The curse of free gear (what should I do?)

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I went to the NorCal DIY audio meeting on Saturday and won some drivers in a raffle. The description said Cerwin Vega/NHT Tower system.

There are 4 16 ohm 6" Mids (Cerwin Vega PM7) and 2 Tweeters from NHT (they're from NHT Sub Zero bookshelves). Googling the NHTs reveals that they were originally crossed over at 2200 Hz. The Cerwin Vegas don't show up anywhere, the closest matches are for automotive speakers.

I mounted one set in a cardboard box, and the sound was not bad. My ears are low grade steel, rather than gold.

Just for grins, I'm building a couple of Monkey Coffin style speakers (11w x 7d x 35h). I'm making the front baffles completely removable. If I like the overall aethetic with my home theater (think WAF), I'll replace the drivers with something nice. I'm looking for Left/Right mains. We don't play music critically; I just want to be able to enjoy watch TV. My rear surrounds are Mimimus 7s with custom crossovers. The sub was $5 at a rummage sale (BIC, originally $100).

I have 3 questions:

1) The NHT Tweeters are obviously meant to be mounted in a custom-fit plastic enclosure. I can use my router to mount them in the MDF baffle, but it will be a PITA. Is it worth it? Should I spend a little on Tweeters (if so, please recommend)

2) I have some cheapo 3500 hz crossovers from another project. Shall I just use them? There are NHT Sub-Zero Crossovers available on eBay for $35 shipped; would they add any audible quality to the project?

3) Can any one supply me with an simple design for a minimal crossover at 2200 I could build? Google yields no usable specs for the drivers. The 'kit' included measurements for the Cerwin Vegas, but not the NHTs. Is there some way to engineer a simple crossover based on the info I have?

Bear in mind that my expectations are very low. I'm mostly experimenting to see if I can build enclosures and grills that my wife will allow in the living room. It would be nice if they turned out good enough to be permanent, but I'm avoiding the high cost of good drivers while I test my woodworking skills (which are generally high, but my wife requires grills, and they worry me).

Thanks for your input.
 
There are no easy or simple answers to your questions.

The less you care about how it sounds, the easier your project becomes?

What is really needed is to put together a crossover that "works" for the actual drivers that you have. The slope of that xover depends on the acoustic response of the drivers. In otherwords, if you have a peak at say 3kHz, and you have a crossover at 3kHz, guess what? You do not have a crossover at 3kHz acoustically, it's higher - somewhere at the end of the peak...

Then too, you can't crossover the tweeters too low, and similarly you can't crossover the mids too high either - both conditions create problems. So, as is the case for virtually all speakers you want the best "fit" you can manage for your drivers.

The enclosure for the mids counts big time, unless they are sealed back mids.

They need to be loaded into the appropriate volume per their T/S parameters.

This is pretty standard speaker building/design stuff... if it appears daunting you have two or three choices:

- post the exact drivers and all the specs you can find here, plus some jpegs and ask for guidance. Some of what you get will be useful, some not, then decide what to do from there.

- read up here on diyaudio on crossovers and other design "sagas" and see what people have done and are doing, and how people solved similar problems

- read up in a book, like Vance Dickason's Loudspeaker Cookbook and figure out the theory

- and 4th, buy a kit from an outfit like Parts Express, one that is already figured out, OR follow a good design from here on diyaudio or another site. (skip the design process, go direct to the construction step)

There's no other or easy way to solve the problem properly or well... otoh, ANY xover will make the drivers make sound, setting driver levels is still another issue, and the minimum requirement for a decent sound with unknown drivers, and no measurements.

Otoh, there is freeware measurement software, and ANY mic will yield reasonable results in the crossover range. You can then use that to set levels and tweak away! :D

_-_-bear
 

Attachments

  • CERWINVEGASPK1.jpg
    CERWINVEGASPK1.jpg
    52.3 KB · Views: 185
The only "equipment" you need is a random microphone and ur soundcard in ur computer...
it won't yield laboratory grade results but you don't need that quality.

Then you can start with an online xover calculator for the xover... that will get you in the ballpark. Next you set up the drivers with two sets of BP on the back, one for each driver, and you try some xover ideas and measure them. I'd start with a parallel network, since you can change each driver's response independently.

I think that 2.5kHz to 3.kHz is likely an ok place to try first.

But otoh the best thing to do is to slam the mid and tweeter into the box and measure them without any xover from a distance of about 6" on axis... then you can see if you have a relatively flat response or some major peaks to cope with... then decide where
to cross them over.

You can use freeware for the freq response checks. Post the results up here.
You can get a lot of good advice.

The reason I say a random mic is that you are really only looking at something in the middle of the response range, and most mics are flat enough there... you only need to see the relative response of the two drivers...

might as well take the leap!

_-_-bear

btw a very good choice for a cheapo mic is almost any electret condenser element, even from Radio Shack... all you need to make it work is a 9v battery, a cap and some cable...
 
Thanks so much for the responses. You've motivated me to do more than just plop them into the cabinets and hope for the best.

What do I measure first? I came across a link that describes measuring resistance with a simple test jig and ARTA. The alternative (mentioned in the same post) is Speaker Workshop and the Wallin jig.

If I take the measurements with the demo version of ARTA (which can't save to files), will I get enough info to model a crossover? What do I want to measure? Can you give me a link to a specific online crossover designer? I want to see the target data so I can figure out what to measure.

The Woofers are 16 Ohm. I'm going to have to run them in series to get down to an 8 ohm system. Do I test with them wired together? Do the woofers in series require a particular arrangement (e.g., TMM vs MTM)?
 
you can grab results with a screen save and paste procedure...

look at the curves of the drivers with nothing in series with them -
for the tweeter put a ~15ufd cap in series to keep from frying them... keep the
levels low for the tweeter... of course if its an impulse or MLSSA (sounds like noise) from the test program, then there is less chance to fry it with normal levels... test first with a known good speaker system? :D Get the procedure down... you'll be surprised how bad a "good" commercial speaker tests...

_-_-bear

You want a program that draws a curve, not a 1/3 octave type bar graph thingie...
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.