It sounds like actually measuring the drivers might be easier than I though! Dragging this thing outside and putting it up in the air sounds like a lot of work!
The woofer and the mid both have about 8 hours of run time now. They both picked up a lot of volume after about an hour but didn't change much after that. Everyone's pretty tired of hearing sine sweeps. The tweeter's been running for a few hours now too, but I haven't noticed much difference in it. I should be almost ready to take some measurements.
What sort of measurements are needed?
The woofer and the mid both have about 8 hours of run time now. They both picked up a lot of volume after about an hour but didn't change much after that. Everyone's pretty tired of hearing sine sweeps. The tweeter's been running for a few hours now too, but I haven't noticed much difference in it. I should be almost ready to take some measurements.
What sort of measurements are needed?
First read the Jeff Bagby paper I linked to in the previous post.
Box
Then move on to his article on finding acoustic centers.
Box
This tutorial is geared toward PCD but works equally well in XSim. In fact it's easier with XSim because you don't have to input x and y coordinates. Just follow the instructions
and substitute an extra unconnected driver (using 'driver only') for the PCD overlay. It's been a little while since I went thru the process so I'm not sure if I missed anything that isn't in the paper. Just ask if there is anything discombobulating.
Box
Then move on to his article on finding acoustic centers.
Box
This tutorial is geared toward PCD but works equally well in XSim. In fact it's easier with XSim because you don't have to input x and y coordinates. Just follow the instructions
and substitute an extra unconnected driver (using 'driver only') for the PCD overlay. It's been a little while since I went thru the process so I'm not sure if I missed anything that isn't in the paper. Just ask if there is anything discombobulating.
I think I have the measurement thing figured out somewhat and have managed to get ARTA running, but I won't be be able to get any measurements until Monday. The second cabinet has got it's second coat of primer on, time to machine a front baffle for it next.
Here's a XSim file that Douglas Blake put together with what he said is a 1970s style crossover that feeds the mid-range from the middle of the woofer's crossover. This still has the higher frequency crossover points that we originally started with. Interesting results with output fairly flat at almost 90dB/1W.
I've been trying to get Response Modeler working but I can't remember how to get around the "Compile error for hidden module" error on startup. Does anyone know how to fix this?
I just figured out my confusion with your tweeter was my mistake - I missed the "T" in the RST28F-4 and was looking at the RS28F-4 instead. My apologies.
Sorry, can't help with the Excel problem. Maybe a Google search?
Sorry, can't help with the Excel problem. Maybe a Google search?
I've tried a Google search on my issue. I saw people having similar issues 8 years ago and the general consensus was it was due to the change to 64 bit operating systems and Jeff Bagby said he didn't have a fix for it (Parts Express forum). I had the same issue on my work PC but can't remember how I solved it.
Last edited:
I just figured out what I did on my work PC. I switched to FRD Response Blender 2.0. Will this software give the same results?
I believe that error occurs when you're running the 64 bit version of Office. If you are able, try switching to the 32 bit version. You'll have to un-install the 64 bit version first.
Blender essentially does the same functions as Response Modeler, but I don't believe it has the capability to simulate the woofer impedance like Response Modeler. It does allow you to add high and low "tails" before you create the minimum phase versions. The tails should make it slightly more accurate than RM.
Last edited:
Welcome back! I'm in the midst of trying to figure out how to properly measure the response of the drivers.
Yes he's pointed me to some excellent resources. I'm trying to figure out how to get around the room reflections now.
Use a gated sweep ... that is, the sweep sound continues like normal but the microphone only listens for a couple of milliseconds at each test point then shuts off before it picks up reflections and nodes.
Hopefully your software allows for this...
Last edited:
I'm using ARTA which allows for gated measurements. jReave recommended a couple papers from Jeff Bagby on what to actually measure. The port graphs are pretty wonky, I haven't figured out how to select the gating points for it yet.
What volume levels should I be taking my measurements at? I haven't seen any actual recommendations on that anywhere.
What volume levels should I be taking my measurements at? I haven't seen any actual recommendations on that anywhere.
I can't help much with the gating specs. I typically use about 2 ms when measuring only inches from a driver, that allows the cone to get moving but precludes any room effects... but that is in my room... yours might be different.
As for volume levels, most people recommend the standard 1 watt level test. But if you're up close to the driver you're going to have to cut that back considerably to avoid microphone overload. I typically measure at less than 40db up close, about 50 to 60db at 1 metre.
Louder isn't always better. (In fact, it seldom is) I've noticed several times that a stressed driver --one being driven hard-- has a very different behaviour than when running in the lower end of it's comfort zone and that could skew your readings somewhat.
I'd say you probably need to experiment to find the best levels and gating as this does depend on room and drivers.