Air leaks ? (Lucia IV P)

[Glug]

Hi everybody,

Im in the process of checking the mechanics of my new 1959 Hohner Lucia IV P, and Im currently checking it for external air leaks using methods from Accordion Revival and here.  The problem Ive got is I dont have much idea of how air tight it should be.

Im using the Accordion Revival drop test [DT] to measure progress:
25 secs          - as bought
33 secs          - treble gasket and 3 other leaks fixed, treble pallets checked

But my 1935 48 bass italian box gives 55 secs, so I dont know yet if Ive got other air leaks or thats just reasonable for a Lucia.

So has somebody please got an air tight Lucia IV P they could test, or some idea of what a reasonable number would be ?

Cheers,
Gareth


[DT] From http://accordionrevival.com/ACCORDION_REPAIR1.php#Air_Leaks

To find out if your leak is internal or external, measure the amount of time it takes the bellows to close under its own weight from the full open position with all valves closed.

If the bellows takes less than 35 seconds to close, it has some external air leaks, as nearly all accordions do.

 

[debra]

Hohner accordions are known for not being the most airtight. 33 seconds isn't bad compared to your Italian box: a small box has fewer pallets.
You can reduce the air leaking through the pallets by strengthening the springs (under the keyboard and in the bass mechanism) but that makes the accordion harder to play.

 

[Glug]

Thanks for that,
I didn't realise Hohner were know for not being airtight

I did wonder if the springs had got a bit tired, but it sounds like they're ok and it's just expected to leak a bit. I'm not going to change the keyboard action from what Hohner think it should be.

 

[Sebastian Bravo]

Im not going to change the keyboard action from what Hohner think it should be.

It couldve been a nice spring tension in 1950s, but in 2018, after 60 years of playing?

Try to check more leaks, in silence. I check air leaks at 3:00 AM, it helps. Remove the grill, open the bellows and close it with a lot of pressure, to find leaking treble notes. Then, adjust the springs of that notes, just a little!

And, to get the best airtight accordion, you need to change every leather!

 

[Glug]

Hmmm, maybe I should buy a new spring and try that to see what the difference is and if there is much difference change the rest. Didn't know adjusting the existing springs was possible, I'll look it up.

What I've had the most success with so far is masking tape:

Step 1: masking tape under bass and treble blocks - only air leaks from the bellows and case will have any effect.
Fill bellows from air button, compress and use hearing, wet lip & stethoscope to find leaks. Fix the leaks.

Step 2: Tape under bass blocks and treble pallets - only leaks from bellows and case and register slide mechanism.

Step ... you get the idea.

Think I saw that in Accordion Repairs Made Easy, which I found a pdf copy of. It makes finding bellows leaks which are small relative to the pallet leaks much easier to find.

 

[boxplayer4000]

Air Leaks.
Glug. You’re on the right track finding air leaks by isolating/taping parts one at a time so that they can be eliminated.
Air leaks can be complicated to locate and fall into two categories. ie. those that occur when notes are being played and those that occur when not being played.
Leaking pallets: My feeling is that it is much more likely to be caused by leather linings that are faulty (compacted/hardened/misaligned), rather than weakened springs. (Why would springs weaken? Their design keeps them well inside elastic limits in normal use.)
With one instrument I had I suspected the pallets but could not locate leaks with the usual methods of stethoscope, chalk dust etc.
A large, domestic (air-tight) plastic bag was pulled over the treble keyboard and and taped and sealed to the treble body near the bellows.
By working the bellows it was soon apparent the pallets were leaking as the plastic bag inflated and deflated.
The reason was minute leakage at each pallet caused by using inferior chamois leather bought from a DIY shop. The lesson was, some leathers are more air-tight than others and only materials from a proper source should be used. After re-lining the problem disappeared.. Obviously inferior leather is unlikely to be the culprate in your case but the plastic bag might help if you can’t locate leaks with any other method. Best of luck.

 

[jollyrogeraccordions]

I don’t think chamois leather is suitable. It doesn’t have a 'shiny side' and is not airtight. Also, new chamois can be a bit sticky when new...not what you need on pallets. If you wash it out, to remove the stickiness, as you know, chamois goes hard and stiff, until you wet it, so it isn’t suitable. Try and get some kid skin or goat skin...the sort used to make gloves, with some furry 'nap' on one side.

 

[boxplayer4000]

Jollyroger:
I agree there are other leathers used for pallet lining (other than chamois). Chamois is on sale here for pallet lining from a very reputable source.

 

[Glug]

Thanks for the ideas.

I'm fairly sure the pallet facings are ok, unless they're not Hohner ones.  They're the right thickness (4mm) and I checked the alignment and brushed the leather face.  I'm going to recheck the pillows in a few days when it's settled properly.  But it seems to me that Paul is saying they're expected to leak on Hohner boxes.

Previously some of them were out of alignment, as seen by looking from inside with a light behind them, and then checking the pillow markings.  So I made a wooden pallet facing, complete with two wooden pillows: just locate the pillows in the air holes and draw a line down each side with a felt tip, then it's obvious if the real pallets are aligned correctly.

This is one of the pallets that was misalligned:

I don't know if you can tell if it's the proper leather from that image.

A cunning plan I haven't tried yet: pump warm air in via the air valve and use a thermal imaging camera to find the leaks  Unfortunately thermal imaging cameras are not cheap, but it would be interesting to know if that works.

 

[jozz]

A cunning plan I havent tried yet: pump warm air in via the air valve and use a thermal imaging camera to find the leaks Unfortunately thermal imaging cameras are not cheap, but it would be interesting to know if that works.

Although that is a cool idea, I think that will not work. The imaging cam will not pickup air. It will only detect surface temps, at least the consumer level stuff does.

 

[Glug]

Yep, not convinced it would work either, but if I had a thermal camera lying around I would give it a try. I've got a spare Canon 20D, but that's the wrong band of IR.

I think one day it would be possible, technology gets better and cheaper every day.

 

[jozz]

maybe if you injected very dense fog machine fluid into it...

:mrgreen:

 

[Glug]

Might look good around Halloween .....

I did think dyes, dry ice, etc but nothing I saw was very reed friendly. And it might need a lot of cleaning after.

 

[Glug]

Update

I carried on finding leaks using masking tape to find which level the leak was in. The drop test value depended on register settings so I then decided to change the treble pallet facings to check, and I've just run the first tests:

25 secs - as bought
33 secs - treble gasket and 3 other leaks fixed, treble pallets checked (not very well apparently)
37 secs - register mechanism sealed
50 secs - new treble pallet facings (3 tests - 50, 50, 53)

all tests done with master/master registers. Base set to light, treble to H (3 voices total) gives 72 secs.

I'm now a very happy bunny.
And I haven't touched the bass yet ......

Gareth.


In case it helps somebody else:

Fixes:
Various screws and bolts - Fabri-Tac on the thread to seal
Bellows quick release levers - replaced rubber washers with rectangle of 2mm neoprene and put Fabri-Tac between neoprene and the body.
Treble gasket - replaced (3 times), 5mm x 2mm foam works the best (Hohner gasket is 4mm x 2mm rubber)
Register slider box - resealed one end with silicone
Register mechanism plastic box - varnished to seal cracks in original glue
Treble pallet facings - replaced and aligned carefully using several blunt objects.

Materials:
Treble gasket: from Charlie at CGM Musical Services
register box sealer: Technicqll Silicone HT from Amazon
Varnish: DecoArt DuraClear gloss from Amazon
Felt glue: Evo Stik 528
Pallet facings: from akkordeonwerkstatt.com via Ebay (very fast delivery)

 

[debra]

Very good job on all that sealing!
As you discovered these Hohners are leaking air in many different places. Back when I had an Atlantic IV N the biggest air los was in the bellows quick release levers. They seem like a good idea to make access for maintenance easier but they are a nightmare to get them sealed properly.
When it takes 50 secs now for the accordion to close on its own you have achieved an excellent result! It won't get much better than that, even on the finest recent Italian boxes.

 

[Glug]

Cheers, I quite like making things work at their best.
And I haven't managed to break anything yet.

If I was doing those levers again I would try the silicone gasket compound (arrived after I did them). It sticks really well to metal, and quite well to neoprene, and sets in about 10 mins.

Another benefit is I did get to try MM+H and LMM+ as voice combinations, they're not available on the actual register switches. Unless I find some way to changed that

 

[Soulsaver]

A quick question: How do you know its a 1959?
And is it this model?
Or this:

 

[Glug]

Hiya,
Its the second link, I think the first one is the 1973 model.

We established the date in https://www.accordionists.co.uk/viewtopic.php?f=25&t=6848&p=61474&hilit=lucia#p61474

Are you still trying to date the design changes ?
And which changes ?

Gareth

 

[Soulsaver]

Hi Glug. No, I've not been active on here for a while, but I do pop in now and then for a read.
I'd forgotten we'd addressed the subject before.. duh!. Thanks, they're great value IMO.

I'm still busy on a project & still retired from playing although tempted occasionally.

 

[Glug]

Update (the last)

Had to check out the bass really, otherwise it would just nag at me.
Final results (all with master/master registers):

25 secs - as bought
33 secs - treble gasket and 3 other leaks fixed, treble pallets checked
37 secs - register mechanism sealed
50 secs - new treble pallet facings, and aligned properly
55 secs - new bass pallet facings, slight alignment fix (4 tests - 55, 55, 57, 58)

There's still a slight leak at the bellows lock lever (soap test) but I can't see an easy way to fix it (it's riveted I think).

Quite a 'fun' experience doing that and I now know all the mechanics are working properly, and I know my way around the internals quite well. Must be time to play it some more ...

Gareth.


Fixes:
Bass bellows bolts - sealed with Fabri-Tac
Bass pallet facings - replaced and aligned.
Bass machine - lash set quite carefully.

NB. You can check if the bass machine lash is too small by running the drop test with the bass machine removed.