The dome midrange thread

@motokok I've used that H304 in a few builds back in the 90s. Its actually not nearly up to the level of performance of the 75MX / D7608. The overall distortion is a bit higher and it has some weird resonance at 1.5 k. The chamber is also not correctly designed to get the best out of it. That seems to be the common issue with many mid domes.

I was looking at the Hificompass data and it doesn't at all reflect the performance I've observed from the H304 vs the D7608. Problem is you can't directly compare the two mids due to lack of chamber on the D7608. I can tell you wholeheartedly the D7608 is significantly better, but the data doesn't show that. I came to this conclusion back when I started using large mid domes and quickly abandoned the H304 despite its built in chamber. Perhaps a better chamber could have helped.

There was a rather rare 4" ITT dome they made for B&O used in a prototype 4 way monitor. It used that huge Vifa 15" acoustic suspension woofer. It was more of a novelty than anything practical. I tried looking it up online and couldn't find anything on it. A 4" dome would be interesting if it had a decent suspension design.
 
Last edited:
  • Like
Reactions: piotr z and motokok
Finally, a significant update on the D7608 chamber progress -

I got the CAD files back from the generous fellow on here who graciously volunteered his time to make this happen. Out of respect for his privacy and sanity that he doesn't get swamped with tons of PMs here, I'll let him decide whether he wants to be mentioned or is able to answer any questions regarding printing them.

I'll be posting the STL file once the design is tested with absorption media in place and it fits the D7608 as intended. The chamber design is for personal use only. Any commercial use should be cleared with me first.

Initially I'll be printing a pair of test chambers from PLA to verify the overall fitment and perform some measurements. I took inspiration from the newer Seas T35C002 design, as it requires minimal mechanical dampening to achieve decent decay spectrum and avoid most common standing wave modes.

Here's a teaser -

20250128_105151.jpg
 
Is this PLA as good as any?
The mostly double curved surfaces of this chamber inherently make it very stiff and PLA should be perfectly fine as material.
For best air tightness it may be necessary to print it with a high infill ratio and/or cover the surface with filler, primer or paint.
For high power applications PLA may be too heat sensitive, it's getting soft above ~ 60-70° C. Also consider heating up of an eventual bass enclosure outside the mid chamber.
 
Last edited:
Yes,, thank you @stv for all your hard work!

I was thinking of using CF loaded ABS. Its not that hard to print if you're hot end can do the extra few degrees. I'd imagine PLA would be more than sufficient temp wise for domestic applications. There are quite a few types of PLA with varying durability. I've used the SunLa PLA pro on several prints with very good results. Its very tough material and cleans up very easily using abrasives. I recommend at least a 50% infill and probably would do 70% if you're not in a hurry. If the printer nozzle is extruding correctly I don't think there should be any concern with getting an air tight seal. I usually put a skim coat of cyano on both inner and outer surfaces if there's any concerns with sealing, but I really doubt there's going to be an issue.

To adhere the chamber, a higher grade hot melt adhesive would be more than adequate and leaves the option of gaining entry if something should need attention inside. The inner ID may need a little tweeking depending if your printer under or overextrudes, or if the scalar is set incorrectly. There are of course production tolerances on the dome itself leading to ID tweaks. You'll have to check your own printer for any stack u tolerances.
 
  • Like
Reactions: stv and jccart
I was thinking of using CF loaded ABS. Its not that hard to print if you're hot end can do the extra few degrees.
You might already know but abrasive filaments need a hardened nozzle to print. I like CF PETG for it's dimensional stability, it's also got a bit of damping, but the filament can be variable by manufacturer. QIDI does a glass fibre ABS that is incredibly tough, although quite expensive. That doesn't warp either. PLA prints easily but I always feel like I'm holding a toy not a product.
I recommend at least a 50% infill and probably would do 70% if you're not in a hurry.
Increasing the number of wall loops is the best way to increase strength. It also makes the print look better.
 
@fluid I agree with the abrasive filaments needing a hardened nozzle. The other issue is the heat brake and the teflon tubing, which can't handle being exposed to the higher temps needed for printing alternate filaments. Trying to dial in hot end and bed temps can be difficult, especially when the weather changes.

Found out the hard way my build plate was warping when it heated up, ruining the first layer and jamming up the nozzle. You need tight mechanical extruder control to get consistently clean layers. The double Z axis is a b**** when it goes out of sync.

I use a tungsten nozzle and titanium heat brake, which I align in a way there's no pockets or internal gaps to catch any liquid filament residue. That causes jam ups and lots of issues with most budget printers.

I had to literally put myself through a steep learning curve to figure out why my (not so)cheap(in the end) creality printer was screwing up. Its the simple crap that will get you, yet there's no specific info on the web you can rely on being accurate. In hind sight I should have paid someone to print it but we have zero walk in tech places down here in Hicksville that do this sort of thing for us peasants.
 
  • Like
Reactions: stv and fluid
I happen to have a pair of H304. If there is a problem are there other ways than 3D printing to fix it?
1. Adding bitumen damping to the outside of the original back chamber?
2. Cutting of the original back chamber and make a way larger one (MDF box, cardboard tube,...)?
 
Found out the hard way my build plate was warping when it heated up, ruining the first layer and jamming up the nozzle.
Heat creep is an issue with a lot of thinner aluminium beds. Enclosed printers are better because you can heat soak the bed to get it to a temperature where it becomes stable and then run a bed mesh on in so any minor level issues are taken up in the mesh. Without heat soaking it for 10 minutes or so before printing the mesh can be quite inaccurate.

I also found with abrasive filaments in enclosed printers attention has to be paid to make the filament path as smooth as possible. I had real trouble to start with prints failing due to filament breaking in the feed tube. I had to print a riser and some mods to improve the filament path, since then I haven't had any prints fail for that reason.

Infill pattern matters too as abrasive filaments can build up a bit on the nozzle and then with some infill patterns it can cause the nozzle to come into contact with the print and dislodge it or at best make the quality rough. Unlocking Klipper on the Creality printers that use it opens up a lot of possibilities for customisation that can improve print quality.
 
  • Like
Reactions: profiguy
@fluid The heating bed on my printer is aluminum with glass build plate. I machined a flat 3/8" aluminum plate, which helped but the glass build plate was always warping. It couldn't move independently of the aluminum expanding at a higher rate. The z axis also needing synchronization which wasn't easy, along with zero lash worm screw couplers. By the time you spend all the money trying to fix all the bugs, you could have bought a used professional printer that has its own product support.
 
@fluid I was looking to get one of those editing interfaces to modify the G code on the printer setting side. I hear these can be helpful with tweaks on the printer in general ie. scalar settings, extrusion offsets, etc. First layer height has proven critical to any print, especially with larger surface prints. I had to buy 3 build plates to get only a single good plate out of them.
 
For a little more context; I would like to try and build a speaker with similar characteristics to the Celestion Ditton 66. I am interested in a detailed hifi sound not a studio monitor sound. I want to produce a wide sound stage with great imaging that's natural, engaging and not at all fatiguing. My aim is an exceptional everyday speaker for home listening, likely to be played towards -85dB but also performs well at lower volumes.

Regarding mid's, my starting point and how I got here, I have been looking at the Volt and Bliesma options so far. My instinct tells me to see if it will be possible with a paper dome, however the paper Bliesma doesn't get as good a write up as the other options. Silk would be my next choice I think, historically I haven't enjoyed listening to beryllium/alloy mids.

Can anyone weigh in on the difference in the paper vs silk? Or indeed Volt vs Bliesma? The paper breaks down sooner in the top end but does it make up for it in characteristics?

Are there other high end options to consider?

@profiguy whilst writing this I have come across your recent considerations and they align with me greatly. I feel like I am in the right place. Thank you. If you had any time to shed some light on my thoughts that would be greatly appreciated.
In short, I like to listen to most styles of music in a real world setting, sometimes at higher volume levels. A wide dynamic range, accurate sound stage and very low distortion with excellent midrange transparency are very important to me

Crossing at 500kz and 5000hz seems like a difficult feet, or at least I haven't found a mid with that range yet. Crossing around 600 and 4000 seems more likely. Although unless I am mistaken it still looks like a stretch and a high order slope may be required, which could result in a less natural sound?

The SB34NRXL75-8 seems like a good LF option if going for a dome mid, and can also be paired with a PBR instead of porting which appeals in the Ditton 66 context. The HF which has caught my eye so far is the Scanspeak Classic D2905/9500.00.

Whilst I have good ears (IMO) and lots of practise setting up big hifi systems for dance parties, designing speakers from scratch is a totally new endeavour. I am not afraid to be told I'm barking up the wrong tree completely, I know very little about crossover design for example. Just keen to learn in a practical application.

Any opinions or nudges in different directions are welcome 🙂

Thanks for reading!

JB
 
To understand better why this speaker appeals as much as it does to you, it would be useful to know all its technical details. One can surmise that the soft dome material and likely shallow crossover filter slopes contribute, along with well matched accompanying drivers.

There's some technical documentation on the Ditton 66 in a brochure downloadable at https://www.hifiengine.com/manual_library/celestion/ditton-66.shtml An actual FR curve is shown with illustrations of the drivers, and even the crossover. No crossover slopes shown but guessing from the crossover pic, unlikely to be greater than 2nd order.

This speaker's 500/5000 Hz crossover points would have been chosen on the basis of on-axis FR curves. It cannot have the smooth directivity transitions most diyaudio builders seek. There will be significant FR discontinuities off axis. And to achieve 500-5Khz with good off axis behavior, you would probably have to seek the best of the available mid domes. But since your volume wants are modest, perhaps a 2" dome with its higher frequency reach & wider dispersion could be persuaded down to 500Hz without too much distortion.

This user review -- https://audionostalgia.co.uk/celestion-ditton-66-review/ -- suggests a somewhat fat and appealing 40-80 Hz bass boost effected by the woofer + passive radiator & a less than thoroughly braced enclosure. The comment that clarity suffers with more complex music may be the negative aspect of this bass.

The SB34NRXL75-8 works technically up to 500Hz, but my guess is that it won't quite give you the lively sound you get from the lighter cone & more flexible surround of the Ditton 66 woofer.
 
@Josh_B There are some dome mids able to cover most of 500 - 5000 hz. It depends on several factors, but the practical useful bandwidth you should aim for is more like 600 to 400, as you pointed out.

The paper dome M74P isn't going to get you to your goal as easily as a silk dome. This is likely the diaphragm material you're going to be happy with if the Ditton 66 is to your liking. I'd probably recommend the M74S as the more ideal choice, but the D7608 would be suitable on a budget if you lower the tweeter transition to 3500 hz. Crossing at 600 2nd order from the LF shouldn't be a problem. The SB34NRXL would be borderline for the application. I'd look at a better 10" instead. Your problem with the SB34NRXL is that kink above 800 hz and that's going to pose a problem with anything less than a 4th order 500 - 600 hz LF LP. You have to obviously keep the woofer directivity in mind as well. I wouldn't cross the SB34NRXL higher than 400 hz 2nd order. SB has a kit using that woofer and I believe they cross at 350 hz (2nd order?). Its a solid 92 dB/W woofer and would pair nicely with your typical 6 - 7" cone midbass as a 4way. A dual 8 - 10" LF 3 (or 3.5) way would be optimal used with a 3" mid dome.

Your tweeter choice is excellent, but may be a little too shy on the sensitivity side. If wide sound stage is what your prefer, a narrow baffle would facilitate this better for wider horizontal dispersion. You could do a tapered round over towards the top of.cabinet, narrowing the baffle past the woofer. That's not easy to execute from a diy perspective.