Official Luthiers Forum!

http://www.ukworkshop.co.uk/forums/hygr ... 28406.html

Bit of a spin-off from grumpy's thread on the MIMF, but I just wanted to seek opinions on how consistent, and how accurate, this system would be, if you extended the length of the indicator from 400mm to say ...900mm ...

Seems to me that it would be an incredibly accurate (and very affordable ) indicator, once it had been calibrated in the first instance ...

I've made a couple of those. I 'calibrated' them using 2 digital Hygrometers and a sling psychrometer (which seemed to agree within 5% or so). They react pretty accurately to the digital hygrometers that I have, although the reaction is somewhat delayed. I suppose they mimic the wood that is in our guitars.
I think they are fine if you are prepared to build within a certain tolerance. 2 or 3% might be asking a bit much but you might say the same of digital Hygrometers.

Thanks Michael.

My thinking is that the longer the endgrain slice, the more accurate the reading would be ...so for example ...900mm would give a really accurate, precise (and repeatable) reading ...???

Is there any reason why they would go "out of calibration" ??? ... maybe there is ...but I can't see why ...

The two that I have made still seem to read fairly close to my sling psychrometer and one of the digital devices. The sling psychrometer can give different results even when taking multiple readings, one after the other. Don't know if that's my technique or ageing eyesight - sometimes the mercury can be pretty difficult to read. I now tend to rely on the wooden hygrometers but periodically check them with the other two devices. I can't be certain of how accurate any of these things are. I've returned a number of digital Hygrometers that seemed to give odd readings. I'm just hoping that I'm within something like 7 or 8%. I can't think of a reason why a wooden hygrometer would become inaccurate. If it did then I guess that our Guitars would become inaccurate as well.
I do know that I have a couple of Guitars that are stored in a non controlled room (have been for years) and they haven't cracked. Risky I know. What is odd is that they are true flat Tops ie. no doming at all.

Murrmac. Here's one of my Hygrometers:




Each division seems to correspond to around 10% humidity swing. As shown in the picture it is reading 45% RH - or at least I hope it is. I build anywhere between half a division either side of 6 'o' clock. I'm not sure that a much longer indicator will be of that much advantage but neither should it be of any harm.
If you can calibrate the indicator to a very accurate 'known' RH device it should be pretty accurate or at least accurate enough for our purposes.
You can only build to a certain accuracy. What happens to the instrument once it leaves your hands is (usually) another matter. The number of Players who take absolutely no notice of RH control might surprise you. Here in the UK It's very rare to come across humidity controlled rooms. For the majority of the year the readings are fairly benign. Winter can be the real problem. The lowest that I've read is 25%, which is real danger.

Michael, could you give the approximate dimensions of that hygrometer... length, and also how thick you made the endgrain strip, and how thick the long grain strip?

I was thinking that some iron-on veneer on the end grain could be an alternative to an actual strip of hardwood glued on....it might give the strip more "mobility" so to speak

TIA.

It's 400 mm's long. 4 mm thick, 2.5 mm endgrain Cedrela, 1.5 mm SA Boxwood.
My other one is shorter at 300 mm's. Endgrain Cedrela, long grain walnut. Unfortunately there is now a distinct crack in the end grain section although I'm quite sure it will glue up fine. I've now bought some Pot. Carbonate to test these using that method. If the result compares favourably with the Psychrometer + my digital device I'll stop worrying and declare them very accurate!

I built some. They are great. Incredibly precise in my opinion. The big problem is the first step: calibration. I yet have to find a truely precise hygrometer. I also bought a psychrometer (not sling). Reading is difficult and so is calibration. Humidity is one of those parameter that unlike length for example, change as soon as you go into the room and try to measure it. A difference of only 0.5°C makes a lot of difference in terms of humidity. So at the moment I have three different hygrometers and all of them give three different readings

These are cool, I think it would at least be real easy to spot trends in RH from across the room.
I would like to build one of these but I'm having a problem understanding the construction of it.
Can someone post a close-up picture of it, especially the "end grain strip"?
Thanks,
Rusty

It's very simple. Take a piece of long grain wood, 400 mm long x 15 mm width x 1.5 mm thick - all finished dimensions. To one side of this strip we glue our end grain pieces, 15 mm wide x 2.5 mm's thick. Of course it's fairly difficult to obtain a length of end grain at 400 mm's long, so we just cut off sections and butt join them until we have the 400 mm length. I think I used as many as 4 or 5 pieces. Use wood that is free of knots and fairly straight grained.
The ideal is to do the glue up at 45% RH. (or whatever you deem to be your correct shop humidity). Hopefully when the glue is dry the indicator will remain very straight at a RH of 45%.
I used epoxy for the glue because I found that water based glue affected the indicators position after the glue had dried. Try to use as little glue as possible i.e. just enough glue to get the job done. You certainly don't want thick glue lines affecting it's performance.
The indicator is attached at one end. I didn't bother gluing this in but simply used a push or interference fit. If I ever need to change it, it's an easy task.

Incidentaly, i was thinking of building myself one of those, but was thinking on using only spruce (because i have plenty and should correlate with tops fairly well ) is there an advantage in using hardwoods? and should the long and end-grain woods be different species?
Thanks in advance,
Miguel.

_________________
member of the guild of professional dilettantes

Carleen Hutchins had one of those in her garage workshop. She simply milled one layer off a cutoff strip from a piece of 3-ply plywood. She hung it horizontally on the wall by one end. On nice days when she had the garage door open, and the weather guy gave the R.H., she marked the wall at the free end. After a while she had it pretty nicely calibrated. The total swing was at least a foot, IIRC.

I made up something similar to control a 'dry box'. I used a short strip of cross grain spruce and a piece of tinned brass shim stock for the sensor. That switched a relay that controlled a light to heat up the air in the box. After a while the spruce cracked and came loose from the metal: it's hard gluing dissimilar stuff.

No idea, try it! I suspect it will be fine but the time it takes to glue up an indicator is very little indeed. I simply used available waste wood. You only really need the indicator and a small block of wood at one of the extreme ends, the rest is bling. Just mount it on a wall and pencil in some divisions, could hardly be simpler.

Here's my 'other' Hygrometer displaying a bit of swing to it:




Actually I tested this one by boiling a few pots on the stove, really increasing the humidity. I seem to recall that the indicator really swung off the scale! Probably overdid it, which is why it's cracked.

Good questions, and I have to say that I do not know the answer, but my gut instinct is that all that matters is that one side of the endgrain strip should be sealed, and that this could be done by simply sealing the side with any kind of moisture resistant sealer, and that it wouldn't actually require a long grain strip glued on.

If I am wrong (and I was wrong once before) I anticipate that Mario will come on to the thread and hopefully correct me

FWIW I am going to make several of these bad boys, and they will be made by ironing on a thin beech iron-on veneer to the end of a sink cut out from a beech countertop, and then cutting off a 3mm strip on a panel saw.

No need to think too hard about what to make these things out of - it's what's in your scrap box. I made these a while ago just from off-cuts of backs and tops. The cross grain (or end grain as it has been called in this thread) is just off-cuts from thicknessed panels, available in lengths up to ~400mm. These are only about 150mm. A neat trick that I learnt from Markus Schmidt is that the "drive" essentially comes from the cross grain strip, so expose plenty of it to speed the reaction time (obvious really, but thanks, Markus, for pointing that out!) Hence the long grain strip only covers part of the cross grain width. The other obvious thing is that how much bend you get depends on the stiffness of the structure - so don't build it too thick. The woods in the pic are WRC, EIR and Englemann and it's pretty obvious which is which. The rule gives an idea of the dimensions, they're glued together with epoxy, no finish, "zero" is at 45% RH and the rest is self explanatory.

_________________
Trevor Gore, Luthier. Australian hand made acoustic guitars, classical guitars; custom guitar design and build; guitar design instruction.

http://www.goreguitars.com.au

Mmh, Trevor, let me guess... if at 45% they are perfectly straight... with that bending humidity should be about 81%. Am I correct?

Anyway the backstrip that one usually use for the back halves is perfect for this. And a low and large brace is more than adequate. I noticed that spruce is more reactive to humidity changes and therefore works better for this purpose. But you get the same bending from all the different species of wood. So I must be wrong.

so just to blather on about this a bit more....i'm thinking the end grain is the strip responsible for the bending motion.
... and the long-grain strip is just there to act as a "base reference" to the end grain, and the tension between the two is what causes the curling/ brings it back to straight....

THUS i am guessing the long-grain part could be sealed, for even more accuracy/more dramatic movement...? i'm trying to think of another substance that a)adheres to wood and b)does not absorb water, and c)has some stiffness. someone mentioned that metal failed.

i love this idea; i'm making one fo sho

You might be right. Yes. This is all the resoning I did at first. Then I realized that I just did not need it. It works just perfectly the way it is, in its simplest state. It's difficult to improve perfection.

Just FTR, when I mentioned "end-grain" earlier, I meant "end-grain" , and not "cross grain".

I envisaged taking a board, for the sake of argument 400mm wide x 19mm thick, truing one end by crosscutting on a panel saw with a sharp blade, gluing a thin 19mm wide veneer strip to the cut end grain, and then cutting off a 4mm strip from the end. So you end up with a piece of end grain (not cross grain) 4mm thick which has a veneer strip glued to one side. Then, ideally, you would seal the veneer strip with something moisture proof, like contact cement.

I could be way off-beam here, but my gut feeling is that this would be a much more sensitive indicator than a piece of "cross grain" since the end grain would be much more receptive to changes in humidity than a cross grain strip.

Once my 3 super duper German-made hygrometers have arrived I will be putting this theory to the test...

I used actual end grain pieces but it seems that both types of cut work. The actual sensitivity isn't a problem, nor is the amount of swing. It's actually very easy for me to read 2 % changes in humidity, alternately simply place 2 marks either side of your central humidity of 45% and use it as a go, no go indicator. There really is no need to aim for accuracy readings of 0.5% RH or for huge swings in the indicator arm.
Your objective should focus on calibrating the indicator to a known RH.

i just made two out of poplar end grain(widest plank i happened to have), with thin maple long grain. one i glued up with 2 part epoxy, the other, with "liquid nails", which stays rubbery/flexible. i was about to use cyanoacrylate but i realized it would wick into the end grain
now i can't wait for the weather to change

Thanks, Murray, for that clarification. I couldn't make that out from the pics and I've only ever seen them cross-grained before. The ever present danger of making unfounded assumptions! I should know better...

Actually, all those pictured bent differently, due to different wood combinations, thicknesses etc. What I was trying to do was to get a fast response and a big response. The fastest is the second strip down, the biggest movement comes from the rosewood cross grain ones, but also quite slow, especially the fully cedar backed one. The bottom one was the least sensitive, but that was because the rosewood backer was quite thick and consequently quite stiff. Of course, longer gives a larger response, too, but in general, as Michael said, getting sufficient amplitude in the response is rarely an issue.

The biggest response is from materials with the largest difference in dimensional change with humidity. For a particular wood species, this is usually between the long grain direction and tangential grain direction. When different species are used this effect can be "amplified" by using a very stable long grain timber and a very unstable tangential or radial grained timber. So the wood combination with the highest "gain" is long grain against flat sawn cross grain (which is the same as tangentially oriented end grain, but less likely to fall apart on you). When you do long grain to long grain, you have a way of assessing the suitability of neck/fretboard wood combinations. Yes, it will increase the "gain", but won't do anything for "accuracy" other than make the readings easier to make.
FWIW, sealing just a single piece on one side works, to an extent, but as the piece equilibrates over the long term it will straighten out irrespective of the RH.
Time for a trip to the bathroom...

_________________
Trevor Gore, Luthier. Australian hand made acoustic guitars, classical guitars; custom guitar design and build; guitar design instruction.

http://www.goreguitars.com.au

That reasoning was exactly what went through my head, but my initial conclusion was to use a relatively unstable wood for the end grain (or cross grain) , and a relatively stable wood for the long grain.

I think I'll have to try making one...

I hate to be the wet blanket on what seems at first to be a sensitive and highly appropriate idea, but I'm afraid that wood, or at least most wood, is not well suited for this purpose.
If you measure the dimensions of a piece of wood at, say, 90% RH, then allow it to reach equilibrium at 40% RH and observe that the shrinkage is X, then expose the piece again to 90% RH, you would think the piece would expand to the original dimension, but it doesn't. It expands to X-Y. This is known as hysteresis, and explains why "seasoned" wood is more stable in service than newly dried wood.
An accurate hygrometer needs a material that shows negibile hysteresis, to be repeatable.

but isnt there a point at which even seasoned wood will respond predictably to RH...? or is that point called "petrification"?