Intonation with a moving-bridge guitar (Excel, etc)

Earnest Bovine · From: Los Angeles CA USA

In another thread I asked whether anyone had ever measured how much the moving bridge (Excel, Anapeg, Marlen, others?) throws intervals out of tune as you move the bar up the neck.
Excel owners tell me that the effect is negligible, but theory convinces me otherwise. In general I trust empirical data over pure theory; who can forget that theory in 1900 denied the anomalous precession of the perihelion of Mercury, and theory came out wrong on that one.

Richard Burton didn't give pitch data, but did figure out that on his Marlen guitar, the 4th string becomes 0.046" longer when raised to F#. In this example, suppose string scale is 24". Suppose that you tune your raised 4th string exactly in unison with the 1st string (high F#).
Now put the bar at the 12th fret, exactly 12" from the bridge of string 1.
String 1 was 24" open, and is now barred at 12" therefore pitch is exactly F#.

But the bar is less than halfway up string 4, so string 4 will be flat. How much flat? Is it flat enough to matter?

The middle of this string would be at half of 24.046", or 12.023" from each end. So the bar is 0.023" further from the bridge than the middle, where F# would be in tune. Distance from fret 11 to fret 12 is about 0.71". So we are (0.023)/(0.71) of a fret flat on the raised 4th string. This number is 0.032, about 3 cents flat.

At fret 24 the effect would be twice as bad.

Ross Shafer · From: Petaluma, California

I think you might be mis-understanding Richard's post (or I am!)...I believe that the changer on his Marlen does not affect the string's length at all as it pulls more tension...all Marlen changers I've seen (pull/release and all pull) roll the string over the finger while pivoting on the changer axle. As such, the string/scale length stays the same....the amount of distance a given point on the finger's circumference moves is .046" (that's 1.17 mm for you metric savvy folks) ....the amount of movement needed to tension the string enough for that particular change.

Richard please set me straight if I've got this wrong!

I have a Lamar which features a true linear changer, so as the changes occur the scale length of the affected string does indeed change....the string is pulled horizontally (parallel) to the top surface of the guitar. I don't pretend to have the greatest ears, but I've experienced no noticeable adverse affect to the intonation by this small amount of movement. I've never measured the actual amount of movement entailed in the various changes on my rig.

Keep in mind the intonation of these beasts we all love so much is also affected by the amount of pressure we put on the slide at different parts of the neck....this of course is due to the string's deflection (stretching, bending) under the pressure of the bar (without any pesky frets to determine when it stops) which is going to change the affected string's tension (and in turn it's intonation). On regular old fretted guitars the intonation is affected by the height of the strings off the neck, which is why you set the intonation after the string height has been dialed in. I'm sure you'll agree most guitars with movable saddles have string lengths that differ a fair bit more than the .046" example we're using here.

I'm sure in the pure empirical sense you are correct the intonation is being affected, but I also believe that its negligible and that those who posses ears sensitive enough to be bothered by it are probably too busy tuning pianos to bother with pedal steel.

...Bottom line is there's a whole mess of mechanical and sonic puzzles going on with these rigs and its up to the player's ear/hand coordination to make it work. Ah the travails/joys of fretless instruments.

Earnest Bovine · From: Los Angeles CA USA

Ross, thanks for correcting me about the Marlen guitar. I've never actually seen one up close.

I wonder if anyone can tell me how much longer the string gets on an Excel, for example in a whole step raise on string 4 or 5. My recollection is that it moves more than 1 mm, and maybe as much as 2 or 3 mm.

Around the 12th fret, the frets are about 18 mm apart.

BTW I fixed a number in my original post. In that example the F# at the 12th fret would be 3 cents lower on the raised 4th string than on the 1st string.

Michael Johnstone · From: Sylmar,Ca. USA

On the Excel the changer is not totally linear BTW. It has a combination of rotating on a cam like any other guitar and linear motion. On raises,it's more rotational than linear(maybe 70/30)and on lowers,it's almost entirely linear. In case anyone wonders how that can be possible,it's done with elliptical holes in the fingers riding on a round cross shaft.The rest is done with string tension,springs and leverage. I agree that in theory the intonation should be all over the place but to my ear,my Excel is very much in tune all over the neck with all that going on. I don't have the measurements of the throws Doug but I'll get them for you next time I take that thing out of it's case,if I ever happen to do that.

Allan Munro · From: Pennsylvania, USA and Scotland

Bob Blair · From: Edmonton, Alberta, Canada

Doug, I have an Excel set up right now, and no way is it moving that far on the 4 and 5 string raises. I find it hard to measure, but maybe one mm on the 5th string and a hair more than that on the 4th. Like Michael says, the lowers are more linear - lowering the E a half tone it is sliding forward about one mm or so, and when you raise the same strings a half tone the change in length is barely perceptible.

chas smith R.I.P. · From: Encino, CA, USA

Jim Palenscar · From: Oceanside, Calif, USA

I recall a quote by Winnie Winston when speaking on the same subject regarding the Anapeg which has a changer that uses linearity as well as rotation for pitch changing that "Tom Brumley- who has ears like a bat- doesn't seem to think there is a problem". That was good enough for me. I don't notice a problem with mine- although "Hughey-land" and I don't hang out for too long together. Smile

richard burton · From: Britain

Ross,
I initially calculated the degree of rotation required to raise the 4th a tone, because someone on the forum was building a steel and needed to know how big to make the changer hole in the wood, so that there was adequate clearance.

I used this info to calculate the elongation of the string with the finger rotating about a pivot, but it is just as relevant when discussing linear movement.

I've just done a non-scientific test on my steel, which anyone can do, to find out if the scale length difference makes an audible difference.

Place the small radius at the back end of the bar, not the bullet nose, exactly on the 24th fret, on any string, I did it on the 4th.

Pick the string gently, and slide back about a hair (this is the unscientific part Very Happy

), certainly no more than 1/32" (1/2 mm)

See if you can hear the note drop in pitch, I certainly could Very Happy

Gary Cosden · From: Florida, USA

For moving .032" I get 1.59 cents flat. When you push the bar down the string goes slightly sharp since you are bending the string and increasing its length. Probably, I figure at least, somewhere near this same neighborhood. So, at least the way I see it, when the pedals down you get a net improvement in your intonation of that one note (though not in relation to other notes you may be playing) Intonation is a strange subject indeed - especially when it come to a fret less instrument like Pedal Steel.

ed packard · From: Show Low AZ

On a Sierra 25" scale keyless, the E4 = 0.014 string under straight linear pull measures as follows:

Open string Fret 12 Hz
Scale @ F#4 = 25.045 12.523 739.989
Scale @ F4 = 25.023 12.512 698.456
Scale @ E4 = 25.000 12.500 659.260
Scale @ Eb4 = 24.978 12.489 622.254

Approximates to 0.022 inches per halftone change at open string for the E4 string. Other strings will be different. Your bar placement to within 0.002 inches is difficult, and your fretboard is off by more than that at any fret.

Measurement repeatability is within 0.002 inches. Peterson VS-II, and digital micrometers were used for the measurements.

E9 players are heavy on 1,2 and 3 string picking, so correcting any "linear" and "bar pressure "related offset is not a big issue. In the heat of the battle on Four Wheel Drive, or Orange Blossom Special these "errors" would probably not be noticed by the majority of players/listeners.

In the discussion of "cents", realize that cents are not linear...they are based upon the 1200th root of 2. At E4 = 329.63 Hz, one cent would be 3.29 Hz...but at F#4,F4,Eb4 it would not.

Try this for a mind bender...When a string vibrates, it must increase in length, so the larger the vibration, the more "off pitch" it must get. Then add in that it must also decrease in dia at the same time and it gets really hazy.

Back to how many cents per mil of displacement...if I get ambitious I will run the numbers based upon the sqrt of tension over mass approach.

richard burton · From: Britain

I must say that I'm feeling slightly proud of myself, my measurements and calculations are only one thousandth of an inch away from Ed's Very Happy

ed packard · From: Show Low AZ

RB...and I am happy to find agreement anywhere on this topic.

Gary Cosden · From: Florida, USA

Here is how I came up with the figure of 1.59 cents flat.
Using Ernest Bovine’s 24” scale analogy:
The 12th fret is 12” from nut. Raising that string by a linear motion of .032” makes the scale length 24.032”.
The 12” fret should be 24.016” from the nut. Difference = .016” which represents .001333 percent of the interval. (.016/12)
Since an octave has 1,200 cents .001333*1,200 = 1.59 cents (at the 12th fret)
I am not exactly Copernicus when it comes to math but I think I am correct here. If not – please correct me. I honestly want to know more about this.
At the end of the day if you sound in tune to yourself and to most of the audience you are essentially "in tune". Your actual intonation with respect to the equal tempered scale might appear to be shockingly bad by the numbers but that is, of course, irrelevant. I think it is way too easy to conclude that a linear pull changer system would be a bad thing without having some kind of understanding regarding the slew of intonation issues you face now with a conventional changer. Still, somehow, a good player can play one "in tune"! I personally feel that a linear pull setup could be designed to have a positive (albeit quite small) impact on intonation if the small issues relating to scale length were to be offset by the small issues of "compensation" were addressed in its design. I also think no one really needs such a design but its fun (for me at least)to contemplate.

Earnest Bovine · From: Los Angeles CA USA

Earnest Bovine · From: Los Angeles CA USA

Gary Cosden · From: Florida, USA

I am pretty sure that Ed is measuring the movement of the finger on the changer and converting that to the equivalent linear motion. On my Emmons PP I measured the finger movement for the low B (10th) string and drew it up in CAD. I was able to determine the string moved .041". ("a" pedal B-C#)I suspect that ED is just doing the math but I may well be wrong.

Earnest Bovine · From: Los Angeles CA USA

ed packard · From: Show Low AZ

Yep Gary...you are wrong. The Sierra session series has the lever arm keyless, which allows quite direct measurement of the string distance changes. Changes were measured by retuning the string at the keyless tuner, not by using pedals.

Earnest Bovine · From: Los Angeles CA USA

CrowBear Schmitt · From: Ariege, - PairO'knees, - France

Gary Cosden · From: Florida, USA

Ed - I have a Sierra Session model myself and I did not think of that. Thats why you make the big bucks.

Gary Cosden · From: Florida, USA

Earnest Bovine - you are correct about the Anapeg and Excel changers from what I have seen posted on this forum pertaining to their designs. The only true linear pull changer being made that I am aware of is the Lamar and from what I can see owners are happy, or at least I don't recall any noise on the forum to the contrary.

Rick Schmidt · From: Prescott AZ, USA

Quote: Chevy Chase (while impersonating Gerald Ford.)

"It was my understanding that there would be no math."

Winking

David Wright · From: Pilot Point ,Tx USA.

Well theres Jay Dee's pust pull Razz

Forum Index > Pedal Steel	Goto page 1, 2, 3 Next	Next oldest topic :: Next newest topic