- This topic has 13 replies, 3 voices, and was last updated 2 months ago by
.
-
Posts
-
-
I’m in the process of making my next DIY speaker, I’ll put a thread on here about the build once I’ve got past this little issue I keep having. The issue is with the signal power on winisd, whenever I upload a new driver following nicks tutorial (so I’m pretty sure I’m doing it right) I go into the signal power tab and input the drivers RMS number, which in this case is 30 watts, then check the cone excursion before I do anything else. The issue is that no matter which driver I use, once I put in the RMS number the green line goes above the red line, a lot below the drivers simulated f3 and by a little bit after it’s f3. I really don’t understand why, all my other builds I have just used 1watt and can literally have the speakers incredibly loud without noticing any problems. Is it really that important to put the signal power higher? I ask becuase not only is it showing an issue in the cone excursion but also makes it impossible to fit a port in the box with 30 watts signal power, where as at 1 watt I can and have before, without getting chuffing or unwanted port noise even at high volumes. It’s a little bit disheartening when you believe you had understood something and can use it properly to then realise something your doing isn’t right, but I guess it’s the way learning goes sometimes.
The driver I’m attempting to model is the ND91-8
-
-
-
-
Hey there 👋.
If you go over to the SPL tab at the top where you move between the different graph types, you’ll be able to see the difference in volume between the different wattages that you enter in the system input power box. 30W (RMS) is what your driver is rated for when under continuous operation (think max volume white noise). If you are happy with the approximate SPL that 1W gets you at 1m distance away on that graph then you can definitely use that figure as a design reference in winISD.
To find out the limits of your current design you can look at these graphs as you increase the wattage to get an estimate of how many watts your system can handle before noticeable distortion.
Just know that turning the volume up beyond this value will then start to give you distortion from either port noise or over-excursion. If you ever hear noticeable distortion (that isn’t port noise), turn it back down a bit to avoid bottoming out your driver and damaging it!
If there’s anything else you want clarification on, please do not hesitate to ask. 👍
Elliott Dyson – Elliott Designs (YouTube) – 3rd year MENG Student-
Thank you for that, I have made quite a few speakers now, and I’ve always just used 1 watt to model the enclosure and port, it’s never caused me any issues even playing them at high volumes, but my concern came from when I did try to up the signal power out of curiosity and the air velocity makes it look as though my speaker would have bad chuffing from the port. But in reality the ports have worked fine, even the measurements of finished speakers taken with the omnimic haven’t shown anything unusual (I don’t think). Basically I’m just unsure now how safe it is to model with 1 watt on a speaker that is likely to cost a lot of money to make. I’m just confused as to why the speakers sound good in reality but winisd is showing that at full volume the port isn’t optimal. Like I said in the above post, the port size with 30 watts signal power is impossible to fit in the box, I could just keep doing what I’ve done before but I really like to know I’m doing something to the best of my ability, so I thought it best to make this post and find out from the experts if what I’m doing is ok.
-
Sorry, I re read your comment as it didn’t sink in the first time, I know what you mean now about raising the wattage to see where the speaker will top out. In my current model I can raise the wattage to 20 and the cone excursion stays just under the red line, but the rear port air velocity is off the scale still, so lets say I wanted to model the speaker to handle the full 30 watts and fit a port in the box, to do that would I have to increase the box volume? Is that where I’m going wrong trying to increase the power in the same air space that works ok for 1 watt?
My plan for this speaker design is to make 5 of them and replace my larger surround speakers with smaller ones to fit a smaller room, so I want to keep the box volume small as possible (but I still struggle with passive radiator designs). My av reciever gives 50 watts per channel so I was thinking at half volume this design would work ok as it never gets turned up to full volume anyway.
-
Ok, a few things here.
I’ll start of with your amplifier.
Usually those 50W per channel specs are lower when all speakers are playing at the same time. Secondly, when playing at half volume it should actually be a lot less than half wattage since dB does not scale the same way as wattage. If you can give me the -dB scale the AVR gives you I should be able to estimate the power to each channel for you at that volume.
So, the other point is port velocity, I’d recommend having a read through this: http://www.eminencedesigner.com/EDTopic14.html
It mentions a given velocity. That should be an upper maximum really, if you want to minimise distortion, aim for 17m/s or below.
To do this it’s not really about enclosure size. A larger enclosure typically makes the port system more efficient. What you’ll want to do to is increase the port area (diameter if circular). This also results in a longer port for a given tuning frequency. Annoyingly winISD doesn’t take into account the volume a port and driver takes up when considering the enclosure volume, so be sure to add that on afterwards when building your enclosure, else your results may not match winISD as closely.
If you don’t want to make your enclosure bigger that’s fine. Just know that there’s always going to be some limitations. If you limit enclosure size then either power handling or distortion/port noise is going to suffer. You have to weigh things up and decide what’s best for you.
Another thing is you should definitely make a prototype before making all the enclosures. You don’t want to make them all and then realise something was wrong in the simulation or that actually you need the port area and enclosure to be bigger.
Again, if there’s anything else, just reply and I’ll be more than happy to help 🙂.
I (and I’m sure many others) would love to see the progress of the build if you want to post about it on this forum as you do it? 😃
Elliott Dyson – Elliott Designs (YouTube) – 3rd year MENG Student -
-
-
I think maybe I’ve just been expecting that becuase a driver is rated at 30 watts that it would automatically mean it would be possible to model it as that without too much cone excursion and be able fit a port in there with no problems, but the port would just be way to long to do that with this particular driver and it exceeds the cone excursion line at 30 watts regardless. Other than that I basically follow all the steps you mentioned above when I start a design, like keeping the port velocity under 17 and adding the ports volume to the box volume once it’s been established, what I mainly found troubling was guessing at how much air the driver would take up in the box, it’s always ended up being a guess, so if you have a method for that id be grateful to hear it. I will be making just one speaker and be doing tests on it before going for all five, once I have finished the design work I will start a dedicated thread on here and hopefully get some constructive criticism to help me come up with the best design, this forum has been a godsend for me since I started the hobby a few years ago.
-
Usually, I just estimate a cylinder from the magnet to the spider and use that. Where the equation is Pi * D^2 * h * 0.25. Where D is the diameter of the magnet and h is the distance from one end of the magnet to the spider. That’s just what I use. I don’t believe box volume is sensitive enough to warrant any closer approximation of the driver volume.
Elliott Dyson – Elliott Designs (YouTube) – 3rd year MENG Student -
Yeah thats a good method, thank you, I’ve mainly just chamfered the inside of the driver hole before to create a bit of extra volume hoping that would be enough, your method is better 😂
You’ve already took a lot if time to help me with this so I understand if you have other things to do, but I wondered if you could help me to design this speaker with a passive radiator rather than a port? I would like to give it a signal power of at least 10 watts but I can’t keep the slot port short enough for the enclosure, but bear in mind I have zero experience with passive radiators so it could be painful for you if you say you can help 🤣
-
@elliottdesigns You are correct, the box is not that sensitive to the percentage error tolerance of assuming the driver to be a cylinder at these frequencies. By the time you get to frequencies high enough that the chamber volume is that critical you are working with drivers that almost always pre sealed in its own enclosure e.g. tweeter, compression driver etc.
Josh Evans, Professional Live Sound Engineer, High End Commercial AV Install Technician
-
-
I’ve managed to find a good balance now using 20 watts, fitting a port into a 1cuft box wasn’t the easiest though so I have two designs to chose from now, the second being an external port that is integrated into the speakers stand, haven’t decided which to use yet though. But thanks for all the help, I new the basics when I made this thread but I feel I understand winisd more now having discussed the different signal powers and how they affect the finished design 👍
Once I’ve got further into the design I’ll post a thread for the build, this one is gonna be an aluminium enclosure, but a small one to start off with then I may attempt the big one I planned on another thread if this goes well, haven’t had the bollocks to do it yet as just the material cost for the enclosure is extortionate!
-
- You must be logged in to reply to this topic.
