| Speedsmile |
Well first post here, I've to say I'm reading along here for a while and have learned a lot in the process.
Some 3 years ago I built 2 floorstanders from a kit (Vifa Filligran), these are still playing in my room as I speak, and I'm very happy with them.
However my hands started itching again and with the last exams in the pocket I have the time to do a little project again.
This time I want to built a small portable but still good quality speaker for use with my PC, and eventually to take it with me to a 3.5 month internship I'll be doing in Norway starting in September. The thought of being without decent quality sound for over 3 months kind of made me nervous.... :whazzat: :eek:
So after some research I came to the conclusion that the FR125S is a very promising driver, and given my space/portability constraints the 4.5 litre aperiodic from planet10 seems like a perfect fit.
The problem is, the proposed enclosure on the planet side doesn't really fit my size constraints regarding depth and width. Therefore I redesigned the volume to a taller box, in the process dismissing the trapezoidal form factor. :smash: So the questions:
- Is it correct that this form factor was mainly chosen to reflect the waves going to the backwall down, and thus preventing reflected radiation through the driver itself as much as possible?
- How much of a problem am I going to get with my design (shown below), and are there any extra steps I can take to prevent this?
- As you can see I placed the 'vent' on the back, but I'm wondering if this is going to cause problems if I place them very close against a wall?. maybe it is better to vent to the side?
So to flourish this long story up a bit , here a few cad drawings, and a 3dsmax render of the finish I had in mind (nevermind the Reference
;) :clown: ).
  
I'd like to add my mind is set on these drivers, and already ordered them, mainly because if this project is not going to work out I always can build something with less compromises like the Fonkens out of them.
Furthermore, if anyone is interested, because I'm doing a master in Mechanical Engineering I've access to quite a few fancy numerical (FEM) programs, and I plan on doing a (structural) eigenmode analysis on this design in the weekend just for the fun of it... |
|
|
| planet10 |
| quote: | Originally posted by Speedsmile
- Is it correct that this form factor was mainly chosen to reflect the waves going to the backwall down, and thus preventing reflected radiation through the driver itself as much as possible?
- How much of a problem am I going to get with my design (shown below), and are there any extra steps I can take to prevent this?
- As you can see I placed the 'vent' on the back, but I'm wondering if this is going to cause problems if I place them very close against a wall?. maybe it is better to vent to the side?
|
1/ yes... the shape can be mutated as required
2/ you might have to play with damping a bit more
3/ as long as there is almost any clearance behing the terminus you should be fine.
dave |
|
|
| chrisb |
Well, I can see that Dave sniped in a quick reply to your questions, so I 'd add the following:
This is quite a surprising performer for its size, and makes for a great nearfield / computer monitor or as part of a HT system with stereo subs, but it will not begin to approach the same driver in the mini-Onken design. (FWIW, allow me to be anal for a minute, as the christener of the design's name; the F in Fonken family moniker designates the use of Fostex drivers :rolleyes: )
If you do build the little guys and find the performance missing a bit, and if the complexity of the onken style remains a deterring factor (having built more than couple pairs of them, I can understand that), you could always consider the more conventional flat front paneled GR design. In or listening comparisons, it's quite close to the beveled front design.
Of course if you're very tight for space, either of the larger cabinets will require some creative juggling - they definitely like several feet of free space surrounding them, and a decent floor stand.
It will be interesting to see what your FEM analysis of the design predicts, and how that correlates (or not?) to your final listening impressions of the speakers in their final environment.
|
|
|
| OzMikeH |
If space is limited you don't have to stick with a small enclosure, make it a 2 piece cabinet, the top half (with the driver) smaller than the bottom half.
Picture the top half as a top hat shape, with a flange.
The bottom half is a larger box that the top sits on and is screwed down to.
for transport you unscrew the top hat, tip it upside down and stick it inside the bucket. As a bonus it protects the driver. |
|
|
| Speedsmile |
Thank you for your comments, so as I understand it I can get away with this configuration without to much problems, nevertheless plan on making a test cabinet from some scrap wood I've lying around, just to get a feel for the overall speaker.
I just finished doing a quick eigenmode analysis in Unigraphics NX4, the results can be found under the link.
Nothing unexpected though, the modes are mainly equivalent to the usual plate modes. I also want to make a analysis with the driver itself in place, but I don't have enough info on the basket material and bulk weight of the driver to do that now, does someone have more info on this subject?
Eigenmode analysis
So I would think that if this is any indication on the sound of the speaker, it would sound a little bit brighter in the upper mid range 600-1000 [Hz], which is not necessarily a bad thing.
Furthermore for the folks who are less familiar with structural vibration modes I made some movies of the first 4 modes to illustrate them...
Mode 1
Mode 2
Mode 3 Mode 4:cool: |
|
|
| consort_ee_um |
Nice pictures and nice FEM modelling Not sure I believe the bending mode because of the stiffness of the corners will tend to stop that but the panel "breathing" modes seem right.
Can you try fitting bracing to the model and see what happens?
The FR125 has a phase plug and air can leak out between the coil and plug i.e there is no dustcap. The cabinet loss due to leakage is therefore high which affects any bass alignment you care to try. |
|
|
| Speedsmile |
| quote: | Originally posted by consort_ee_um
Nice pictures and nice FEM modelling Not sure I believe the bending mode because of the stiffness of the corners will tend to stop that but the panel "breathing" modes seem right.
Can you try fitting bracing to the model and see what happens?
The FR125 has a phase plug and air can leak out between the coil and plug i.e there is no dustcap. The cabinet loss due to leakage is therefore high which affects any bass alignment you care to try. |
You're right regarding the 2nd and 4rd mode, these will probably not occur very fast, and only do show up this low because there is such a big hole in the upper part of the cabinet, with the drivers modelled along I expect them to completely disappear. |
|
|
| binarywhisper |
I would think that bracing would make a substantial difference in the modeled results.
You think much air passes thought the voice coil gap? I'd be interested to see that modeled.
I have a pair of 12" woofers with no dust caps running in sealed boxes. I really can't imagine that much air passes thought that space. Many of the vintage woofers had very flimsy dust caps and I've never seen one appear to act, or react, in any way. Not very scientific :) but none the less, being a huge fan of sealed inclosure's I would have thought I'd have seen a dented cap on one of many old 12"ers I've owned at least try and pop out when I was thrashing out the metal tunes on em.
It's an interesting thought. Sounds like a great project for an engineer to figure out :) |
|
|
| Speedsmile |
| quote: | Originally posted by binarywhisper
I would think that bracing would make a substantial difference in the modeled results.
You think much air passes thought the voice coil gap? I'd be interested to see that modeled.
.........
It's an interesting thought. Sounds like a great project for an engineer to figure out :) |
1. Bracing will probably make a significant difference, the problem is, I don't have much room for it. However as you see in the drawing I'm planning to use 12 mm MDF on the front and 8 mm for the remaining of the cabinet, an interesting question could be what would be more (weight/volume) efficient? Partially braced 8 mm, or plain 12 mm.
But I'm not planning to model that before I know more of the driver itself because the metal will seriously effect the stiffness in that part of the cabinet. So again, anyone of the CSS owners..how much does that driver weigh? CSS doesn't list it in the spec, searching on internet gives 1.3 kg as an approximate...
2. I think the airgap around the voicecoil will not have a very big impact if you consider how small it is, furthermore the spider limits airflow even further. If I had some exact dimensions I could probably model the static flow through it, and from the timescale involved say something about this effect in dynamic behavior, but I doubt it will be small enough to effect say 60 Hz bass. A thought experiment with a 'squeezing' baloon shows exactly what I mean....;)
3. Indeed a great project for someone like me, finally all things I learned come to good use....:idea: |
|
|
| binarywhisper |
Sorry, I don't have any non-installed css125's to weigh although I have 5 of them ordered.
They do not have a metal basket, its plastic so I'm not sure how much reinforcement they will add to the hole. Some obviously but its not a particularly ridged plastic although it does appear quite inert.
well on a box that small I don't think there is a need for a regular brace. Two, maybe three hardwood dowels would probably be plenty. Use one as a cross brace, another from the magnet to the back wall of the box and maybe a third bracing the front panel to the rear below the driver. Just a thought. I'd personally brace the driver against the back panel at the least.
If my drivers show up before yours I'll weigh them and post. |
|
|
| Speedsmile |
| quote: | Originally posted by binarywhisper
Sorry, I don't have any non-installed css125's to weigh although I have 5 of them ordered.
They do not have a metal basket, its plastic so I'm not sure how much reinforcement they will add to the hole. Some obviously but its not a particularly ridged plastic although it does appear quite inert.
well on a box that small I don't think there is a need for a regular brace. Two, maybe three hardwood dowels would probably be plenty. Use one as a cross brace, another from the magnet to the back wall of the box and maybe a third bracing the front panel to the rear below the driver. Just a thought. I'd personally brace the driver against the back panel at the least.
If my drivers show up before yours I'll weigh them and post. |
1. I just got mine, and to my surprise they actually have a metal basket...which is stiff enough to indeed alter the whole analysis.
I wonder how you came to think these were made off plastic? :confused:
2. I also think that bracing will not be necessary in this small box, however the drivers are heavier then I suspected, so a front to back brace is probably necessary.
The drivers are playing in as I speak, and the sensitivity is indeed not very good, but that was to be expected. I'm surprised by the sound however, especially the treble is better then I suspected. They have a very laid back sound to them (IMO), which is very nice for the purpose I intend to use them. |
|
|
| binarywhisper |
| quote: | Originally posted by Speedsmile
1. I just got mine, and to my surprise they actually have a metal basket...which is stiff enough to indeed alter the whole analysis.
I wonder how you came to think these were made off plastic? :confused: |
Well I probably came to that conclusion because I've owned a couple of them for quite sometime and they have always had a plastic basket ... till I read this and went over and pulled a speaker. Somehow you claiming yours was metal converted mine to metal. Thank you but in the future I'd prefer if you ask before you modify my drivers telepathically.
Thinking about it they have been installed in their current box for over a year and somehow in that time I decided they were plasitic baskets. I id notice while removing and reinstalling the driver just now that the basket flexes quite a bit as the pressure of the screws is applied.
| quote: |
The drivers are playing in as I speak, and the sensitivity is indeed not very good, but that was to be expected. I'm surprised by the sound however, especially the treble is better then I suspected. They have a very laid back sound to them (IMO), which is very nice for the purpose I intend to use them. |
they have quite a long breakin period during which they get smoother and if memory serves the bottom end got a bit fuller, deeper and faster but I could be wrong about that :) |
|
|
| planet10 |
| quote: | Originally posted by Speedsmile
1. Bracing will probably make a significant difference, the problem is, I don't have much room for it. However as you see in the drawing I'm planning to use 12 mm MDF on the front and 8 mm for the remaining of the cabinet, an interesting question could be what would be more (weight/volume) efficient? Partially braced 8 mm, or plain 12 mm. |
I strongly recommend plywood -- particularily is this thin. If you have to use MDF at least laminate it both sides with a skin of something like fiberglass.
A driver magnet brace never hurts, but in a box with the goal of small it occupies significant space.
| quote: | | how much does that driver weigh? CSS doesn't list it in the spec, searching on internet gives 1.3 kg as an approximate... |
The FR125S weighs 972g, the FR125SR 981g. The metal baskets (not plastic) are different. The old driver has 6 legs the new RoHS driver 4.
dave |
|
|
| planet10 |
| quote: | Originally posted by binarywhisper
well on a box that small I don't think there is a need for a regular brace. Two, maybe three hardwood dowels would probably be plenty. Use one as a cross brace, another from the magnet to the back wall of the box and maybe a third bracing the front panel to the rear below the driver. Just a thought. I'd personally brace the driver against the back panel at the least. |
Dowels are surprisingly ineffective...
dave |
|
|
| planet10 |
| quote: | Originally posted by Speedsmile
The drivers are playing in as I speak, and the sensitivity is indeed not very good, but that was to be expected. I'm surprised by the sound however, especially the treble is better then I suspected. They have a very laid back sound to them (IMO), which is very nice for the purpose I intend to use them. |
The low sensitivity is the biggest detraction from the FR125. It doesn't like the amps that i tend to prefer. When it comes to my recommending FR125 or FE127 is usually straight forward based on the amp the person has.
Also, a couple vanishingly thing coats of C37 dramatically improve the FR125. Puzzlekoat is almost as good (but there are potential gotchas with installation) and maxro is about to embark on an enabling of his (i expect even just the gloss coat will have similar affects as the C37 or PK)
dave |
|
|
| Speedsmile |
| quote: | Originally posted by planet10
I strongly recommend plywood -- particularily is this thin. If you have to use MDF at least laminate it both sides with a skin of something like fiberglass.
A driver magnet brace never hurts, but in a box with the goal of small it occupies significant space.
The FR125S weighs 972g, the FR125SR 981g. The metal baskets (not plastic) are different. The old driver has 6 legs the new RoHS driver 4.
dave |
Thanks for the info, and I agree that any bracing will bring the cabinet volume significantly down when using the same outer dimensions, so I want to avoid that for now.
Regarding the plywood: Maybe you're right, a quick re-analysis showed the first eigenmode was excited at approx. 1200 Hz using plywood, so about double the 547 Hz of the MDF enclosure. I will fiddle with the question a little longer and see what 12 mm MDF brings. I prefer to use MDF with respect to finishing, I never worked in a precise way with plywood... |
|
|
| planet10 |
| quote: | Originally posted by Speedsmile
Regarding the plywood: Maybe you're right, a quick re-analysis showed the first eigenmode was excited at approx. 1200 Hz using plywood, so about double the 547 Hz of the MDF enclosure. |
That is a very interesting analytical result. With the energy available to excite a resonance decreasing at approx the square of the frequency, that gives plywood a 4-fold advantage over the MDF.
I'll let Chris comment of the difference between finishing ply & MDF...
dave |
|
|
| Peter Menting |
Dave; "a couple of vanishingly thin coats of C37"-----not familiar with the product or the process. Could you elaborate?
Pete |
|
|
| chrisb |
| quote: | Originally posted by planet10
I'll let Chris comment of the difference between finishing ply & MDF...
dave |
here we go again ...........
my 2 cents worth:
depending on the grade of finish you're looking for, MDF can be as easy as a quick couple of coats of texture spray paint and clear top coat polyurethane, or a whole 'nother world of pain for high gloss "piano" grade finish.
As others have noted extensively in another thread, conditioning, sealing, priming and prepping MDF for a high quality, durable paint job can easily take several times longer than the enclosure fabrication. I have neither the skill or patience to undertake this, nor frankly any appreciation for the end result - except for window and door trims, I just don't care for painted wood. .
Except for car audio boxes where contour or compound radiuses are common and painting / carpeting are only real options, I tend to like to cover all my enclosures with wood veneers.
Finishing then becomes a question of stain or not?, and then several spray coats of post catalyzed nitro cellulose lacquer. As Ed Schilling would say: "After over 10 years of doing this, I've found something that works for me - so 'nuff said" |
|
|
| planet10 |
| quote: | Originally posted by Peter Menting
"a couple of vanishingly thin coats of C37"-----not familiar with the product or the process. Could you elaborate?
|
http://www.ennemoser.com/
C37 is a musical instrument lacquer. It gets a lot of stain as a snake-oil product because of the nature of the claims made, and the (supposed) high price. Yet a number of well respected audio people rave about it.
I don't care about the claims, just the results.
A while back (probably close to 2 years ago) a diyAudio member bought an ounce & sent it to SY for chemical analysis and then the rest was sent here for real world trials. I have been playing with it for about a year and keeping results close to my chest.
I found it not to be a panacea, but in some cases did make dramatic improvements. Puzzlekoat -- which i have been using for 30 years -- was used as a benchmark.
How well it works seems almost to parallel how easy it flows onto the substrate being done (you use a brush). For example, it goes onto a FR125 cone (or a Visaton B200) readily and can be spread very thin very easily. On a Fostex cone it actually a struggle to get it to cover the cone... on the FR the C37 edges the puzzlekoat , on the Fostex those results flip.
In both cases treated beats untreated.
As to cost, the problem really lies in the minimum amount you can buy. An ounce of C37 is $150 which may seem like a lot. But i'd guess you could do 50-100 FRs with that. Still not as cheap as puzzlekoat (where it would cost about $6 for sufficient to do the same number of cones). Relative to the cost of the driver, the treatment costs nothing. With the PK on the FR125 it is far to easy to slop some over the edge of the VC and glue it to the phase plug.
What is the sonic difference? Tonal balance seems pretty much unchanged, but the noise floor drops dramatically. This reveals all sorts of detail that wasn't there before.
Once i've done a bit more work with it i will likely offer it up as pre-prepped kits for people to treat their FRs (or WRs). Most of the cost will be labour for repackaging.
dave |
|
|
| Speedsmile |
| quote: | Originally posted by planet10
That is a very interesting analytical result. With the energy available to excite a resonance decreasing at approx the square of the frequency, that gives plywood a 4-fold advantage over the MDF.
I'll let Chris comment of the difference between finishing ply & MDF...
dave |
Very interesting indeed, however the real question is how much absolute energy is available to excite resonance at 500 Hz in the first place...
Nevertheless this result causes me to consider plywood as a very attractive alternative. Actually when I look at the mechanical material properties plywood has a superior strength to weight ratio.
| quote: | Originally posted by chrisb
here we go again ...........
my 2 cents worth:
depending on the grade of finish you're looking for, MDF can be as easy as a quick couple of coats of texture spray paint and clear top coat polyurethane, or a whole 'nother world of pain for high gloss "piano" grade finish.
As others have noted extensively in another thread, conditioning, sealing, priming and prepping MDF for a high quality, durable paint job can easily take several times longer than the enclosure fabrication
.....blabla.....
Finishing then becomes a question of stain or not?, and then several spray coats of post catalyzed nitro cellulose lacquer. As Ed Schilling would say: "After over 10 years of doing this, I've found something that works for me - so 'nuff said" |
Thank you for the info, based on my experience I agree with you that finishing MDF in a very nice way will indeed be very time consuming, but to me this is'nt a big problem.
The thing is; I have a finish in mind using a part veneered look, and part black paint, something like these:
http://img183.imageshack.us/my.php?image=e21jm9.jpg
(Or the other way around, black front and veneered back)
I plan on using pre-veneered mdf or plywood for the veneered part.
Furthermore, could you elaborate a bit regarding milling the baffle using plywood?, because that is my main concern, MDF will be very easy and smooth, using plywood I have no idea....
Oh yeah, I take it for granted that shrinking etc. using plywood is just as unimportant as with MDF? |
|
|
| Peter Menting |
Thanks Dave. Hope I didn't interfere with this thread. I look forward to your progress as I have eight well worked in WR's and FR's to play with.
Pete |
|
|
| Speedsmile |
Update!:
In the past days I was busy making a test cabinet, but had some delays due the fact that we needed a new circular saw.
Finished the cabinet yesterday, and today I took them for a spin, on purpose I first tested the fully closed version of 4.5 litre, without any internal stuffing. Off-course this one had a very heavy 100-150 Hz region, and sounded very hollow, but this was quickly solved by adding some stuffing.
Next I made the 7, 13 and finally 20 aperiodic holes , and used the same damping material to create the resistance material (pritex).
Adding holes did decrease the 'boominess' a bit, but I'm not entirely convinced. Measurements (uncalibrated and cheap mic) showed there is a indeed still a bump in the bass region, and a dip from 500-1000 Hz, this is exactly what I heard, the speaker sounds to muffled for my liking.
Playing dire strait's sultans of swing for example exactly 'attacked' this behavior and the bass line is far top pronounced, while the voices are too muffled. Below some of the measurements, beware that this is measured with not too great accuracy, but they show some of the trends I'm hearing.
Any suggestions as to what I can do next?, I have tried pritex, and real sheep wool (we had sheep 10 years ago :D) alone and combined. At the moment I have some piece of pritex pressed against the aperiodic holes. I also have acces to some cork but I have still to try it.
Maybe I make my final version a bit wider and less deep, any comments on how this affects the response? |
|
|
| Speedsmile |
| I just stumbled across the hobbyhifi measurements and the peak between 1000-2000 Hz is clearly a driver property, their measurements do also show the other global properties, so actually there is not much going wrong here. But this doesn't solve the problem although after some more listening, I think that if I can reduce the bass peak I'm rather happy. |
|
|
| chrisb |
Well, the 4.5litre box is definitely a compromise in terms of trying to squeeze maximum performance from this driver.
FWIW, having built a few of them, I think the best sounding Planet10 enclosure for either the FR125 (or the Fostex FE127E), in terms of extension and smooth control of bass roll-off in the smallest cabinet, is the mini-onken / Fonken. It is however, several times the net cubic volume of this little guy.
Measurements are fun, but for a quick test to verify if what you're hearing is the box or the driver, I'd suggest quickly mounting them on a small open baffle ( 4ft^2 or so - even stiff cardboard would suffice for low volume levels). All "full-range" speakers have some degree of compromise, and it could be that some of what you're missing from the vocal projection is inherent in the driver. |
|
|
| Speedsmile |
| Yes I can understand that, and I do not expect a perfect speaker, maybe I'm just to accustomed to my other speakers. I've to say that these FR125's do immediately reveal if a recording is clipped even the slightest bit. I first thought foobar's replay gain plugin messed my music collection up :bawling: :bawling: , but comparing the files with the ones on my 20 gb mp3 (wich I've not updated for several months) revealed I was mistaken. :rolleyes: (I'm not totally sure however) |
|
|
| Speedsmile |
Well it is time for an update :smash: :smash:
Last weeks I was busy buying materials and getting my hand dirty....
so here we go...
First of all, I choose 10 mm birch plywood for the cabinet, except the baffle which is 12 mm.
Then I choose to veneer the baffle with watery (don't know if this is the correct english term) maple. The maple sheets were just a little wider then my baffle so that was very handy.
I'm compressor spray painting the remainder of the cabinet satin black.
Some pictures:
The test cabinet:
The veneered baffle:
The 'watery' effect:
The cabinets in my paint 'room' ;)
 
Monitoring temperature and humidity to prevent paint surprises :apathic: :clown:
TEASER; laying out the parts for the accompanying amplifier ;:smash:
To do list:
- Finishing the paint job
- Sanding an applying danish oil to the veneer ,any tips to what the finest grade sandpaper is I should use?.
- Testing these cabinets and determining final stuffing
- Manufacturing the case of my amplifier project :smash:
- Finish and listen :) |
|
|
| Bob Brines |
| quote: | Originally posted by Speedsmile
- Sanding an applying danish oil to the veneer ,any tips to what the finest grade sandpaper is I should use?. |
When doing an oil finish, I quit at 220 grit. Between coats, I rub with 0000 steel wool.
Bob |
|
|
| Speedsmile |
| Ha ha, just too late, but thanks anyway, I have sanded them now, ended at 400 grit very nice and smooth now. |
|
|
| Bob Brines |
| quote: | Originally posted by Speedsmile
Ha ha, just too late, but thanks anyway, I have sanded them now, ended at 400 grit very nice and smooth now. |
Great! The only problem of going that high is that if staining a hard, non-porous wood, the stain may not penetrate very well.
Bob |
|
|
| Speedsmile |
Yes I expected that, but I don't think it will be a problem in this case, however I suspect that if you are a real diehard you could let maple look like a mirror....
Anyway because it is too rainy (humidity too high) to paint I couldn't resist and took one of the cabinets out of my paint room, lay the baffle on top of it, including the driver, and I'm already very pleased with the look:
 |
|
|
| GM |
| quote: | Originally posted by Speedsmile
...........I'm already very pleased with the look:.......... |
DUH! :cheers: |
|
|
| Cal Weldon |
| Very nice indeed. |
|
|
|
