Homebuilt Changer Fingers

Eldon Cangas · Post by **Eldon Cangas** » 24 Sep 2009 4:41 pm

I'm looking for thoughts and recommendations

I've just modeled (wood $ aluminum flashing) a changer finger and lower lever. I haven't tried to move the lower lever pivot point around yet so before I do. How much does the Top Dead Center have to move to get a single and double lower. Mine loosens 5/16", the way I have it attached right now.

I think I'm on the right track.

Thanks Folks, Eldon

Bent Romnes · Post by **Bent Romnes** » 25 Sep 2009 5:58 am

Eldon, interesting little subject. It forces us to analyze what's going on.
I hope I am understanding you correctly when you talk about
I just discovered that what you mean by TDC is not TDC of the actual top pf the finger that the string lies on, but rather the swing of the top of the lower bar.
Please confirm this and we might take it from there.

Bobby Burns · Post by **Bobby Burns** » 25 Sep 2009 6:07 am

Mount your changer finger and a tuner to a board, and put a string on it. I think you will be amazed at the small amount of movement required to change the pitch a whole step. The actual distance will depend on a lot of geometric variables in you design, It may be diferent than any one else's, and still work. Also, a design change that effects the distance it moves may also effect the force required for that move. The easiest way for you to figure it out, is to tune up a string on a board and pull it.
It is easier to test this now, after making only one, then to figure out you need to change it after you have made all ten fingers.

Bent Romnes · Post by **Bent Romnes** » 25 Sep 2009 12:10 pm

Bobby, you are giving me ideas here. I should do exactly that, mount it to a board etc. But to arrive at what Eldon wants to know, I can't simply mount just a string on a board, I would have to mount the whole pulling system, raise and lower fingers and string to the board. I already mounted just a string to find the pounds of pull needed to pull the 3rd G# to A. As I remember,the distance pull was something in the area of 3 mm, in other words, like you say, very minuscule indeed.

To arrive at distances for the various components, the whole assembly has to be mounted. I will publish my results when and if I get around to trying this.

Bobby Burns · Post by **Bobby Burns** » 25 Sep 2009 12:50 pm

Bent, I am flattered to have maybe given you an idea. You are an inspiration to all who desire to build their own steel guitar. Most who express an intrest are promptly told by most folks that it is foolish and next to impossible to build their own, You have proven that it IS possible, and without a complete machine shop full of expensive tools. My hat is off to you sir.

Yes, when I said mount the finger, tuner and string, I meant the whole raise and lower system of that one string, not just the string and finger. Mounting it this way also lets you play with all the dimensions that effect the leverage of all the parts that can change the distance and tension of the pulls. It is nice to be able to make a few different parts and try them out, before you make ten complete assemblies, and then find out they require a very long travel, or they are just too hard to pull.

Allan Munro · Post by **Allan Munro** » 25 Sep 2009 2:21 pm

Following this with interest. I am also at the 'mounted on a board' stage...

Good luck with your project, Allan.....

Eldon Cangas · Post by **Eldon Cangas** » 25 Sep 2009 2:49 pm

Man You Guys Are Awsome. To answer Bent's question......... What my meaning of TDC is the top of the changer finger because that's where you measure your scale length. Yes I am in the analysis stage and it's a real interesting journey.

Just thought of something........ It seems that you should be able to put a pencil mark on the TDC of the changer finger, step on a pedal, mark TDC again and measure the difference.

I'm heading over to a friend's place next week to take some pictures and get some measurements. Erol has been nothing but suportive and helpful just as he was when I built my first lap. Will keep you guys up to date.

Bent Romnes · Post by **Bent Romnes** » 25 Sep 2009 3:07 pm

Eldon, Ok top of finger it is, and you might be surprised at my results which I measured on my prototype this morning. You're right; I put a pencil mark on tdc of the finger and onto a spacer beside it.
First I checked to see which finger turned the most when activated. And here is the surprise: It was the 3rd finger, G# to A raise! I always thought that it would be a wound string LOWER that had the longest travel.

Anyway, this finger with the longest visible travel, moved .060" off the marks. This is just under 1/16 of an inch. In other words with your 5/16" of travel you should theoretically be able to raise a string 5 half tones. Of course there are other factors at play here, that's why I said theoretically

Bobby's advice and suggestion is more definitive. Something to think about doing on a slow day

Also, maybe this test could be done easier, and just as accurate in a good drawing program like Inventor. I'll check with a friend on exactly that.

Russ Wever · Post by **Russ Wever** » 25 Sep 2009 5:17 pm

Sounds like you're using linear
units of measurement (inches, and
fractions thereof, in this case) to
measure a radial movement.
Might it be more conversant, and
perhaps more precise, to use radial
units of measurement such as degrees
(and minutes and seconds) to measure
the radial motion?
'Just a thought', as they say.
~Russ

Bobby Burns · Post by **Bobby Burns** » 25 Sep 2009 5:33 pm

That's a good point Russ. I think for a lot of uses, degrees are the way to measure this movement. However, For a given degree of movement, the string may move a different distance, if the distance from the center of the axle to the bearing surface of the sting on the finger is different. Also, on a given finger, the distance on top of the finger, and the degree of movement will be the same for the same interval pull, but the pull rod will travel different, and pull with different tension, depending on the distance from the center of the axle to the pull rods, and the point that the finger hinges. So I think the distance answers this particular question better.

Eldon Cangas · Post by **Eldon Cangas** » 25 Sep 2009 6:41 pm

Note to Bobby & Russ,

The original measurement I used was linear it was quick dirty and easy.

I did a few calculations tonight base on Arc Length. I compared a 3/4 inch vs 1 inch changer width because Bent & I have sent a few emails back and forth about related items. Because of the materials I have on hand I may end up with a 1 inch wide changer finger and I was intrigued by the fact that a shorter radius in theory will have a smaller string contact point at TDC of the finger.

If my calcs are right the difference between the two is .02 inches which should equate to about 1/64 inches. At this point I'm thinking we're building a pedal steel not a watch. I don't think I have much to worry about.

Warmest Regards, Eldon

Russ Wever · Post by **Russ Wever** » 25 Sep 2009 7:10 pm

I compared a 3/4 inch vs 1 inch changer
width because Bent & I have sent a few
emails back and forth about related items.
Because of the materials I have on hand I
may end up with a 1 inch wide changer finger

Okay, I obviously wasn't getting the gist of
the lingo usage. When fingers are 1 inch wide,
of course the thickness comes to mind, and I'm
visualizing a ten-string guitar with the changer
nearly a foot wide, and the strings being an inch
from each other!!
Typically when discussing the 3/4 and/or 1 inch
measurements that are mentioned above, it's stated
in terms of the 'radius' of the finger, which would
be 3/8" radius and 1/2" radius, respectively.
'Scuze me,
~Russ

Eldon Cangas · Post by **Eldon Cangas** » 25 Sep 2009 7:21 pm

Thanks Russ, I appretiate the clarification, I'll think a bit more before I type the next time.

Warmest Regards! Eldon

richard burton · Post by **richard burton** » 25 Sep 2009 9:56 pm

I've measured the movement of the 4th string full-tone raise on my Marlen.
As it is a pull-release system, the finger is one-piece, and therefore it's very easy to get an accurate measurement.

I measured the movement at the bottom of the finger, and have calculated (by trigonometry) that the degree movement to raise a 0.014 plain string from E to F# over a 3/4 diameter changer finger is 7 degrees 6 minutes (7.1 degrees), which equates to 0.046" travel on the circumference.

As my kids are always telling me, I am a sad b@st@@d

Russ Wever · Post by **Russ Wever** » 25 Sep 2009 10:36 pm

I measured the movement at the bottom
of the finger, and have calculated (by
trigonometry) that the degree movement
to raise a 0.014 plain string from E to
F# over a 3/4 diameter changer finger is
7 degrees 6 minutes (7.1 degrees), which
equates to 0.046" travel on the circumference.

Of course altering the guage of the string,
and/or the radius of the finger would have
affect on the radial measurement (7.1 degrees
in Richards case).

Another variable to consider in the equation
might be the relationship of the 'unplayed'
distance of string (ie, distance between
stringpost and roller) vs. the scale length.

~Russ

Bobby Burns · Post by **Bobby Burns** » 26 Sep 2009 4:39 am

You can spend a lot of time working out on paper, but you will get there quicker and visualize it better, if you put it on a board and play with it. I guess some kids like to play with numbers, just like some kids like to play with erector sets. I always liked to build things myself.

Actually, I work things out (and re-work them out) on paper quite a few times before I ever make the first prototype. After I play with the prototype a little, I often go back to the paper. And it really does help to know the angles too.
As far as the precision, you can vary things quite a bit, and still have a playable, tunable guitar, as long as things are adjustable, and the parts fit together good. The precision comes into play in making repeatable cuts, so that all your fingers are similar, and the parts fit. If things get sloppy, you may not be able to get them to move to the same note repeatably. If they fit and things aren't too sloppy, when you adjust your pull rods they will return to the same pitch repeatably. It won't really matter if your circumference is a little smaller than planned, and it has to move an extra fraction of a degree, it will still tune. Of course, it will play better if the circumference of all the fingers is the same. The repeatability is where the precision comes in.
I think everyone is surprised by the small amount of movement required at the string edge of the changer the first time they ever measure it.

Bent Romnes · Post by **Bent Romnes** » 26 Sep 2009 8:15 am

Bobby, I appreciate so much your kind words about my building efforts. Believe me, it makes a huge difference to get approval from a guy with machining AND musical experience.

As for my feeble attempts at setting up a jig for the purposes we are discussing here, I am starting to realize that to to a believable test, one does need to set up the whole pulling train like you said, and then go through the various string gauges, wound and plain. Make accurate measurements on all points of the raise/lower bar and fingers, Where measuring is required in degrees, minutes and seconds appropriate measuring equipment is needed.

Richard Burton, a machinist, gave the whole thing a good start. I already see that he is miles ahead of me with his trigonometry etc.

The whole thing is so interesting to me that I would be willing to make such a jig and waste a half dozen sets of strings and changer parts in the process. But of course I would need y'all's help. You yourself are a machinist. Russ Wever certainly shows that he knows his stuff..Richard I already mentioned.

People with new ideas for changer mechanisms are most welcome to participate. I know about a couple of you, you know who you are, let yourself be heard.

Edit: Yes, Bobby, I was totally taken aback by the small amount of movement needed to lower a stringa couple of frets.

Marty Smith · Post by **Marty Smith** » 17 Mar 2010 8:33 pm

heres some i made

J D Sauser · Post by **J D Sauser** » 17 Mar 2010 9:19 pm

The longest pull on E9th, C6th and it's universals is a W22 at the E9th 6th position (long key head) tuned to G# and LOWERED two semi tones to F#. If I remember right, it's about 3/32" movement on a 24" to 24 1/2" scale. I was told that early PSG's changers lack of ability to complete a movement that large, is what turned many PSG players on to plain o.o22 strings for that position after Buddy Emmons came up with that change.

Next would be the 3rd E9th string G# to A B-pedal raise and, if I am not mistaken the 4th string E to F# C-pedal raise. Which are both just bellow 1/16" of stretch movement.
Al Packard will probably have some exact measurement/calculations available.

... J-D.

Bent Romnes · Post by **Bent Romnes** » 18 Mar 2010 5:50 am

Thanks JD, and it's nice to see this thread re-awakened.

Jim Palenscar · Post by **Jim Palenscar** » 18 Mar 2010 6:17 am

Actually I believe that typically the next longest one after the W22 6th lower a full step is the 1st string raise a whole step.

Bent Romnes · Post by **Bent Romnes** » 18 Mar 2010 8:26 am

From the wee bit of experimenting I have done, I arrived at the longest pull to be lower the 3rd .011 a whole tone. Plus the G# to A raise like I mentioned earlier.

Can't wait to start making some more and rigging it up on a board to determine this for sure.

Jerry Jones · Post by **Jerry Jones** » 18 Mar 2010 11:15 am

I’m with Richard on the “degree method” of measurement. That should be very accurate and easy to transfer to a simple CAD drawing. Those degree readings can always be translated to linear movement if needed. If you are using a mock-up to test
different string gauges and frequencies, don’t forget to allow for the extra string length and stretch that occurs between the tuning key and the roller nut. Good luck! Let us know what your results are. I’ve been curious about this myself.

Bent Romnes · Post by **Bent Romnes** » 18 Mar 2010 1:51 pm

Jerry, I tend to agree. It makes sense out from the standpoint of determining the actual swing of the finger needed. I can see where applying this to a CAD drawing like Inventor. That way, if you have the linear travel needed, the radial travel can be worked out in the 3D drawing to make sure you have all the parts moving as intended without hitting anything. Of course linear travel must be determined in order to learn the radial finger movement. In other words: The linear movement is always the same. The radial movement of the finger is dependent on the finger radius.