Which Tannoys, folks?

[walpurgis]

Yes, I too have come across the occasional eBay genius. It happens.


I don't know what type of adhesive you plan using for the veneer on the bendy bits, but I've built large cabinets with expensive veneer onto curved surfaces and found using a good contact adhesive (Thixofix) on both surfaces and allowed to go only just tacky dry was best for me.

Obviously great accuracy is needed to get the veneer lined up before the glued faces come into contact and 'grab', as you only get one chance.

The trick is to first lightly press the veneer down, then rub firmly along the grain with a softwood block, working from the centre of the panel outwards. This gets the veneer tightly down onto the panel with no lifted areas. I'd suggest a couple of trial pieces if you adopt this method.

Should any bubbles or lifted areas become apparent, they can be persuaded to stick down with a hot iron as long as they have had a proper adhesive coating originally.

I've had beautiful results.

[Reffc]

I'll be using PVA in all probability Geoff. The veneer is too expensive to risk mucking it up so the veneering will be done in the same way as the curved panels were formed. The PVA will be evenly applied to the plywood face, the veneer placed on and rubbed down from the centre outwards, then placed inside the vacuum former and left for at least 4 hours to set before removal. The PVA is more than strong enough and this method allows complete confidence that sufficient even pressure is applied to prevent bubbling up of the veneer. It's very time consuming as only one panel can be done at a time but hopefully by this time next week, there'll have been enough time to book match and glue the first few veneers. I'll post some photos when done.

[Reffc]

Side panels now veneered using the vacuum former and work started tidying the panels up ready for the first lacquer coat. Photos to come. Veneer looks great and really suits the profile.

Have had a think about the internal damping and reckon I can get away with slightly less except for behind the drive unit, by introducing some "V" deflector panels. The volume lost will be balanced by the volume gained omitting the wool felt to the side panels so I plan for two Vee deflectors: one in the base, to reduce unwanted standing waves along the length of the cabinets, and one set of two (ie "W" deflectors) on the rear panel behind the drive unit. The chamber containing the drive unit will be fully lagged and damped using bitumen, wool felt, acoustic foam and long lambswool. The front and rear baffle panels will also have wool felt applied to deaden them but I'll now omit this felt from the side panels as I don't want to risk over damping the panels.

[walpurgis]

There was some talk years ago about damping materials causing the speaker to 'see' a larger cabinet volume than it does without them. How factual this is I don't know, but it may be of interest.

[Reffc]

There was some talk years ago about damping materials causing the speaker to 'see' a larger cabinet volume than it does without them. How factual this is I don't know, but it may be of interest.

Its perfectly true Geoff. When you stuff something like long lambswool onto a speaker enclosure, the path travelled by the backwave from the speaker driver is increased, fooling the speaker into thinking it's in a larger cabinet than it is. You can increase apparent volume by up to 25% by stuffing the whole enclosure. Whilst this is usually a good thing for sealed cabinets (the larger the cabinet the smoother the response and the less rolled off the bass) it isn't a good thing for ported cabinets. This is because you have tuned the ports to a specific frequency and the act of increasing apparent volume will lower the resonant frequency of the tuned enclosure. "great" I hear you say, but not necessarily so.

Every driver has what is called a "free air resonance" which is it's natural harmonic resonant frequency when unloaded (ie free air or dipole conditions). If the cabinet tuning falls on or very close to this frequency then all sorts of horrible LF distortion is the result. Its important to also realise that a manufacturer's stated FAR for a loudspeaker driver is often not the same as that measured! Some Tannoy MG 12 inch drivers I know have a FAR of approximately 50Hz which is way above the stated FAR, so its important that cabinets for MGs are not tuned or resonate at 50Hz or close to that figure.

I'll bet you're sorry you asked now!

Tuning a reflex cabinet should be done as precisely as possible and the cabinet should not be over damped. The only area which you can really go to town on the damping is behind the driver to prevent unwanted reflections straight back through the driver. Everywhere else should just get the minimum needed to absorb HF reflections and mid frequency reflections. A few inches of acoustic foam or a single 35mm layer of baff/wool elsewhere is fine.

Because mine are long cabinets, I have to be concerned with half and quarter wavelengths at LF's and work out where the nodes and anti-nodes form. To reduce the possibility of these forming I have added the sloped or "Vee" panels to the design (all will become clear when the photos are posted) and also introduced damping at the critical areas such as around the ports barrels themselves and in the centre of the chamber.

Closed infinite baffle boxes are far easier to design for, especially if large. Predicting what happens with smaller boxes is more tricky and these do by and large need to be much more solid and heavily damped than large cabinets.

Beware any floor-stander that feels light. Probably an indication of being under-braced and under damped using thin panels. Particular attention needs to be paid to the front baffle which should be as acoustically dead as possible (very heavy and well damped).

[walpurgis]

You've been browsing your textbooks again Paul.

No, I'm not sorry I asked. It's nice to see that somebody else has a good grasp of speakers. They're my favourite Hi-Fi subject.

Having built around fifty speaker systems in the past I have actually learned a lot. I'm aware that to have port resonances tuned close to that of a main driver can create problems. That applies to all vented systems, not just reflex (Hemholtz) types. You can end up with a huge bass hump, cancellation and all sorts of dodgy harmonic effects.

I personally like transmission line or acoustic column (which are not quite the same thing) speakers and have built several. Horn speakers are nice, but I don't like the immediate 'backloading' Lowther tend to use as it creates all sorts of colourations due to the too close proximity of reflective surfaces. A big flare/horn coupled reflex or quarter wave horn can sound great though.

The original 12" Tannoy Monitor Gold, with the pleated paper cone surround was basically a Monitor Red with higher power handling and at one time had a quoted free air resonance of 23 Hz. which I never believed, nearer 40Hz. I'd guess. The next version with the rubber roll surround had a lower resonance and may well have been around 30Hz. The last 12" gold had the unfortunate 'Tanoplas' foam surround which was slightly more pliant and might well have produced a resonance getting down into the twenties. Although I've owned these, I never bothered to measure their characteristics.

The quoted resonance for the little IIILZ was the biggest porky, Tannoy claimed 30Hz, but my guess would be getting on for twice that. You just could not get proper deep bass out of them. I've used these in pretty big cabinets and their limitations were obvious.

[Reffc]

Yes, I have a soft spot for 1/4 wave loaded speakers. Castle made quite a few along these lines and I owned three pairs (Chesters and few pairs of Howards) and the Howard S2's were amongst my all time favourite loudspeakers although the treble could be a little coarse.

Tannoy claim 26Hz for the HPD and a similar figure for the MG which simply cannot be true. The HPD was engineered for more (and lower) bass so I can't and don't believe that the later MGs achieve the same power handling or bass response (they don't). The problem with HPDs is that unless the cabinet is optimised to make the best of that response, they can sound toppy and require some increase in the HF series resistor, and really benefit from a notch filter at the crossover point. (in fact ALL Tannoys benefit from a notch filter to cure the impedance characteristics at the crossover point to tame the honk!). Once you have the HPDs in proper cabinets, they start to make far more sense as the balance is restored. You simply cannot achieve this in undersized cabinets with them, without getting a very peaky bass response. 150 litres or more for a vented enclosure is needed. Anywhere between 150 and 170 litres is fine. If you make the cabinets any bigger, the bass starts to roll off and if you tune for a lower resonant frequency, you cross over the drivers own FAR, which ruins the performance. It's no accident that Alex Garner chose 150 litres as the optimum for the HPD 12 inch drivers (and it's also pretty much the optimum for 12 inch MGs, although their response stays a little smoother down to say 130 litres, so they can be used to good effect in smaller enclosures but wont have the bass of the HPDs).

The other thing to remember when DIY-ing loudspeaker cabinets is that baffle width and shape matters! if the front baffle is too narrow, then the frequency at which baffle step loss occurs is higher than for a wider baffle. The can be easily calculated but the important thing is to remember that when radiating into full space as opposed to half space a 6dB loss occurs at the transition frequency from half to full space. Most crossovers for commercial speakers are designed to compensate for the "apparent" rise in frequencies occurring at twice the width of the baffle (so for a baffle of 600mm wide, maximum amplitude occurs at 1.1kHz which also happens to be close to the crossover point of many Tannoy crossovers).

This last point does mean that crossovers may need to be tweaked to account for the amplitude rise. It's also another reason why the same Tannoy drivers and crossovers can sound much brighter in some cabinets than in others. There's a hell of a lot to take in when considering cabinet and crossover design!

[Reffc]

Finally had a chance to cobble together an update. Not a great deal of progress as it's been really time consuming getting all the laminations done then doing the veneering and shaping each individual brace for an exact fit.

Anyway, some photos:


Cabinets look enormous sat on these 3 inch trolley plinths! Just the right size for a 100 BHP Mercury outboard motor on the back of each!

Base and top panels cut, joints formed and glued to veneered side panels:



Side view of one of the veneered side panels. Its only up to a 20 or 30 sheen, so the grain will come out more with some more finishing work. That won't happen until I have the interiors fitted out and the front and back panels are completed. Still, I like the ripple Birch. Very contemporary and light and I reckon it'll be a nice contrast with the Black Walnut front edges:



Close up on one of the panels. The veneers have been cut and matched as well as possible before going into the vacuum press for gluing:



Front view...braces removed for the moment whilst top an bottom panels are glued:



Side view:




Detail of top panel joint:



Front of cabinet. 30mm deep baffle panel will be glued into this recess formed by the sloped face and the straight (internal) face beyon the step you can just about make out behind the strapping. A really awkward cut to make:



Rear view:



The cabs will sit there until the glue has set and hardened so 24 to 48 hours before they're touched. Next job will be to fit the rear panel after first forming a cut-out in the lower panel for the removable section housing the crossover and binding posts. On the base internally there'll be a sloped internal baffle cut to shape from 24mm ply and the void between it and the cabinet base packed with wool felt. The slope is being put in to ensure that the internal base is not parallel to the top panel to help avoid standing waves. Won't have that issue with the curved sides!

It'll be a week to 10 days before I update with more photos and should be close to completion by then.

[Alex_UK]

Just the right size for a 100 BHP Mercury outboard motor on the back of each!

Top work though Paul - I really admire people who can do this sort of job to a decent standard - apart from the time issue, I just know I'd cock it up somehow. Look forward to seeing the next instalment but they look very professional so far.

[Reffc]

Thanks Alex. I am getting help with them as parts of the construction are much easier as a two man job, but it's not all been plain sailing with some scrapped panels due to the odd mistake, but we're getting there! I've a fair amount of internal work to do to them following the rear panel being fitted before the baffle can go on. To help speed things up, I'll be getting on with the internals whilst my colleague gets on with the front panel.