latest email to Jim into this post. Also I'd like to thank all those who have helped out thusfar.
Here's the email:
**************************************************
Jim,
Thanks for being so willing to help. Here's the information you requested:
1) Finish Method:
a) The instrument (Indian Rosewood and Engelmann Spruce) was sealed and pore filled with epoxy ala Doolin. I did not use his (SB3??) epoxy, though, I used ZPoxy 15 minute formula. Doolin felt that any epoxy should yield good results, and I shouldn't worry too much about which one I used. I intended to use the ZPoxy Finishing Resin as a filler like most of the users on the Luthier's Forum use, but could only find their 15 minute adhesive locally, and didn't have time for mail order. The ZPoxy was scraped on with a razor blade, even the top, as epoxy brings out the depth in the grain better than shellac in my observation. The epoxy filler/sealer was applied in straight lines with the grain. With the razor blade, 99% of the epoxy went in the pores, and an imperceptible amount stayed on the surface, evidenced only by the much darker and richer color of the wood. Only a scuff sanding here and there was needed for a couple of very minor epoxy ridges. The angles of the "blistering"/pitting do not correspond the the angle of epoxy application.
b) The KTM-9 was applied during the following two days (basically per Doolin's schedule) with two foam brushes - a wider one for the top/back, and a narrower one for the sides. It was applied with a method similar to Turko method you sent me. Prior to starting, I transferred the finish from the quart can into several plastic bottles obtained from a plastics dealer. I'm not sure of the material (I may be able to figure it out if I kept invoices), but I know at least some were meant for chemicals and such. Using a squeeze top on one of the bottles (roughly 8 oz capacity), the finish was squeezed onto the end of the foam brush and applied to the instrument. Perhaps it's because of this way of wetting the end of the brush that I got what I thought were surprisingly thin coats. Each linear pass of the brush was refreshed with more KTM-9, and delivered slowly and evenly. No problem with bubbles (a pleasant surprise!). Application temperature was at about 72 - 76 degrees, and relative humidity was at about 40% or a few points lower at the high end of the temperature range.
c) After each coat, the finish was allowed to dry with the instrument suspended and a fan a few feet away, and an infrared lamp (??Watts?) about 18 " away. The fan imparted a swinging/spinning motion to the instrument, so all sides got some infrared and direct breeze. Between coats the foam brush was rinsed with warm water until the runoff was clear and no finish was seen when squeezing the brush. It was then placed in a zip-loc bag after shaking excess moisture from it. Six coats were applied the first day, each on the hour.
d) The second day I leveled the surface with 3m Fre-Cut 220 or 320 through perhaps 600 grit by hand using a hard block. After leveling the surface was wiped with alcohol to clean and neutralize. I believe I used ETHANOL rubbing alcohol (hard to find), or I may have used isopropyl. Six more coats of KTM-9 were applied in the manner described above. The second day, I may have done a very minor leveling with finer grit prior to the last three coats ... I don't recall. If I did, I probably wiped the sanding dust off with a water moistened paper towel.
e) The instrument was then hung up to cure for what ended up being a whole month at room temperature (60's). For a lot of that time, a set a small fan blowing on it from a few feet away in hopes of aiding off-gassing.
One month later ...
f) After the cure, I did final leveling and buffing. Due to mental fatigue and a temporary lapse of judgment (or just totally being out of my mind), I began the final leveling with 3M Free-Cut 320 - way too coarse. As a result, I was never able to recover from the sandpaper scratches completely. The scratches are a separate issue, one that I'm not puzzled by; I know the cause and solution. The leveling progressed through Free-Cut 800, and then went on to StewMac Micro Finishing Papers 1000 - 2000 grit, these used wet with soapy water, and each grit in alternating directions (0/45/90 degrees). No finish anomalies were noted up to this point ... not even the scratches (which do not seem to be related to the problem in question).
g) The next step was using 3M Finesse It II, applying by hand with a cotton rag (T-Shirt). At this point a gloss started to form. With the slight gloss present, the scratches became evident, but no sign of the problem in question. But, after doing the rest of the guitar, the mysterious problem became evident in one area of the previously worked back. The problem, in a longer view than in the photo provided at the top of this thread, was in the form of a line or several lines (like in the photo) that were a couple to perhaps 6-8 inches long, and with a gentle curve to them. I worked those areas particularly hard with the Finesse It II, and was able to reduce some of the defect, but not eliminate it. That was the stopping point for the day.
h) The next morning I resumed, inspecting the guitar relatively early in the process below. Surprisingly, several more of these lines had formed, in varying angles across and with the grain. They don't seem to correspond well to any angles that were used in the application of finish or filler/sealer. I then applied Perfect It II, which bought the finish to a higher gloss. A few more lines appeared on the back (none on the front yet). At this point I was thinking if there was a contaminant on the carpet/sheet under the instrument, rotating the guitar as I work on the top could account for the various angles of the lines.
i) I wasn't very satisfied with the gloss, so I tried an alternative approach that has worked in the past on my KTM-9 test swatches. I powered up a hand drill with a new 6" loose flannel buff, and applied Menzerna Fine compound. This was a step backward in grit, so the Perfect It II gloss became a little hazy. The anomalous lines seemed to diminish slightly with application of the Menzerna Fine.
j) Afterwards, with a separate buff only used for Menzerna Glans Wachs, I applied that in the same manner. This restored the gloss, and perhaps reduced the lines a little more. To remove any Glanz Wachs residue, I used a dry paper towel, which seemed to do a better job than cloth
.
This left the instrument glossy but with minute scratches and the defect in question. Neither of these can be seen in the long shot below. They become very evident with closer inspection.
At this point I considered the finish "done", though I was not satisfied with the results. Because I have a well known luthier visiting me in a couple of weeks, I was motivated to get this instrument done before his arrival. (That plan has since gone by the wayside, as I've decided to try to fix the finish). With the finish considered done (at the time), I went on to fit the neck. This involved handling the instrument and changing it's position on the blue sheet/carpet shown. Over the next day I noticed a couple of more new "lines" in place previously OK. On the third day after buffing I noted two lines on the TOP this time. Regarding potential transference from the blue cloth, I will say that the instrument has spent 90% of the time on it's back, and only occasionally lying on its top. Hmmmmm ... is that why there are far more lines on the back than the front? Interestingly, NO LINES have appeared on the sides. The guitar has only rested on it's sides for seconds on a few occasions. I don't know if that's important, but it's something to consider.
2) Regarding the effect from the carpet/sheet ... I really don't know. My last few sentences above, though, might indicate some involvement. The carpet is some unused carpet from when my house was built about 15 years ago. It's been rolled up in the basement for that time. It was rigorously vacuumed with a Fein Shop Vac prior to use. I have no clue what a carpet might exude, and if it still would do so after 15 years. Regarding edges, the carpet has a pretty thick nap (is that the right word?), so there are no real sharp edges. If it were pressure concentrated on the edge of the carpet and imprinting through the sheet, I'd expect it to be a lot more diffuse, and not 1/16" wide. Also, some of the lines go through the center portions of the guitar, and I NEVER had the instrument lying half off the carpet/sheet such that the carpet edge would be at the center of the guitar.
By the way, you see the line of white "trim" in the blue sheet (actually pillowcase) above? Thinking that might be the cause (it's about the right width), I placed one of my test swatches directly over that and weighted it overnight. No defect in the test finish this morning. But then, the test swatch is a couple of years old ... perhaps curing that long it would not be susceptible? Also I placed test swatches on that very same carpet/sheet last night, as well as contacting the plastic bottle that dispensed the KTM-9, and those did not come out with any defects. Again, the swatches were a at least a year old.
3) Regarding problems in the substrate ... there is a glue line down the middle of the back ... actually a whole strip of inlay. The defect lines appear everywhere, including across that. So it's not that. Resin in the substrate? Well, probably 100%, as I filled/sealed with epoxy. But maybe a cured epoxy is not considered a resin? Other than those two things, there is nothing that was done in the construction (intentionally at least) that would cause substrate problems where the lines actually are.
Well, I've written a WHOLE LOT for you here, hopefully not too much! But I guess the answer is in one of these details somewhere, or perhaps a detail I haven't pieced together or remembered yet.
I truly appreciate your support, Jim.
Thank you so much,
Pete
Pete Licis38721.6492824074