Top thicknesses :)

You can ask questions here about Trevor and Gerard's exciting new book on Luthiery.

Moderators: kiwigeo, Jeremy D

Post Reply
JurgenV
Myrtle
Posts: 74
Joined: Mon Dec 10, 2018 6:58 am
Location: Bavaria, Germany

Top thicknesses :)

Post by JurgenV » Wed Oct 05, 2022 5:44 am

Looks like it´s getting a bit more active nowadays. And ... some quite similar problems.

So I´ll join the club. After 4 guitars following the principles of Trevors book and me thinking yepp, that´s no big problem anymore. I posted some information about those guitars in the forum.

I finally have a customer not only paying for the material but willing to pay 1500€ for one of my guitars. Wants some rosewood 000 with sitka top and some additional features. Me: yeahh, I can do that in 3 month. No problem. But reality caught up with me
Everything went fine until I started with the top. I´m now at top number three and no matter what I do I always end up at around 180Hz with the just closed soundbox (no binding, no nothing).
I determined the long grain, cross grain and twisting frequencies using the same way as before with the same microphone and computer. I used these values and got thicknesses ranging from 2.6mm to around 3mm. After the first two desasters I raised the primary braces to 9.5mm and still I end up at around 180Hz for .
I´m running out of time. There is no way I´m going to finish the guitar at the promised date. Getting a bit frustrated. And I´m now down to my last two tops.

I have one sitka top and one engelmann top left. I did my tests again and come up with the following values:
sitka:
sitka.jpg
for the engelmann top:
engelmann.jpg
Before I have to scrap those again is there anyone who can give me some advice if these values make sense? I thought the difference between sitka and engelmann would be bigger and actually I never had a suggested thickness of around 3.3mm for a sitka top.
And yes, I did the tap testing for both panels several times. The values changed almost not at all. 1 or 2 Hz at most.

As I said I would be grateful for some advice. I´m getting frustrated. Especially because I don´t find the reason for what´s happening.

Btw. the same tests for the back gave me a reasonable 228Hz after closing the box and before shaving the main back brace.

JurgenV
Myrtle
Posts: 74
Joined: Mon Dec 10, 2018 6:58 am
Location: Bavaria, Germany

Re: Top thicknesses :)

Post by JurgenV » Thu Oct 06, 2022 6:35 am

Ahh, I forgot: you need the box size for the guitar

a=520mm
b=385mm

If I use the values from table 4.5-3 I can reproduce the target thickness.

User avatar
Trevor Gore
Blackwood
Posts: 1605
Joined: Mon Jun 20, 2011 8:11 pm

Re: Top thicknesses :)

Post by Trevor Gore » Tue Oct 11, 2022 8:57 pm

Jurgen, I put your numbers through my spreadsheet and got 3.15mm for the Sitka (twice) and 3.22mm for the Engelmann. So I suspect you may have got the data entry wrong on the Sitka, unless the data you posted is your actual spreadsheet, in which case something really weird is happening.

As for always ending up at 180 Hz, I can only speculate. I know that my builds nearly always tap out a lot stiffer (higher frequencies) than most people's, using nominally the same material and procedures. Are you measuring the stiffness of your brace material?

JurgenV
Myrtle
Posts: 74
Joined: Mon Dec 10, 2018 6:58 am
Location: Bavaria, Germany

Re: Top thicknesses :)

Post by JurgenV » Sat Oct 15, 2022 7:14 am

Thanks for your reply Trevor.

For the sitka I found a small mistake. After correcting it I got the same result. So that seems to be ok. In the mean time I also checked with the deflection method and found quite similar results. So for this time I think I can trust the values for the target thickness.

Your values for a closed sound box are higher as the ones I usually reach but I never came out that low. Btw. what contributes more to the top frequencies? Braces or top thickness? At the moment my guess is the top thickness because when the last one came out with 180Hz again I just wanted to know what happens when I thin the top further (to see if the frequency increase due to mass reduction or the loss of stability wins)
Result: the T(1,1)2 dropped quite fast even with braces 9.5mm high

At the moment my guess is there was some problem with the tap testing. Some of those tops were slightly bent in the short axis (cross grain) while being totally straight in the long grain direction. The last two remaining sets didn´t show any bending so the probability for trustworthy results is higher.
Just to make sure I don´t have any problem with the epoxy I also bought new epoxy. Never leave Murphy a chance.

As I´m going on vacation for a week there will be no update on the issue for the next ten days. I also didn´t want to go on vacation with a bad feeling if it turns out I have the same problem again :)

And yes, I did also measure the stiffness of the braces (sitka). According to the excel sheet I should go down to around 6mm :)

User avatar
Trevor Gore
Blackwood
Posts: 1605
Joined: Mon Jun 20, 2011 8:11 pm

Re: Top thicknesses :)

Post by Trevor Gore » Sat Oct 15, 2022 11:04 am

JurgenV wrote:
Sat Oct 15, 2022 7:14 am
At the moment my guess is there was some problem with the tap testing. Some of those tops were slightly bent in the short axis (cross grain) while being totally straight in the long grain direction.
That would explain things. Cupping across the grain will make the long grain Young's modulus measure higher than it actually is, resulting in your resonant frequencies coming out lower.

JurgenV
Myrtle
Posts: 74
Joined: Mon Dec 10, 2018 6:58 am
Location: Bavaria, Germany

Re: Top thicknesses :)

Post by JurgenV » Sat Oct 15, 2022 4:57 pm

Yepp, that would explain a lot.

What do you do with such a set of top panels? The deflection method will also not work on those. Is there any way to still use them?

User avatar
Trevor Gore
Blackwood
Posts: 1605
Joined: Mon Jun 20, 2011 8:11 pm

Re: Top thicknesses :)

Post by Trevor Gore » Sun Oct 16, 2022 11:34 am

JurgenV wrote:
Sat Oct 15, 2022 4:57 pm
The deflection method will also not work on those. Is there any way to still use them?
It's still possible to use the deflection method if the cupping is not too bad. Basically, pre-load the panel with enough weight to flatten the cupping, zero the deflection gauge, then add more weights to take the deflection reading. Flip the panel and repeat, bending in the opposite direction. Take the average of the results, and, if the cupping was not too bad, you should get somewhere near an accurate reading.

Another (arguably preferable) method is to flatten the cupping. This method sounds counterintuitive but it works. Put the panel on the bench, convex side up. Wet the convex (top) side with plenty of water, but not so much that it runs off (a liberal spraying). Leave it on the bench to dry. The panel will get even more convex initially, but as it dries it will dry with less or no cupping in the original direction. One wetting is often sufficient to flatten the panel, but repeat up to three times if necessary. If it doesn't flatten after three tries, it is not going to, but you may be able to flatten it sufficiently to to get a decent result from the first method.

Post Reply

Who is online

Users browsing this forum: No registered users and 31 guests