Getting started on a new project/some questions...

[Unclemo]

I am about to embark on a new smaller 2 way, I guess bookshelf set of speakers for my Dad's birthday. I have made a few sets of other people's designs (been really happy with them all) and a set of speakers for my nephew that I did myself but those were all based on sims and manufactured FRD/ZMA files using info I could find around here and with the help and suggestions from a few of you fine folks (thanks everyone.)
Well since then I have gotten a mic (umik-1) and have played around a bit with it measuring some speakers around the house trying to get a feel for the REW program. I also just received the DATS system in the mail today so I am going to start playing with that soon.
Anyways, so far I have my drivers that I will be going to try and work with, these are going to be smaller speakers with the SB 12NRX25F-8 for the woofer and the Fountek NEOCD 1.0 for the tweeter. I was/ am a bit concerned/ hoping the SB will be able to reach up high enough to meet the fountek, playing around in xover designers seems like it could be ok. I'm thinking somewhere between 3200-4000 for an xover point, but I will see what my initial measurements of each driver look like first. Well so far I modeled the woofer in WinISD and came up with a 6.5L vented box tuned to 55 hertz. I've tried to check all the graphs to make sure I was on the right track and I don't think anything looks too out of whack, if anyone is bored and wants to double check me/ offer opinions all are welcome, I will not be offended. In the 6.5L box I am getting an F3 of about 49 and about 98-99db before I am hitting xmax. The modeling was done before I had the DATS so I am going to run the woofer with that and compare my numbers, I have knocked up a test box and been trying to break in the woofer, got about 10 hours on it so we will see what I can come up with.
I guess my question is, assuming I can get the DATS up and running, should I bother with the DATS and the Fountek? Is there going to be much to be gained from that? I have never worked with a ribbon before. I also ordered a 10 and a 20 cap for when it comes time to measure the frd. Second question, do I just put one inline with the positive lead of the tweeter for measuring? And I would do this to protect the tweeter from the lower frequencies? Also it said on the website that I got it from that a ribbon is a dead short to the amp by itself so the cap helps with that? Sorry, electrical properties are really not my area of expertise. Is there any other tips I need to know for measuring the tweeter? I would prefer not to blow it or my amp up. Once I get measurements on the raw drivers, do I leave the cap in when I start doing my sim in PCD?
Thanks for reading everyone, I will try and post progress along the way with a ton more questions I'm sure.
Also for what it's worth, I realize that these speakers aren't going to shake the walls with earth shattering base or play at concert levels. I just want my Dad and I to be able to enjoy a beer or two and rock out to our favorite Zappa album.
Thanks again and I appreciate any feedback you guys/gals wish to throw my way. Have a great day! And if anyone is in the Portland Or. area and wants to play around with speakers, let me know.

[Psycoacoustics]

Good Luck! Aaaaaaa, what’s your and your dads, favorite Zappa album?

Later, Mark

[Unclemo]

Thanks. Man he's got so many, we tend to go with either apostrophe or overnight sensation most of the time .

[rpb]

I just posted in another thread that if you can measure, you can do trial and error x-overs without PCD. You need to have parts on hand though. That can be expensive.

[Kornbread]

Although I have yet to work with ribbons, my understanding is similar to yours. The cap is used to protect the ribbon from damage and the amp from being unloaded at low frequencies.

From a competitor's description for the Neocd 1.0

Ribbon tweeters should always be used with a passive crossover! The ribbon is a dead short to your amp.

Do not test without a capacitor. Do not use an electronic crossover without a capacitor (we use 20+ mfd).

Just my uneducated guess and no sim so take it for what it cost you, $0.00. I think they may be made to meet but the slopes will need to be steep.

[anotherTodd]

Thanks. Man he's got so many, we tend to go with either apostrophe or overnight sensation most of the time .

Excellent. Those are two of my favs. as well. I thought it was fabulous when Rykodisc put nearly everything on CD. I recall an interview where the President of Rykodisc(Art Mann?) said that sales of Zappa CDs kept his/their company alive when all he wanted to do was keep Zappa alive - by going digital. They did really nice work on all their productions.

[Unclemo]

ok, so I fired up the DATS to get some measurements on the woofer, now this is my first time using it (seemed pretty straightforward) however I was getting some different measurements than the manufactured specs most notably the Fs. The published Fs is 63Hz and when I measured I got 79Hz. I did have the driver playing a 70 Hz test tone for about 10 hours the other day. So being that I have never measured T/S parameters before, is that fairly normal? Does the driver need more time to loosen up? For the record I used the added mass technique with a few nickles to get the Vas.

As to Zappa, I was on spotify today looking at his discography... dang it would take 2 months to get through all of that. He was a machine, I read a biography about him and from what I got out of it, he was just a perfectionist and somewhat of a pain to work with but hey. I know he's probably not everyone's cup of tea but I think he was one of the most underrated musicians of his time.

[Chris Roemer]

Cone straight up? Be sure not to block any vent out the back of the magnet (rear plate).
If your measured Fs was about 16Hz high(er than spec), was your Qts also higher (like upper 0.50s), and your Vas lower (<0.10cf)?
If so, your box model will probably be nearly the same.

[Unclemo]

Qts was .553, measured Vas was .124 and the spec sheet was 0.14cf. So yeah, I will try measuring again. Thanks Chris.

[Chris Roemer]

Even using THOSE numbers, you can get nearly the same results in that box by raising the tuning about 10Hz.
The nice thing about that is the port gets SHORTer! (Maybe 1-1/2"id x 4"?)
Power handling goes down a bit, but efficiency goes up, so SPL's about the same.

[Unclemo]

That's promising. I'm hoping to bring my laptop to work with me this weekend so I can play around with it some.

[Unclemo]

Ok, well I measured the other woofer with the dats and got the same results, not sure why they are different from the specs but whatever. I ran them through unibox and was able to get back to where I started like Chris said. I've been playing around with my umik- 1 and REW and I think I'm getting the hang of it.
One silly question, since I'm measuring them in the test box, am I correct in assuming that the baffle loss and diffraction will be contained in the measurements?
So really all I would need to do would be to extract min phase and I would be ready to work in pcd or xsim?
Also if anyone could let me know how I determine the z offset that would be sweet. I'm hoping to get some measurements tomorrow and I could try and post them if anyone is interested.
Thanks,
Michael

Well I think I found the info for measuring the time delay. I will have a go at it tomorrow and see if I'm smart enough to put it into action.

[Wolf]

One silly question, since I'm measuring them in the test box, am I correct in assuming that the baffle loss and diffraction will be contained in the measurements?

Yes. It will.

So really all I would need to do would be to extract min phase and I would be ready to work in pcd or xsim?

No, not necessarily... If you are measuring in box, there are 2 methods.
- Measure on one axis (tweeter, usually), and keep the time of flight phase of the measurements and go to PCD/Xsim with no offsets required. Use the files as they are, even though the gated summed with nearfield is not in the picture.
- If you measure on driver individual axes, then you extract min phase at some point. Usually, measure individual axes at far and near, splice/join, then extract. This also means that you'll require offsets in PCD and have to take the acoustic center measurement trifecta to get them. Xsim has a delay function to adjust for Z-axis, but that is the only offset it entertains to my knowledge.

Also if anyone could let me know how I determine the z offset that would be sweet. I'm hoping to get some measurements tomorrow and I could try and post them if anyone is interested.
Thanks,
Michael

Well I think I found the info for measuring the time delay. I will have a go at it tomorrow and see if I'm smart enough to put it into action.

A quick search on this forum will help you find it if you don't. wogg even showed it in his Supernova Minimus build thread.

Later,
Wolf

[jhollander]

- If you measure on driver individual axes, then you extract min phase at some point. Usually, measure individual axes at far and near, splice/join, then extract. This also means that you'll require offsets in PCD and have to take the acoustic center measurement trifecta to get them. Xsim has a delay function to adjust for Z-axis, but that is the only offset it entertains to my knowledge.

You should not be measuring on individual driver axis. No sim program recreates the FRDs on the design axis. All drivers should be measured on the design axis without moving the Mic.

[jhollander]

Try this link for the older JB white paper http://techtalk.parts-express.com/fi...tch?id=1149302