Deep End DIY - My first speaker project

[Cloth Ears]

acg,

 

PLEASE! Post a warning before pictures!!!

 

I had to go change my pants...

 

23 hours ago, acg said:

 

Must say that thought ran through my mind too.

 

 

Speaker cables are built into the horns and soldered directly to the output transformers in the amplifiers so I guess short speaker cables is my preference.  But also short interconnects in general, mainly because I love the contrast that bright thoriated tungsten glow provides to the soft, warm orange glow of the power amps, and because I want to point the remote control forwards and not 90 degrees to the right...now that's lazy (and it's the primary reason).  The phonostage will barely make any light out there on the side and will therefore require longer interconnects, as will the tuner...now that thing is genuinely beautiful, especially when lit up...I wonder if I can make it fit in front too.

 

 

 

 

This photo does not come close to doing her justice...she is truly beautiful and sometimes I just turn her on at night just to look at her.

 

 

 

 

[acg]

This morning I have been revisiting this great album from my younger years:

 

 

 

Man, the things they do on this album that you can hear on a good system are crazy...the imaging, the bass, the layering, the harmony.   This song ^^^ I reckon is what a record must feel like in an ultrasonic cleaner...I know I felt like I was tripping while being washed clean by some awesome defibrilator level bass...my clothes shaking, the chair shaking, my chest and body being gently massaged by my Bass Cannons.  So playful and beautiful and why doesn't it ever sound like this in the ute?  Haha..love it.

Edited August 26, 2020 by acg

[acg]

Phonostage Fiddling

 

As part of the recent shuffle of stuff in the room, I have found a new home for the Helix 2 turntable on the side wall.  Seeing as though this is to be it's final move I thought I would do what needed to be done to get the system as optimised as possible in terms of sound.  Rather than make a wall shelf I opted to buy one and promptly screwed it to the wall in its intended location.  The height is excellent for general use.  There were, however, several issues to solve.

 

First up was getting rid of the last bit of noise from the phonostage.  The gain structure of my system is such that I need a lot of gain from the phonostage and quite a bit of gain from the preamp, so whatever comes out of the phono has to be excellent in terms of noise because it is further amplified by the preamp.  The LR Phono that I built a while back is from some pcbs made available by Pete Millet.  He did a presentation at ETF13 about this stage and this peculiar method of RIAA correction which you can find here, and the sound is very good, but there was always a little bit of "ocean noise" and hum.  I measured the ripple of the linear power supply at load at the input of the phono amplifier and it was 0.2mVac, which sounds ok, but then it has to be amplified 100 times in the phono and another 8 times in the preamplifier and suddenly that 0.2mV multiplied 800 times turns into 160mVac at the input to the amplifiers and with 110dB sensitive horns that is quite a bit of noise and hum.  I want silent...vacuum of space silent with my head in the horns because everything sounds better without background noise and if you have hiss or hum then you have less dynamic range plus it adds to listening fatigue and the stuffed ears feeling.  This is a tall order when using inductors in any circuit let alone very small signal circuit such as a phono corrector.

 

Got to get that power supply ripple down.  Thankfully, some time ago I made up a number of LiFePO4 battery power supplies (which is the go-to chemistry for low noise batteries) which are mechanically galvanically isolated from their chargers and thus from mains power.  All the safety features for the batteries are built into the charging logic which means no fires in the music room.  The chargers themselves are some big, high current linear supplies I built ages ago based on the old high current LT1083.  They knock out 19v @ 10amp each and two of them service the battery supplies for the preamp.  Noise at output (not input as previously measured) of the phono preamp is 0.4mVac at full gain, which equates to a 100 fold, or two orders of magnitude, improvement in noise output.  Ocean noise...gone...sweet utter silence...completely black.  But the hum has not gone anywhere!  Bugger.

 

 

 

Above is the LR Phono Chassis open for a heart transplant.  Same chassis constructio principles as with my other gear with vibration damped panels.  Notice I left loads of space on the power supply side of the barrier to fit the battery supplies.  At the time this chassis was built replacement front and back panels were made to facilitate the change of power supplies...just screw the old panels off and the new ones on...worked a charm.

 

 

About two days were spent chasing the hum.  Lots of measuring and grounding and other shenanigans were undertaken, all the while in contact with the manufacturer of my turntable who was bored crazy under lockdown in Sicktoria.  Never before have I been forced to chase down cause of turntable hum and what ultimately turned out to be a grounding issue and it certainly was great to have Mark offering advice and experience.  Along the way I determined just how much noise at the input to the preamp was necessary before the hum became audible and therefore had goalposts, and with guidance I tried what seems like a million combinations and permutations of grounding, thought I had traced the problem first to the turntable, then to the tonearm, the tonearm wiring, the preamp, but finally to the phonostage.  In the end I figured out that I needed to connect the phonostage chassis to signal ground just like they used to do in the good old days.  Because the chassis was now battery powered and hence had no connection to protective earth, inducted noise (50Hz) needed grounding and that was the only way that worked.  In the end, with a 0.2mV low MC cart, single ended tonearm wiring, a solid state phonostage with freaking inductors in the corrector, and 6m of single ended cable to the breadboarded preamp that threads past 100kg of transformers and inductors in one DSET amplifier power supplies, the noise at output of the preamp is 0.5mVac...you bloody ripper.  The result is utter silence like the vacuum of space, from a record player...yessss.

 

That was a job that needed doing, and by which I gained a lot of useful experience which will be very useful shortly when building the no-punches-pulled LCR Phonostage.

 

 

  

 

Here we have the Helix 2 on her new shelf with the LR Phonostage and her new power supply next door.  On the shelf below is a great big chassis I made several years ago with some enormous heatsinks.  The idea of that box is that I will put all the ancillary power supplies and chargers for various things in that box so they are all in one place rather than making clutter all over the place.  I've installed usb and network ports on the back with the hope that at some time in high school my son will have to do a tech project and I will be able to convince him that he should put an Arduino in that box and write an app that controls my extensive audio system from a phone/tablet.  I could be driving home from work, ping my audio to turn on over the networks so it is all warmed up by the time I get in...hehe, we can all dream.  Now that the noise is gone from the phono the peripherals chassis will go back to the very bottom of the rack with the LR Phono on the shelf above.

 

 

[acg]

Damnable 30Hz.

 

TT is right where I want her to be, the phono is noisless and sounding great, nothing but silence from the horns with the needle in the groove at full system gain.  Perfect right?

 

Nope.

 

These bloody Bass Cannons are capable things.  My lowest room mode is circa 30Hz based on the axial dimension of about 6m.  This means lots of 30Hz pressure at the side walls and the right amount of (sound) pressure right in the middle at the listening position, which is exactly why the chair is smack bang laser measured bang-on balls-accurate midway.  Nobody wants to deal with 30Hz cause it is just too hard.

 

This funny thing is happening...funny, not funny...dammit, not funny at all.  The TT sits on its shelf and with its built in Minus-K is super duper excellent at dealing with 30Hz vibrations coming through the wall and shelf...too easy.  But the direct airborne 30Hz assault is going straight at the cartridge/tonearm...airborne, not through the house shaking...and giving me runaway feedback at 30Hz and boy does it shake things up!  The room mode starts pumping, feedsback through the cartridge making it a bit louder and so on and so forth until windows are rattling, things are shaking off my desk and I am uncertain if I've been hit by a defibrilator or not...

 

A bit of investigation with a spectrum analyser app on my phone shows that 30Hz is about 18dB louder at the sidewalls than at the listening position.  Very easy to hear and feel when you walk from the chair to the sidewall or turntable.  Bugger.

 

So I figure there are two broad options:

 

(1) Move the tt back to the middle of the room.  This sucks...I like it where it is.

 

(2) Try to treat that 30Hz room mode with some kind of resonators and hope for 10dB attenuation at the turntable so that I can play super loud without runaway.  Might be possible if both sidewalls are treated.  Afterall it is a nasty high pressure with a lot of energy so resonators of some description may work.

 

 

Don't want to do either, but two is more difficult, so perhaps I should give it a try.  Any suggestions for taming 30Hz?  

[Jone5y]

7 hours ago, acg said:

Damnable 30Hz.

 

The room mode starts pumping, feedsback through the cartridge making it a bit louder and so on and so forth until windows are rattling, things are shaking off my desk and I am uncertain if I've been hit by a defibrilator or not...

 

Wowser !!

 

7 hours ago, acg said:

Don't want to do either, but two is more difficult, so perhaps I should give it a try.  

LOL. But having followed all of this thread, i've got a feeling...

[acg]

On 04/09/2020 at 4:17 PM, Jone5y said:

LOL. But having followed all of this thread, i've got a feeling...

It is moving in that direction...of course.

 

 

Above are three foam contenders for construction of some CPA's (Compound Panel Absorber), also know as VPR, CBA and marketed under the RPG Modex Plate family of bass absorbers.  These guys are quite thin, circa 100mm and apparently are capable of effectiveness perhaps as low as 30Hz, certainly at 40Hz, depending how they are implemented and can be configured to work broadband as high as 5kHz.  The basic formula is to glue some steel to some foam, attach it to the wall and try to make it look pretty.

 

Anyway, in the photo above you can see three different polyester foam/batt samples to which I have glued some remnant steel sheets in an effort to gauge their suitability to the task of (a) actually retaining a firm bond between poly and steel (b) the amount of give, or softness in the panels (c) identify potential issues before I lay down the cash to actually buy this stuff.  Left is Megasorber PM100, middle is Martini Absorb XHD100, right is Martini Absorb HD75 (in lieu of HD100 which was unavailable).  For these resonators, it is not the gas flow resistivity that really matters in the region of effectiveness (well it is a little, but only a little in the higher targeted absorbtion frequencies) but it is the ability of the foam to support the weight of the glued-on steel panel which may be up to 2.5mm thick and keep said steel from touching the floor/sides/anything.  Much of the diy work overseas with CBA's has been done with a poly foam that is not available in Aus, some guys have used normal old mattress foam, other fibreglass batts, some melamine foam. 

 

The original patent (and products) specified melamine foam but later moved to poly foam that is more supple.  The more supple support allows the panels to resonate lower and therefore absorb bass at lower frequencies.  I can get melamine foam in Brisbane, but why go for silver when you can have gold, hence the foam samples.  All three above are suitable.  The Megasorber product is more even in its density from one side to the other whereas the Martini Absorb products are definitely more dense one side than the other.  My understanding of the purpose of the foam is fourfold: adequately support the steel plate; allow pistonic movement; allow the steel plate to resonate; damp said resonances/movements.  So the foam needs to have enough structural integrity to actually hold the steel panel, a light enough grip on the panel to allow it to move, but a firm  enough grip to help control its movement so it does not become a noise source itself.  Sounds a reasonably fine line, and perhaps it is, but to my way of thinking best results are probably had with foam with just enough structural integrity to hold the steel plate in place.

 

Which one to get?  All three should be able to support a steel panel.  To go low I am going to start with 2.5m zinc or gal sheet.  The Megasorber stuff has the most even density and is probably the best option but that quality comes at  price which is triple the Martini Absorb products, plus the liklihood of getting anything out of Melbourne for a couple of months is almost zero.  XHD100 is unnecessarily stiff so is out of the equation, but the HD100 looks the goods.  One side is noticeably more dense than the other which to one way of thinking is an issue, but is also gives me two different sides to trial which perhaps will give better results, or at least a direction to advance the experiments. 

 

Tomorrow I will order some packs of HD100 which should be here in two or three weeks, and the bass trapping games can resume.  It will be very interesting to see how the room modes are affected and the frequency resonse at the listening chair and other parts of the room.

[BioBrian]

This is good reading - surely a hot topic for many of us.

 

My 'room' has a 6 m width, but I don't have a problem below about 60 Hz, and I wonder how much of that is due to having 2 large Rolladoors, which would be acting as limp mass membranes, with secondary roles as dispersers.

 

For your 30 Hz problem, I had wondered about cutting a hole in the wall and attaching a 2.85 m cube on the outside, filled with stuffing (doubling as an architectural statement of course). So alternatives are good news.

 

I'm sure you've done all the right research and maths as usual, so I'm not sure what you mean by "2.5m steel"? 2.5 mm thick (?) would be incredibly heavy, and I can't see the poly material supporting it.

 

I would have thought that to get the panels to vibrate, the thickness should be relative to the driver cones creating the sound in the first place - made of paper (?) and only about 0.5 mm thick? Is the thickness calculated according to the higher pressure next the walls?

 

Are you planning to have holes in the steel? Air gaps behind?

[acg]

@BioBrian My room is also 6m wide which is where the 30Hz issue is coming from...and because I want to put my "stuff" on the side walls this is the reinforcement or high pressure zone for that frequency.  You can feel the bass resonances change when you walk around the room with the volume turned right up.

 

38 minutes ago, BioBrian said:

My 'room' has a 6 m width, but I don't have a problem below about 60 Hz, and I wonder how much of that is due to having 2 large Rolladoors, which would be acting as limp mass membranes, with secondary roles as dispersers.

The Rolladoors would certainly help by letting most of the bass pass right through.  My room is "soundproofed" more or less which means these frequencies are reflected back into the room rather than passing through and those reflections create the pressure nulls and reinforcements of room modes...complete pita but necessary to listen to music any time I wish at any volume.

 

41 minutes ago, BioBrian said:

I'm sure you've done all the right research and maths as usual, so I'm not sure what you mean by "2.5m steel"? 2.5 mm thick (?) would be incredibly heavy, and I can't see the poly material supporting it.

 

I would have thought that to get the panels to vibrate, the thickness should be relative to the driver cones creating the sound in the first place - made of paper (?) and only about 0.5 mm thick? Is the thickness calculated according to the higher pressure next the walls?

 

Are you planning to have holes in the steel? Air gaps behind?

 

No holes, no air gaps, 2.5mm steel sheets for a start, no relationship to area of driver cones, these traps go right up against the wall.  The stiffer/thicker steel is necessary to get absorbtion at those lower frequencies.  Will also trial 1mm thick steel.  Depending on where they go in the room CPA/VPR absorbers are in two sizes: 2.2m x 1.2m; 1.8m x 1.2m.  A while back I had these frames  (see below) fabricated to screw together to hold the absorbers.  On the face of them I will be able to install the final room decor. 

 

 

[BioBrian]

15 minutes ago, acg said:

2.5mm steel sheets for a start

 

So each sheet for your 2.2 x 1.2 m trap would weigh about 52 kg. It looks like your edge frames will support that no worries, without the filler having to do it. Also that edge air sealing will be irrelevant (?). I'd expect the effective resonance/absorption to decrease in proportion to the amount of edge frame used, but will look forward further progress.

 

My question re the driver cones was about Mms rather than Sd - I presumed there would be a relationship between the  combined cone mass, and the absorber mass, ie would the 2.5 mm steel be limp enough to vibrate with that specific energy from the shaking air? I'm sure you've done the calcs and all that, but having a bit of 2.0 mm steel plate here as shelving, I'm a bit amazed.

 

55 minutes ago, acg said:

these traps go right up against the wall

But not quite touch it? Doesn't the rear metal sheet can vibrate too, and is part of the whole?

 

It's hard not to get excited by your design aspects, forgive me.

 

Love the infra-LP traps too.

[Red MacKay]

13 minutes ago, BioBrian said:

So each sheet for your 2.2 x 1.2 m trap would weigh about 52 kg.

 

52kgs?  Really - I don't believe this.

 

But isn't 2.5mm thick sheet too thick?  I would have thought something like printing plate sheets of aluminium would be more compliant and better suited for the movement to absorb.  From memory they are 0.2mm thick.  I will go measure one later and report back.

[acg]

3 hours ago, BioBrian said:

So each sheet for your 2.2 x 1.2 m trap would weigh about 52 kg. It looks like your edge frames will support that no worries, without the filler having to do it. Also that edge air sealing will be irrelevant (?). I'd expect the effective resonance/absorption to decrease in proportion to the amount of edge frame used, but will look forward further progress.

That weight is in the ballpark I reckon, but the steel panel must only be supported by the foam and cannot touch the frame, floor, wall, anything otherwise it will not resonate, and the resonation (wobbling) is the thing that does the heavy lifting with these traps.  Am also toying with adding a 1mm steel sheet on the back (wall) side so the total weight per absorber will be closer to 100kg.

 

3 hours ago, Red MacKay said:

52kgs?  Really - I don't believe this.

Yes, that is about the correct weight.

 

3 hours ago, BioBrian said:

My question re the driver cones was about Mms rather than Sd - I presumed there would be a relationship between the  combined cone mass, and the absorber mass, ie would the 2.5 mm steel be limp enough to vibrate with that specific energy from the shaking air? I'm sure you've done the calcs and all that, but having a bit of 2.0 mm steel plate here as shelving, I'm a bit amazed.

 

The effectiveness of these CPA/VPR has no correlation to any speaker parameters other than SPL.  They are absolutely ace at eliminating very large resonances or low frequency ringing from rooms...nothing that I am aware of that is better at anywhere near the size of these things.  According to the research (which I've purchased) the effectiveness of these things is determined by:

1. The size of the plate.  Minimum 1.5m x 1.0m is recommended, but larger will go lower.  It is to do with the eighenmodes or the frequencies at which the steel panels are able to be excited and larger panels get excited by lower frequencies (I've checked the maths).
2. The thickness of the plate.  1mm steel thickness will probably be effective down to 50Hz or 60Hz depending on how thick the foam is behind it.  2.5mm steel plate may be effective perhaps to 30Hz with large panels like mine (fingers crossed - would be a great result) and thicker steel even lower again.  However, the thicker the steel panel to less effective they are in higher frequencies, which is totally not a problem for me.
3. Suppleness of the foam and elasticity of the glue.  

 

3 hours ago, Red MacKay said:

But isn't 2.5mm thick sheet too thick?  I would have thought something like printing plate sheets of aluminium would be more compliant and better suited for the movement to absorb.  From memory they are 0.2mm thick.  I will go measure one later and report back.

 

Ultrafonic in Brisbane were kind enough to give me the measurements they did on their VPR's from some time ago.  They built some for a hospital but had them measured first in a lab and they absolutely killed it 60Hz-1kHz.  If higher absorption is needed then you just put thinner foam on the front of the steel panel and you have a broadband absorber 60Hz-5,000Hz.  These things were 70mm thick Basotect melamine foam and 1mm zincalume 1.5m x 1m in size.  I am planning 2.2m x 1.2m, 100mm thick poly foam, and circa 2.5mm steel panel in an effort to really hit the lows.  I might make it...30Hz is freakin' low and I think is just on the edge of possibility. 

[almikel]

I can't wait to see the outcome of the VPRs - the info on GearSlutz was always a bit sketchy, but given you've paid for the research, and built based on the research....

6 hours ago, acg said:

My room is "soundproofed" more or less which means these frequencies are reflected back into the room rather than passing through and those reflections create the pressure nulls and reinforcements of room modes...

Obviously with higher isolation you increase the potential for reflection back into the room...but with your approach to everything else, I'm assuming your room has multiple layers of Gyprock, greenglue (or equivalent), whisper clips or double studs etc etc...

...IME the double layer Gyprock/greenglue approach is still a "rigid" boundary for higher frequencies requiring the usual absorption approaches, but maintains the compliance of the room in the bottom end, requiring minimal management in the bottom end (say <150Hz)?...but you still have low frequency issues in your room?

 

7 hours ago, acg said:

You can feel the bass resonances change when you walk around the room with the volume turned right up.

I agree you should manage any room resonances you have, but you will hear the bass change when you walk the room at all volume levels - bass changing throughout the room is normal - unless your room is a tent or an anechoic chamber.

What you should target is smooth bass at 1 or more listening positions - the bass elsewhere doesn't matter too much...and expecting the same bass everywhere in your room is unrealistic...

 

Stick your head into a tri-corner - in normal rooms (not tents or anechoic chambers) the bass will be higher - this doesn't matter...as long as the bass is "as you like it" in other key areas of the room that you want the bass to be "as you like it"...

 

As above I'm keen to see the outcomes of the VPRs.

 

cheers

Mike

[acg]

 

11 hours ago, almikel said:

I can't wait to see the outcome of the VPRs - the info on GearSlutz was always a bit sketchy, but given you've paid for the research, and built based on the research....

 

I concur.

 

11 hours ago, almikel said:

Obviously with higher isolation you increase the potential for reflection back into the room...but with your approach to everything else, I'm assuming your room has multiple layers of Gyprock, greenglue (or equivalent), whisper clips or double studs etc etc...

...IME the double layer Gyprock/greenglue approach is still a "rigid" boundary for higher frequencies requiring the usual absorption approaches, but maintains the compliance of the room in the bottom end, requiring minimal management in the bottom end (say <150Hz)?...but you still have low frequency issues in your room?

 

 

Greenglue/gyprock layers do work to assuage the bass but probably only down to 60Hz at best, and even then not by much.  I have been quite successful sealing up the room (literally with sealant) to "soundproof" it which has been quite effective when looking at the rooms frequency response in the lower octave or two where I am getting pretty close to maximum room gain.  The room seems to ring at about 30Hz which is not noticeable at the listening chair because it is exactly in the midpoint between the side walls, but if you move to the sidewalls, where much of my equipment is, you are getting a 30Hz massage (18dB louder!).

 

I suspect that some of the issue is a floor resonance, and you can feel the floor moving with loud music, and even though it is two layers of floorboards with greenglue in between you can go downstairs and find the low frequency room modes down there as well.  It is possible, and I should measure it with the accelerometer, that the floor is also singing at about that 30Hz, though I suspect it may be a little higher, and a combination of my room height and the room height below (which is tiles on concrete slab) are combining to about 5.5m which is another near 30Hz reinforcement.  That is just a theory.  Whichever way things go, I need to reduce the low frequency ringing in the room and VPR's are Kong at this...just not sure if I can hit 30Hz with them but we will see.

 

11 hours ago, almikel said:

I agree you should manage any room resonances you have, but you will hear the bass change when you walk the room at all volume levels - bass changing throughout the room is normal - unless your room is a tent or an anechoic chamber.

What you should target is smooth bass at 1 or more listening positions - the bass elsewhere doesn't matter too much...and expecting the same bass everywhere in your room is unrealistic...

 

Stick your head into a tri-corner - in normal rooms (not tents or anechoic chambers) the bass will be higher - this doesn't matter...as long as the bass is "as you like it" in other key areas of the room that you want the bass to be "as you like it"...

 

 

I understand this, but there are a few things that will improve for me if I am able to get the low frequency ringing sorted:

• Less likely to shake the house to bits
• A cart/tonearm resonance will not be excited and my TT will sound better, plus I can keep it where I want it in the room.
• My office desk is in the (tri) corner of the room and the 40Hz/80Hz/120Hz reinforcements annoy the **** out of me there...I reckon I can fix that one

[almikel]

On 14/09/2020 at 7:49 AM, acg said:

My office desk is in the (tri) corner of the room and the 40Hz/80Hz/120Hz reinforcements annoy the **** out of me there...I reckon I can fix that one

fixing that and keeping the bass great at the listening position will be a challenge - which I'm sure you're up to

[Silent Screamer]

Just doing my annual check-in to see if there is any speaker progress...

[acg]

Happy new year peeps!  Strap yourselves in...this will be a big post.

 

So, I actually took some time off work.  Shouldn't have, but did, I was wearing myself out.  Straight to the shed on Boxing Day to start to reconfigure the lathe for turning the front half of the 1m diameter Upperbass Horns.  I had designed the table quite some time ago (like years) and completely forgot how it was all supposed to go together and there was a box of bolts and screws supposedly for this project but nothing was written down so I eased myself into the task. 

 

First, the existing lathe and table was completely dismantled and a shifted forwards on the frame to fit the larger faceplate for mounting the horns.  Table materials are 32mm MDF, same as the horns and a steel turning template that I had laser-cut a long time ago.  Took a couple of days but the table came together...

 

 

 

The garden shed is pretty full!

 

The first couple of rings on the front half of the horn wree turned with the back half of the horns some time ago.  This will give me the neatest joint between the two halves.  The 10mm steel faceplate "expander"  bolts to the normal lathe faceplate and is centred manually.  Then the first two rings are bolted to the expander using the same nuts that will be used to bolt the twos horn halves together...

 

 

 

 

Those nuts are strengthened with six M10 bolts into the back of them so nothing gets ripped apart unexpectedly.

 

 

 

If you look closely at this photo you will see that the inside face of the horn profile sits 32mm from the expander, which will make it easy to paint and trim.  I am going to have trouble with the outside face of the horns because they are bolted tight against the  expander and will take some effort to not look untidy when painted.  Security on the lathe versus an easy paint job...easy choice for me.

 

 

For those that have forgotten about what I am up to, see below for the 13 x 32mm thick rings up to a bit over 1m diameter that will form the front half of the Upperbass Horn...

 

 

 

...you can also see the lathe table-top with the steel cutting template in place.

 

Now that the table is levelled and aligned properly it is time to glue the first three of the new rings to the pair that are already formed...

 

 

 

 

All new rings will be glued in place while on the lathe...things are lined up using a couple of dowels between the new and old groups.

 

And now with the table-top in place and the freshly made cutting tool sitting in place.  There was some trepidation with this cutting layout because the cutting tool has a tendency to bite into the relatively soft MDF causing a not only a big divot but a shock to the system.  I was not sure how this was going to play out, and in the end it works BUT I need the longer handle on the tool to give me some more control of the tip and to stop it jumping about when it tries to take a bigger bite than intended (handle not pictured below).  I made big mistake by adding three new rings as the first group because for the very inside runs where new abuts old the handle rubs up against the third (outer) ring and I lose square on the template.  Only solution is to take the handle off for that little section and try to take it easy.  

 

 

If you look at the photo below you will see that I have turned the first set of three rings and then glued on a second set of three rings...if you look closely you will see a couple of big gouges near the expander plate.  These are where the damn tool bit in a couple of times and jumped about generally pinching/crushing/abusing my hands, which now look like I have been in a dust-up with someone.  Oh well...lesson learned...two rings up front when I turn the second horn, then groups of three after that. 

 

 

 

 

 Happy days!  Things are happening again.

 

 

Edited January 1, 2021 by acg

[Red MacKay]

My God that looks dangerous!  Is there no easier way of doing this?

[acg]

Could be Red, but at least for now the whole thing is stable.  Want to do it myself and this is the safest way I can reasonably do it. 

[Red MacKay]

How did you generate your curves?  Are they simple radii, or rather French Curve/ elliptical like?

 

[acg]

The horn profile is tractrix...all the horns in thi project are tractrix. 

[acg]

Got through this mornings turning without any surprises...glued another three rings on to turn tomorrow. 

 

 

At this stage I plan another two groups of two rings to finish the horn but to this point everything seems very stable and strong when spun up so depending on how tomorrow morning goes I may just throw on the final four and have all the turning for the first horn complete on Monday...a day early...if all goes to plan.  Hopefully a support roller will not be necessary to stabilise the horn when it is full size...they are a pain to get right. 

 

Edited January 2, 2021 by acg

[Red MacKay]

4 hours ago, acg said:

The horn profile is tractrix...all the horns in thi project are tractrix. 

 

So how exactly are you going to turn a tractrix curve here?  Are you using a copy turning attachment kind of arrangement from that profile jig you had made?  I am more than curious.

[acg]

Have a look at this photo Red...

 

 

You can see the tool that I have made for this job (copper and black in colour).  See those two little bolts at the back separated maybe 40mm centre to centre...those are the guides...and they are sitting against the edge of the template that will guide the final trim cuts.  So I calculated the tractrix horn profile to suit my project, put it into AutoCAD, designed the finishing tool with two guide bolts so I get the cutting angle correct at all times, and with these tool dimensions I offset the tractrix profile to suit (not a straight offset because of the two guide bolts on the tool), decided on 6mm thick steel for the template in a shape that would also reinforce/stiffen the table top and sent it of to be laser cut.  The template is designed so that the far right edge (as seen in the photo) must be perfectly aligned to the back of the lathe faceplate (so I can use a straight edge to align it) and with a short, circa 200mm straight section at the left side so I can align the template to the centreline of the spindle (it is hollow and I can use the tailstock live end at the far end of the lathe bar to align by sight). 

 

That is my copy tool and it let's me use a conventional square edge tool to rough out quickly and then a couple of finishing runs with the tool to bring it to the correct shape.

 

 

[Red MacKay]

Very clever - but jez - it does look really dangerous still.

 

Might a mini router mounted horizontal on your little topslide be safer?

 

[acg]

This is getting serious now...last four rings glued on for turning tomorrow some time...over one metre diameter.

 

 

 

 

As of tomorrow all the ring groups for the second horn will be glued together which will save time each day when the first horn is all painted and off the lathe.

 

 

 

20 hours ago, Red MacKay said:

Very clever - but jez - it does look really dangerous still.

 

Might a mini router mounted horizontal on your little topslide be safer?

 

Perhaps, but a router will buck and bite and will take its pound of flesh if it gets a chance...I did consider this at the start but in the end prefer the lathe  Beauty of the lathe is that sanding is a doddle and happens in minutes not hours/days with the router.  Plus for painting you just set the lathe to slow and hold the spray stationary...very even coverage and very efficient (for a painting novice like me). 

 

Edited January 3, 2021 by acg