Piano World Home Page Forums Our Most Popular Forums Piano Tuner-Technicians Forum Temp/humidity control for Mason and Hamlin BB

I am having trouble with the effects of "stage lighting" and "lighting technicians" being in conflict with piano tuners.

First of all, in the vast majority of piano performances, there is no "lighting technician" involved in the process. "Stage lighting" in a concert/recital hall is a whole different ball of wax than an opera house or a theater.

In the concert/recital hall the lighting is of a fixed design. It does not change from performance to performance. There are two systems involved: "Stage Lights" and "Work Lights." Stage lights are used for performances and rehearsals, while work lights are used when the "stage hands" are doing such work as setting up risers, chairs, music racks, and various other paraphernalia needed for a given performance. Depending on the type of lighting fixture used for "Stage Lights," there can be, and often is, a temperature rise in the performance area. Other than during rehearsals and performances, the temperature effect of "Work Lights" on the stage or piano is minimal. By that I mean, it has no more effect on the piano than does the lighting in a home.

For a solo recital, the piano is tuned in situ and most often in "Work Light" conditions. There is no way that the finances will be expended to light the stage in the same way as it is in performance or rehearsal. Thus, there is a temperature change and this is not a new phenomenon. Concert tuners have been dealing with the situation for as long as they have been in business.

Often a concert hall will have a "Piano Garage" for performance pianos. It is a separate room with controlled temperature and humidity to keep the pianos in the most stable environment to insure that the instruments are as "tuning stable" as they possibly can be. In a concerto performance, the piano is usually tuned, multiple times, in this area. Remember, it is not the norm to have the piano sitting on the stage for the entire concert. It is brought to the stage prior to a piano/orchestra performance, and is removed afterward. It is for this simple reason that concerti are programmed just prior to intermission or at the end of the concert. It is less disruptive to the audience to move the piano as little as possible within the performance sections of the programs. Intermissions are a great opportunity to shuttle the piano.

In multipurpose facilities, the stage lighting for a classical performance is of fixed design. Most often, it is built into the orchestra shell, and even if that is not the case, the "lighting plot," the specifications for locations and types of lighting instruments, is unchanging. The lighting for the stage is the same as it was last week, and will be the same next week.

The assumption that the stage is usually in an uproar is simply not true for the vast majority of performances. Things can, and do, get a little crazy in the situation of a festival with numerous different performing ensembles. However, for all of the professionals involved, those events are not the norm.

Many of the newest of performance facilities are using the latest LED technology and this completely negates the temperature variances due to high wattage stage lighting fixtures. It will greatly help with the stability of all tuned instruments and then maintaining a stable RH in relation to temperature will be much easier to attain.

However, based on my experience with the D-C systems, I would still recommend its installation for the OP. After all, that is the topic.

I have checked many, many sources for the rate of change in the Equilibrium Moisture Content of wood, in particular, spruce, and, as a result, the dimensional change of the wood both tangentially and radially. No where is the EMC and dimensional change mentioned as significant in the very short term, i.e., two hours. All cited changes in the +1% range occur over several weeks to a year, depending on whether or not the environment is enclosed and heated or open and unheated, which is consistent with the often mentioned advice to piano owners to let the piano acclimatize to the new environment for several weeks before having it tuned/regulated.

So let's think about this - steel and wood expand when the temperature increases. This is related to atomic motion - literally the electrons are moving around faster and occupy more space.

If the plate expands, or the case, or the wood - all would serve to increase tension on the strings and make the piano sharper.

Instead the pitch goes flat. This is because the soundboard is loosing crown as the humidity leaves the wood.

The increase in temperature is affecting the relative humidity about the soundboard, and the change in shape of the soundboard modifies the pitch of the piano by going flat.

It would take far more than 2 minutes for the plate to come to equilibrium and change the pitch.

My guess is that the rate and amplitude of the dimensional change, over a short time span (2 hours) in the string length due to added energy (heat) to the system (piano) is several orders of magnitude higher than the same changes due to the change in the Equilibrium Moisture Content of the soundboard and peripheral wood over the same time span as a result of the same input energy.

In other words, sure, the RH in the environment might reduce by 10-15% in the two hours the stage lights are on over the piano, but the metal will expand far more than the wood will shrink.

Strings, Grotriman?

What interests me about this discussion is that there is a huge amount of data available on moisture content in wood, how to dry it, protecting it, what the amount of EMC in the wood is for a given RH and temperature - everything a house, boat, or furniture builder could ever want to know, including precise data, measured in 1/000ths of an inch of the seasonal variations expected in every region of the world of just about every species of wood, but NOTHING about short time span (2 hours) changes.

Exactly. Grotriman, you acknowledge that metal reacts quickly to temperature changes - well, what are strings made out of? And, given the differences in mass and density between the plate and the strings, what do you think will react first?

If you're still skeptical try this experiment: On a well tuned unison on your piano (or any piano you have access to), rub your finger on one string of the unison for 10 seconds, then play the unison. Wait a couple minutes and play it again. What happened? Increasing the temperature of the string (through friction in this instance) caused the pitch to drop, throwing the unison out. After a couple minutes, the temperature returns to normal and the pitch of the string returns as well. Note that your finger had no effect on the humidity of the soundboard .

Now try the same experiment on the plate. I doubt you'll be able to throw unisons out by rubbing the plate for 10 seconds.

YES, prout, that is what has fascinated me for a long time now: there is a lot of well-reseachered data out there on spruce and EMC, just not exactly what we want to know regarding its relationship to the piano!

We could probably test this--and produce some data--if someone has an idea of how to stabilise RH/temp in an controlled environment (e.g., a small box or something). Any ideas?

Adamp88 - I will try your experiment. I don't want to put the acidity of my fingers on the string but I'll try through a cloth. However you are speaking of a 20C change in temperature which in this case is extreme.

So - I'm going by the physical properties. For a vibrating string the equation is L=1/2f * Sqrt(T/mu) where L is length, f frequency, T tension and mu=linear density of the string.

Assume L1 is one meter. A change in five degrees and a starting pitch of a=440 just to rattle 443's cage .

Now find the resulting frequency of L2 given the thermal expansion coefficient of steel is 0.000016 m/m*C the calculation shows that a five degree C change will result in a 0.0001M change in the length of the string. Now the problem is I don't know what the new T would be, but I will assume the same magnitude of change (can anybody help here?) T2 would be T1/1.0001.

Setting up the two equations you get two equations in two unknowns

L1=1/(2*f1) * sqrt(T1/mu)
L2=1/(2*f2) * sqrt (T1/(1.0001*mu))

We know L1 and L1 (1, and 1.0001 meter respectively). We know f1, so solve for f2

4*f1^2 = T1/mu
1.0001*4*f2^2= T/(1.0001*mu)

rearrange

f2^2= (440)^2*4/4.0008

f2=439.9

Barely a shift. Assuming the change in tension assumption is correct. Alternatively if we can find out the tension of a string at 442.5 and at 443 (2 cents off) we could just back substitute and find out the change in length. (working this now).

I'm not quite sure where you get a 20C change from, Grotriman. Regardless, even if your equation is taking everything into account that it should, a change of .1hz is still roughly .4 cents (at A4), which is quite outside the tolerances necessary for fine tuning.

Hello . About 4 hours for the plate to expand ans slightly raise the lowered pitch . It does not get back to original however.

If you have an ETD just laying the thumb for 20 seconds on a string will show some lowering..

Interesting.. i have read those tests made in the 90 . It was in PTG journal.

Regards

FINALLY, some data that helps.

Moisture Content eventually comes to an Equilibrium with air moisture, approximately, in the case of sitka spruce to

relative humidity % 30 35 40 45 50 55 60 65 70 75 80
moisture content % 6.2 6.9 7.7 8.5 9.2 10.1 11.0 12.0 13.1 14.4 16.0

(Notice that temperature is not a factor. The variation in MC, for a given RH, due to a temperature change of ±10 C is ≈0.1)

At MC lower than the fiber saturation point (usually 35%), moisture change takes place by diffusion within the wood. The standard diffusion equation may be written as

t = L² / D

where

D = 1x10-6 cm²/s transverse and radial.
L is the length along the direction of diffusion.
t is the time to 1/e of the moisture change, that is to 63% of the equilibrium change.

So, if you have a soundboard approx. 1 cm thick that is at 8.5% MC (RH 40%) and you want to estimate how fast it will come to equilibrium in your home at 70% RH (13.1% MC) if exposed to air both sides, L = 0.5 cm and the diffusion equation gives t = 2.5x105 s, or about 3 days. The equilibrium MC change required is 4.6%, so in 3 days you can expect 63% of 4.6% = 2.9% higher MC, that is 11.4% total MC. That leaves 1.7% to go, and you can expect 63% of that 1.7% to take place over the next 3 days, to 12.4% MC. So, a little over 1 week should be enough time for 1 cm thick wood to come into equilibrium. According to the literature, most seasoned temperate woods change moisture at a rate within ±20% of this.

The MC change over two hours for a soundboard in the environment above, would be negligible, on the order of 0.08%.

Source: http://web.ncf.ca/bf250/wetwood.html

Edit: "Approximate" symbol not shown

String length does not change when temperature changes, only the tension changes. You need the tensile modulus to figure this out.

Kees

Ohhhhh...that is something that I have never thought about considering...in other words: is the piano's string length actually shorter than what it appears to be because there is tension involved?!?

See what happens when we all take the time to have a thoughtful discussion; it's nice, right?!? People from different fields actually have a lot to contribute. :-D

prout, THANK YOU SO MUCH for finding info on soundboard EMC equilibrium! I am still mentally processing the info--I'm sure I'll have some questions for you later so I can do the math on my own and adjust for variables.

We'd also need to factor in the bridge height, but it looks like this can easily be calculated...

Any info/idea on how a layer of lacquer/polyester/shellac would slow down this diffusion (i.e., how to calculate for this)?

Only that it does slow down the diffusion. Too many variables.

Grotriman, prout, et.al, if anyone could figure out the tension issue to predict the change in pitch via temperature, that would be amazingly helpful...to many technicians!

I guess, though, it would need to be the entire length of the string, not just the speaking length. Right? Since the whole thing is expanding...

The plate also needs to be factored in. It takes more time to change, but I know from experience [a very unfortunate one] that it too matters and has a big impact on the pitch level.

Are you familiar with Don Gilmore's "Self-tuning Piano"? The concept is each string is heated, which adjusts the pitch to a pre-set tuning. He would probably have some of the information you are seeking.

I did the calculation based on these numbers which I happen to have for an A4 string on my piano.
-expansion coeff: 1.6e-5 /C
- Young modulus 1.75e11 N/m^2
- mass density 7850 kg/m^3
- freq 440Hz
- speaking length .375m

I get 2.8 cent pitch drop per degree (assuming only string changes in temp). Seems high, anyone who can check this?

Added: I just tried something on the A4 string. I measured pitch then applied hot air from heater for about 5 sec towards the piano from about 50cm away, and the pitch had dropped by a whopping 13 cent. Maybe 3 cent/degree is correct.

Kees

lol...now, you're cooking with gas!!!

But, this is where I am lost...math and I are not friends. :-(

On a related topic: if the ambient temperature is 20C, what is the surface temperature of the steel wire? Or, what should it be? Same question for the plate: what should the surface temperature be?