Polymate3D FD61 - 3D Printed Full Range Speaker Driver- In development

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
The elephant in the room question is :
Why? Even a capable printer setup that can do this credibly is erm.. Expensive and I suspect, beyond most basement fiddler's budgets.
$$$ that could otherwise allow one to purchase the Best drivers out there.
Beyond that... it seems an interesting adventure.
 
Thank you for the kind word ghemml. I have seen your work and it is truly a work of art!

The elephant in the room question is :
Why? Even a capable printer setup that can do this credibly is erm.. Expensive and I suspect, beyond most basement fiddler's budgets.
$$$ that could otherwise allow one to purchase the Best drivers out there.
Beyond that... it seems an interesting adventure.

It is a very valid set of points and it certainly is not aimed at people who have experience and know what they want to do. My initial drive to do such a thing is to simply see if it's possible, but this project its more from my own personal background.

Through my time building speakers, I have freedom over the cabinet and crossover etc, but the only option to delve deeper into the understanding of how a driver works was to destroy drivers. The only way to learn what I liked was to build multiple cabinets suited to the driver of choice each time. This was both time consuming and takes up space. Each driver requiring a different baffle cut out and screw holes...

So this has been built with a kit design approach for a few reasons:

1) Couldn't locate another company doing the same. I especially couldn't find something that would also be usable in the end as well (Unlike most science project kits). I didn't want something that served the principle, but a real world case.

2) With drivers like the Alpair 7M and 7P, I have wanted there to be a platform where the material is changed and very little else. This would help me understand the differences in material choices, along with cone profiles. With the drive train being interchangeable on FD61, I can attempt many different materials and hear in person the differences.

3) If I was going to design a driver, I didn't like the idea of the magnets being glued down and then cannot be salvaged. This is a non issue with FD61 as well. They can be removed and used on the next project, or possibly added to a new stronger magnet motor, increasing the flexibility of the driver.

I therefore made the design small so that you can print cabinets if you have a printer, but also build with a relatively small amount of material.

So if you can picture a small cabinet with the option of making it a sealed, vented, passive radiator and aperiodic enclosure, and then fit it with a FD61 which you may end up having a Wood PETG, Diamond dust PLA and Carbon Fibre reinforced Polycarbonate drivetrains. You could then change out the material and the enclosure type, allowing very quick learning and adaptation.

Then exploring it further, you can have new suspension options made available with varying compliance, effectively with enough parts creating a driver you can tailor to your own taste.

It's intention is a kit for learning, and the added bonus of it's flexibility. It's not planned to be released as STL's for people to print due to the nature of each printers quality and results vary. The intention is I make all of printed components and supply the magnets etc required in a kit to make at home. My designs are printed on my own design 3D Printer which I have 2 off, and had experience with my first design and ironed out those issues. 1 of the current printers has printed over 60kg of material and only 1 single component failure.

Sorry for the long winded message, and I appreciate the questions and concerns Bare. I'm just a mad man trying to push what is possible:D
 
Last edited:
Paul,

Very interesting work - thanks for sharing. The idea of having a design with interchangeable parts is a good one IMHO - such a driver would be a tweaker and experimenter's dream. But having said that, it is also true that a product like that might not appeal to everybody, especially if we consider those who don't want to experiment a lot. :) If I am not wrong, Mark Fenlon discussed a basket with interchangeable components at one point of time, but cost, health issues, and lack of clear market demand got in the way of further exploration.

https://www.diyaudio.com/forums/full-range/258370-term-ownership-alpair-drivers.html#post3975187
 
Last edited:
Thank you for the kind word ghemml. I have seen your work and it is truly a work of art!



It is a very valid set of points and it certainly is not aimed at people who have experience and know what they want to do. My initial drive to do such a thing is to simply see if it's possible, but this project its more from my own personal background.

Through my time building speakers, I have freedom over the cabinet and crossover etc, but the only option to delve deeper into the understanding of how a driver works was to destroy drivers. The only way to learn what I liked was to build multiple cabinets suited to the driver of choice each time. This was both time consuming and takes up space. Each driver requiring a different baffle cut out and screw holes...

So this has been built with a kit design approach for a few reasons:

1) Couldn't locate another company doing the same. I especially couldn't find something that would also be usable in the end as well (Unlike most science project kits). I didn't want something that served the principle, but a real world case.

2) With drivers like the Alpair 7M and 7P, I have wanted there to be a platform where the material is changed and very little else. This would help me understand the differences in material choices, along with cone profiles. With the drive train being interchangeable on FD61, I can attempt many different materials and hear in person the differences.

3) If I was going to design a driver, I didn't like the idea of the magnets being glued down and then cannot be salvaged. This is a non issue with FD61 as well. They can be removed and used on the next project, or possibly added to a new stronger magnet motor, increasing the flexibility of the driver.

I therefore made the design small so that you can print cabinets if you have a printer, but also build with a relatively small amount of material.

So if you can picture a small cabinet with the option of making it a sealed, vented, passive radiator and aperiodic enclosure, and then fit it with a FD61 which you may end up having a Wood PETG, Diamond dust PLA and Carbon Fibre reinforced Polycarbonate drivetrains. You could then change out the material and the enclosure type, allowing very quick learning and adaptation.

Then exploring it further, you can have new suspension options made available with varying compliance, effectively with enough parts creating a driver you can tailor to your own taste.

It's intention is a kit for learning, and the added bonus of it's flexibility. It's not planned to be released as STL's for people to print due to the nature of each printers quality and results vary. The intention is I make all of printed components and supply the magnets etc required in a kit to make at home. My designs are printed on my own design 3D Printer which I have 2 off, and had experience with my first design and ironed out those issues. 1 of the current printers has printed over 60kg of material and only 1 single component failure.

Sorry for the long winded message, and I appreciate the questions and concerns Bare. I'm just a mad man trying to push what is possible:D


:D I get weak knees when I see slick speaker frame.

Just curious, what type magnet you planning to use?
 
Paul,

Very interesting work - thanks for sharing. The idea of having a design with interchangeable parts is a good one IMHO - such a driver would be a tweaker and experimenter's dream. But having said that, it is also true that a product like that might not appeal to everybody, especially if we consider those who don't want to experiment a lot. :) If I am not wrong, Mark Fenlon discussed a basket with interchangeable components at one point of time, but cost, health issues, and lack of clear market demand got in the way of further exploration.

https://www.diyaudio.com/forums/full-range/258370-term-ownership-alpair-drivers.html#post3975187

Thanks for your post zman01

With the intention of making the driver parts myself, I almost hope it isn't for everyone as I wouldn't be able to keep up :p. The companies out there like Mark Audio, Fostex, Scan Speak, etc clearly know there stuff and I wouldn't want to take someone away from these drivers if they match what they are looking for.

Thanks for the link. I believe Mark Fenlon does make some drivers which you can take apart and reassemble for military purposes. Unfortunately these are not available to us or in different materials due to it's use case.

I will most certainly will keep sharing updates as it progresses closer to a final build. I intend on making a pair of drivers again (Like I did in earlier builds) to make sure consistancy is still going as planned. Then after that it will be attempting to reduce inductance and improve high frequency response.

Paul
 
:D I get weak knees when I see slick speaker frame.

Just curious, what type magnet you planning to use?

I am glad you like it ghemml. It is much better than the early builds. May still have some changes.

At the moment the magnets in use are N35 disc magnets. They then have there north poles compressed against one another to create a strong magnetic gap in the center height wise. The design is all subject to change though.

This is all contained in a tube with a screw on back cap with a indent going inside. This is so the tube of closed over during the screwing down and the compression of the magnets. Don't really fancy all the bruises of magnets coming together on my fingers!

Due to the tube design, it is possible to replace the motor out for another design and layout quite easily. The voice coil is a underhung design and current Xmax is simulated to be approximately 1.5mm, making it's real world power handling around 2.5 to 5W, which helps with temperature concerns as well with certain materials. Tesla for the current motor is approximately 0.67T.
 
Neo magnets are probably easiest accessible magnetic material. It be great to have close up photo of the magnet structures.


It is nice to have something out of the box like what you did.

Due to the printed tube stopping you actually seeing the magnets, I can't show that at this point in time. I will attempt an 270 degree print however this weekend and see if that works as that would allow you to see them.

The design despite using some steel pieces originates from attempting to get as much power from using just magnets on there own, and maintaining a reasonable size magnetic gap to allow ease of making it at home.

On a side note, has anyone found a supplier of 1KG of less of copper clad aluminium wire?

Paul
 
Due to the printed tube stopping you actually seeing the magnets, I can't show that at this point in time. I will attempt an 270 degree print however this weekend and see if that works as that would allow you to see them.

The design despite using some steel pieces originates from attempting to get as much power from using just magnets on there own, and maintaining a reasonable size magnetic gap to allow ease of making it at home.

On a side note, has anyone found a supplier of 1KG of less of copper clad aluminium wire?

Paul


Awesome, do update when u done with the 270 degree print. I also keen in getting a 3D printer.


For enamel wire you can buy from them at 500g


wires.co.uk : Solderable Enamelled Copper Winding Wire
 
Hello everyone. Weekend is here!

This is an update in relation to the magnet motor in response to what Ghemml was looking for. Here are 2 pictures. One before assembled showing the 2 opposing neodymium magnet arrays along with some steel pieces, and the other with it assembled but showing them compressed together.

68295354_463536554201864_5050365491185451008_n.jpg


67890535_463536620868524_4995215405275414528_n.jpg


The array shown in the picture is a bit different to the best so far, but it uses varying thicknesses of magnets to allow me to alter the design and analyse quickly. You can see on the open one that the back cap goes into the tube. This is for 2 reasons:

1) It alleviates the pressure when compressing the magnets, along with enclosing the magnets before the pressure gets too high

2) It opens up the potential to flip the cap around and extend the magnet array without having to adjust the design and print another iteration. This has the added effect of longer test times with each one of these components

Next stages are being worked on. I am altering the front of the basket slightly internally to allow longer throw without chance of contact. The suspension components have had there design changed to create a stronger bond at the tips of the rib sections. Previous suspensions had tears in them and although the current ones have gone fine, I am improving this more now.

Finally, these new suspensions will be created in a 83A shore hardness flexible material.

This is a very highly flexible filament, which in most cases would cause problems on standard extruders, requiring something like a flexion extruder to achieve the results. During the development of this driver, this was a key issue, and my own design extruder was not cutting it. This then indirectly created my new extruder design which was added to my own design printers. Hopefully we will see a higher compliance in the next model.

The final stage will be adding some ridges back into the moving mass to improve rigidity.

As always, any questions, queries or concerns are very much welcome. Will see if I get enough time in the rest of this weekend to bring all the new components together.

Paul
 
Hello Ghemml

I have been measuring in Tesla throughout my testing and the average over the area where the copper wire is placed is 0.67T. I believe this translates to 6,700 Gauss which is weaker than most magnet motors for drivers this size (28.3cm2 / 60mm diameter cone), but not miles off.

The most common polymers used are rather rigid. Especially PLA which is the most common material printed due to ease of printing, but it has a softening temperature of around 55 to 60c. Now for my design using 5W or less, this is okay, but I am working with PETG with a softening temperature of around 70 to 75c.

There are many options though and there properties can be quite varied. The stiffness is determined by its young's modulus value. Here are some examples of materials and properties I look at:

PLA:
1.24g/cm3
55<60c Temperature
3300 < 3800 Young's Modulus

PETG:
1.27g/cm3
70<75c Temperature
2200 < 2600 Young's Modulus

Polywood:
0.85g/cm3
55<60c Temperature
2600 Young's Modulus

PC Max
1.20g/cm3
108<113c Temperature
2000 Young's Modulus

Carbon Fibre PETG
1.19g/cm3
80<85c Temperature
4600 Youngs Modulus

So you can hopefully see from some of the above as to why I use Polywood and Carbon Fibre for some of my tests. Polywood has a lighter mass so I can print thicker or add shape to improve stiffness whilst keeping weight down.

With the Carbon Fibre the same print will be slightly lighter and significantly more rigid. There are materials that push these parameters further, but this gives you an idea.

On the other side of the materials, you have the flexible ones of which I use a type of TPE

TPE
1.20g/cm3
Temperature N/A (Will melt around 160c)
186 Young Modulus

This is the kind of material which is allowing me to create spiders and surrounds which have a compliance within the region I am looking for. Do look up all the materials you want to print before getting a printer, as most cannot do such a varied set of materials.

Paul
 
Hello Ghemml

I have used it before as a sealed between the driver and the cabinet. I don't know what parameter exactly you are looking for or if they have a measurement of it, but I will PM you with a link to the datasheet. Hopefully it will have what you are looking for.

Question time from me to anyone reading this. What are you looking for when it comes to a small driver like this?

Colour wise what would you prefer?

The classics like...

Black
White
Silver
Copper
Wood like

Or would you prefer something different...

White with Blue phase plug
Maroon with Copper phase plug
Black with White phase plug

Would the option to swap out the phase plug as well to different styles and colours appeal?

Very interested in what people would like to see. Off course the sound and material choice comes first, but I can see what I can do.

Paul
 
They look really cool and is a nice idea. I was originally thinking to make the phase plug fit on via a screw on method but with that idea in mind, maybe adopting Planet10's method of a magnetically conductive screw is a better method.

I have created a identical driver to my most recent to run these in a stereo system whilst I continue development. These are in my Fastidious vented cabinet.

Fastidious is a second attempt at a small bookshelf speaker cabinet. My first used a 0.4mm nozzle, 800g of material, and took 29 hours to print it's 3.3L, 6mm thick with 20% infill cabinet. This second model is a revision of the original, 10mm thick, 960g of material, and only takes 12 hours this time. Cabinet designs are open to change but it's a good baseline.

FD61 paired with this 3.3L vented cabinet produces a flatish response down to 55Hz before starting to roll off. I also have printed TPE seals for the port holes to allow testing as a sealed cabinet as well. The little red cabinet in the youtube video is a 0.6L sealed box I use for SPL and power handling testing. Is also handy to determine if it will work well for a bluetooth speaker.

Paul
 
Hello everyone. Quick update.

P28 is now complete in a stereo pair. These are the first to exceed 80dB @ 1W on a 60mm driver. Next stage is to reduce inductance and improve cone for high frequency response. Just some pictures of the pair for you guys and gals.

An externally hosted image should be here but it was not working when we last tested it.


An externally hosted image should be here but it was not working when we last tested it.


An externally hosted image should be here but it was not working when we last tested it.


P29 will likely not be done this weekend, but the one after due to waiting on some parts. Till next time!

Paul
 
Came across this thread and felt I had to make this my first reply on this forum: Incredible! I've been working in 3D printing for about 7 years now and I'm extremely impressed with the quality of what you've got. I had considered trying to make my own passive radiator with printing but never a driver, really kudos!

Have you tried using wood-filled material for your enclosures? I've made a few speakers with it, and compared with standard PLA or ABS I have found it far preferable.
 
Hello Singaud

Thanks for the kind words. My 3D printers I use are my own builds. One started as a Prusa i3 clone and the other is a straight from scratch build. The Prusa i3 clone has had every printed component redesigned and replaced, and during the 2.5 years I have spent on this specific project, certain parts have been redesigned again. I have my own torture tests, weigh prints to 0.01g accuracy, and digital calipers to measure filament diameter to confirm or deny if printer has been moving in the right direction.

Too your question, I have steered away from the composites for the enclosure partly due to cost, and partly due to them usually having a lower young's modulus (Wood in particular). This would then require more material. On the other side of the argument though, The wood based and other composites when used as the speaker cone to my ear sound more pleasing...so there may be more to it. We ultimately want the enclosure to not flex and emit its down discoloration, but if it is in some form, it's possible the wood composite emits a sound signature more pleasing to the ear.

For now with enclosures, I will probably make a few minor changes, but my main focus is the driver. If you do some designs for other speaker manufacturers, why not make your own thread and show what you have been up to?

Whilst I am making a post, I am waiting on some parts still and the pair of speakers are playing nicely. I have been working on my cone profiles to help extend high frequency response. See the render below on how it is looking. I like the phase plug in the render too but both the cone and phase plug here are subject to change

An externally hosted image should be here but it was not working when we last tested it.


Hope you have enjoyed the small update.

Paul
 
Hello everyone

Minor update. I have run the aluminium tube just under the voice coil yielded a very minor change in inductance. New cone profiles done in the desired materials and thinner has come out well. Next stage will take some work but worked well before on a test and that is to encase the magnets in a copper pipe rather than a printed one. I decided against this before for a few reasons, but considering the advantages, I will be spending the next week or so working on implementing this in a clean and reliable manor.

The bookshelf speakers as shown before continue to be used and I have uploaded a short Youtube video of them running a classical piece of music. Gives a much better view of what they are capable of for sound clarity then previous videos.

YouTube

As always, any questions and queries are welcome. To the viewers but not posting, what do you think so far? What would you change? What would you like to see?

Paul
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.