mic ot not?
I"ve been messing around with Passive Crossover Designer for my 3-way tower design. I'm getting the overall summed frequency response to look somewhat flat.
My question is am I wasting my time by only using the factory spec's? (zma and frd files). I get this will give me a decent "starting point" However, once I build the crossover and physically listen to the speakers, that's it. (Other than fine-tune the L-pads on the midrange or tweeter by ear.) Maybe those so called golden-ear people would be like "Yepppp, I better change that cap to a 4uF instead of that dang 2.8uF."
I guess my point is how close can crossover software get without using a mic to perform measurments? Or should I get a mic? Looks like one would set me back $250 at PE. The speakers I want to build will not be "garage speakers" or a first time build for kicks. My goal is quality high-end listening. (As high-end as one could get with RS drivers that is.)
It really does depend on the QC of the manufacturer. Some drivers are incredibly similar to the original specs others not. Depending on this factor I might not even buy the drivers because if you have two slightly different drivers (in terms of ts parameters), you won't get anywhere near as good imaging as you would with closely matched ones. But in terms of getting a mic. I would, because after you have finished making the speakers you can use REW measurements to find the speakers' sweet spots in the room (place where the speaker creates the least nulls and peaks in the room). Then if your planning on using a PC or NUC in there you can use these measurements and import them into Equaliser APO to equalise your whole system to what you want. Bear in mind not to try to compensate for nulls or peaks as componsating usually makes them even worse! For these reasons alone (and many more) I definitely recommend buying a nice mic like the Umik-1. Or the dayton audio one if you have enough money. (I just bought the umik-1, other than the software (where you are comparing free to ~£200) there isn't much difference in the mics. I think REW is decent enough for most things anyway! You could go a step further and get a nad pre/pro with dirac live (that software can do amazing things with lining up different phases throughout the audio spectrum if you can afford it!)
@nick7676 I would never say you are wasting your time using factory specs. In fact, I have built some speakers using good guesstimates using factory specs to great success. However, there is a lot that needs to go right when doing this. The first, is the factory spec sheets need to close what you actually measure. This is not always the case. Typically Dayton is pretty close, but even they aren't perfect. The Esoteric drivers for example measured completely differently than I was expecting. But using the factory specs can also help by giving you decent idea if the drivers will work well together, before purchasing them. For best results, a measurement microphone is the best idea. If I am not mistaken, later this month the Umik-1 will be on sale. Might be the best time to snag one.
One big reason to get a Mic is to get accurate measurements for baffle step compensation. The factory measurements are infinite baffle and will not represent what the speaker's measurement would be in a typical cabinet like you or I would build. Admitantly, I'm not a PCD guy, maybe it takes baffle width into consideration, if so, I'm betting it would only do so for the most basic of box shapes. The Mic is a must in my opinion if you're planning your own crossovers and want a high quality result. Best of luck on your build.
I'm pretty much convinced on getting a mic. My only issue is the extra cost of a mic and a dats. Thanks for the heads up. I will keep checking the umik-1 to see when it goes on sale. I may look around for a used dats v2 as well.
I did watch 123toid's (sorry i don't know your real name) video on Omnimic vs Umik. Great video btw. I would love to have the Omnimic but I can't justify getting it for it will sit in it's box for years after I finish my project.
Nick, you might reach out here or on the TT Forum to see if there is anyone near you that has a mic and DATs and would be willing to help you out. Not sure where you live but I would be glad to help if we are close enough for you to consider the drive.
I fully understand your hesitation to spend too much on the equipment, especially if it's for a one time build and nothing else to follow for quite some time.
Funny you mention having someone use their mic and dats to do the testing. I thought about the same exact thing. I live in canton ohio near the football hall of fame. I would never ask anyone to drive long-distance just to measure my speakers. But if they lived close I would more than welcome them to help me with the measurements.
I do have some questions as to what steps I would take when I obtain some measuring equipment.
1) Start by using Dats to get the T/S parameters for each driver. (Do I use a generic square box to do this or my actual enclosure?)
2) Enter my new T/S parameters into free box calculation software to determine optimum volume and tuning frequencies.
4) Build test/prototype cabinets.
5) With the drivers installed, do the following testing:
5A) Use a mic (like Umik-1 and REW software) to get the frequency response of each individual driver (with NO crossover installed). This will create the "frd" files I need for each individual driver. ( this will give me a realistic frequency response of each driver from "my baffle" instead of an infinite baffle like the factory specs use.) (Right? Or wrong?)
5B) use Dats again to measure impedance of each driver installed in the test cabinets. This will give me "zma" files for each driver. (Right? Or wrong?)
6) Import my new "zma" and "frd" files into free crossover software like PCD and create a crossover
7) With crossover installed, use mic to measure/check combined frequency response and make tweaks.
With you living where you do, should be easy enough to find a fellow DIYer to help out. Not sure about him coming to you, it might have to be the other way around. That and maybe a 12-pack of his beer of choice, lol.
You've got the basic steps all laid out pretty well. Couple of things to clarify. When you measure the woofer with the Dats the first time, it doesn't need to be in a box. Free air measurements are all you need, if the driver has a rear vent, elevate the speaker enough for it to breath.
After the DATs measurements, no need to build a test enclosure, build your final cabinets. If it's ported, try to design it in a manner that allows you to make adjustments to the length of the port without having to disassemble the cab. Yes, you will most likely have to make a small adjustment to get the tuning right, many things can impact it such as damping, proximity to a wall, etc. DATs is great for confirming the tuning and one of us would be glad to explain how once you get to that point.
Everything else in your steps looks spot on, all you need now are the drivers and a local buddy to help out on the measurements. Best of luck man.
Yeah I would gladly go to where they live, too. And I would bring their poison of choice lol.
Thanks for clarifying some of the steps. I found some good youtube videos on measuring the free air of the drivers like you mentioned. Now I fully understand how to use dats. I may pick up the Umik-1 if it goes on sale. Thanks again for the help.
DATS is quite an expensive package, I use REW for speaker measurements as it is free. I thought I might notify you of that. 🙂
Of course if you want to run your own TS parameters then you will need DATS, but usually these are very close to the specs given anyway, you can get most of the way there with frequency measurements of the drivers