Design your own speaker from scratch discussion thread

Great thread, thanks for starting it! I am one of the novices this thread is for.

I'm hoping you'll include a section on phase, and what the design goal of a crossover is with regards to phase.

I also wanted to mention that I very quickly decided to go with an active setup. I do not have a drawer full of capacitors and coils, and the ability to change crossover points on the fly, as well as having access to live equalization is just a great help. It allows for a comparative evaluation of a single set of speakers.

Of course the cost of caps and coils is now replaced by multiple amps...
 
Just another Moderator
Joined 2003
Paid Member
Well it's a work in progress. Idea is that those placeholders can get filled (or deleted). It's not easy to insert something later.

6L6 has done it with his build threads. It'a bit of an experiment. I'm quite happy for people to add more content to the thread. Just not general discussion. :)

Tony.
 
Over all, I think it is too much focus on box simulations, I have come to the belief that it is better to do simple/crude/basic simulations and then focus on measurements and box adjustments. There is just to much that is going on in a loudspeaker box and driver then what is possible to simulate
 
Just another Moderator
Joined 2003
Paid Member
Hi celef, there is certainly no substitute for building and measuring, you won't get an argument from me on that. A simulation should give a solid starting point though. I should add a link to Troels page on how what is calculated often doesn't end up being reality vent tuning .

I put a lot of graphs in there to show that there are many aspects to consider (on a ported box).

I did say this at the begining of the box simulations too ;)

As mentioned I use Unibox for doing simulations so the following graphs are produced with unibox. You can of course do manual calculations, lookup tables or whatever other method you want to arrive at your box volume, but I like to simulate.


Tony.
 
Just another Moderator
Joined 2003
Paid Member
I've added the following to the box modeling post

As celef has pointed out in the discussion thread, a simulation is only an approximation and there is no substitute for actually building and measuring a box. A simulation is a good starting point and if you have the time and resources available, building a prototype before committing to your final build is always a good idea! This page from Troels is worth a read if you are planning to do a bass reflex enclosure vent tuning


Thanks for the feedback! :)
 
Hi Tony,
Excellent work! This should be required read for anyone starting off in this hobby.:) I wish this was in place a few years ago when I started out. :D The box simulation part I found very informative. I use Unibox myself and yet I still learned some things from your illustrations. Could you expound some on the comment you made about avoiding port resonances in the passband if possible?

Thanks so very much!
Rich
 
Just another Moderator
Joined 2003
Paid Member
Hi Rich, basically what I mean by that comment is that if you intend to run the woofer say from 20 Hz to 500 Hz then in an ideal world you should aim to have the port resonance higher than 500Hz, not say at 200Hz. The port resonance frequency shows on the left panel in Unibox.

In the example I gave the port resonance definitely would be within the pass band being at 642Hz (though it is minor) see below: So I should update that comment to make things clearer, as for most two ways the port resonance will be within the passband.

729678d1437374089-designing-speaker-scratch-images-thread-vented_31l_params-png


If you look at the extreme (contrived) example I put in the post
729688d1437374168-designing-speaker-scratch-images-thread-22l_big_port_spl-png


There is a very large aberration in the frequency response at around 240Hz. You would definitely not want to cross this speaker over higher that that, and due to the magnitude of the resonance, you would want to cross over significantly below that point with a steep crossover so as to not have that resonance in the crossover region as well.

It is not always possible to keep the port resonance out of the pass band, but if it is small it shouldn't be an issue. It's just another of the compromises you have to juggle if you decide to go with a ported speaker.

Bigger ports will reduce the chance of wind noise, but at the same time lower the resonant frequency (and usually increase the magnitude of the resonance).

I believe that slot ports(if done correctly) can help in this respect too and depending on the design will behave somewhat differently to a round port. I'm a bit of a neophyte with respect to slot ports. Certainly the box I made with a slot port had a tuning significantly different to what I was expecting based on sims using the area of the port.

Tony.
 
  • Like
Reactions: 1 user
Tony

Looking great and can appreciate the effort you have put into this.

In the crossover section you have shown textbook (nominal impedance) and modelled crossovers and maybe a third option could be shown. That being a textbook xo based on the impedance at the crossover point which can vary dramatically from the nominal due to the rising impedance of the driver and the higher the xo point, the greater the variation. Certainly not as good as a modelled xo but the effect could be shown.

One thing that certainly needs to be added eventually is the padding of the tweeter. Series resistor, amp or driver side, L-Pad all have a different effect on the filter. Any changes in the padding throws the filter design out the window and has be designed all over again. I think it's one of part of crossover design that it thought to be easy and is misunderstood but can cause an xo design to end in tears. Of course software modelling takes away the pain but worthwhile to show the effects of a padding resistor on the network.

Keep up the excellent work.

Peter
 
  • Like
Reactions: 1 user
Just another Moderator
Joined 2003
Paid Member
Thanks Peter, Yes When I was doing the textbook stuff I was thinking I should put in another one with the resistance set to the actual impedance at the crossover frequency (I may have even done the sim) I think the drivers I chose were pretty close to the nominal values so I didn't. But it is definitely worthwhile showing.

I'll tell you a funny story about lpads (and I still haven't fixed this!) When I did the crossover on my MTM's knowing that changes to the lpad would affect the tweeters transfer function I decided to be "smart" and put the lpad before the rest of the crossover components. I didn't think about the fact that meant that it was now in parallel with the woofer and reduced the impedance of an already low impedance on the woofers.

Someone pulled me up on it a few years ago and the lightbulb went off. Luckily the shunt resistor is I think 22 ohms so not causing as much damage as it could have, but I was a little embarrased :)

Thanks for the suggestion it is important and I'll make sure to include some discussion on it as well as some demos of what happens when you change the position or the value!

I've been a bit tired this week and struggling to make any headway on the next installment. I think that I'll have at least four posts in the crossover development part. I'll need someone else to volunteer to do the DSP one though as I've not done any DSP work.

Tony.
 
Just another Moderator
Joined 2003
Paid Member
OK Thanks Peter, but I did most of what has been presented so far quite a while ago (though admittedly it got quite a going over again last week before I was happy to release it into the wild). I'm getting back into gear now after being in neutral for an extended period! :)

Rich, thank you again. I wanted to try and do this in as organised way as possible (part of the reason it has taken me a long time to release it. I hope now that I have I can keep at least some semblance of organization to the material which follows :)

Tony.
 
This is the discussion thread for the So you want to design your own speaker from scratch sticky.

Please use this thread for general discussion rather than the sticky to help keep it uncluttered.
Tony.

Hi, old friend. Superb project in the true spirit of giving back to the community. (I've often wondered whether the designs I build and publish do enough in the giving-back dept. I'd be nowhere in this hobby without diyaudio.) Good to see you active. I've been away for what feels like years. I'm still thoroughly enjoying the performance of my last build, though.

Can i fill up the placeholder post #10? On bracing, cabinet construction, etc? As you may remember from my website, I have some thoughts on that matter, plus a lot of material I can reference as illustration. :D