Home Theater Forum and Systems banner
1 - 9 of 9 Posts

·
Registered
Joined
·
168 Posts
Discussion Starter · #1 ·
Related to this thread http://www.hometheatershack.com/forums/diy-speakers/44040-my-multi-media-computer-speakers.html

I can not put my finger on why my tweets are getting static, interference, noise, whatever you want to call it.
I have tried every variation I can think of rewiring how they are setup. I think at first I was not picking up on the unwanted sound, both because of the general excitement of having them done and the fact it isnt so obvious at higher volumes. One thing that is out of design criteria initially were the L-Pads. I did not intend to purchase 100w models, but did. Way over kill, but cant imagine it would hurt a tweeter, or am I wrong? They are of course 5 ohm models that I added a 3 ohm resistor on the + terms. in series/parallel for a nominal 8 ohm at the crossover. The amp driving this is the Dayton APA 150 (75w RMS per channel). PE shelf 3500hz crossover between the amp and tweets/mids. I tried bypassing the LPad and letting them run wide open individually, still have issue. Changed from the cheesy RCA cable to a new monster cable. Bought a $100 monster power filter........... I can say that the breakup, static, whatever seems to be more pronounced in the beginning of the song and seems to atleast soften. I cant handle it though, especially on softer acoustic songs the static cuts through me like a knife. Suggestions please?
 

·
Registered
Joined
·
348 Posts
What happens when you use a receiver as the source? I would first run things all the same but just with an AVR and see what happens, imo. Also, are your ears literally that close to these? I'm referencing the table and where your listening position is relative to your computer.
 

·
Registered
Joined
·
168 Posts
Discussion Starter · #3 · (Edited)
What happens when you use a receiver as the source? I would first run things all the same but just with an AVR and see what happens, imo. Also, are your ears literally that close to these? I'm referencing the table and where your listening position is relative to your computer.
Firstly, I just measured my listening position to the center of the speakers and its 42-48". Used my SPL meter at what I consider a comfortable level and its 70-73db. I have not tried to use a receiver as the source ILO of the Mac. I can say that the previous Hivi mid and same Hivi tweet with the same crossover, resisters, etc did not do this. It was with a S-onance (it wont let me type the name) amp though........... So its possible. I wish I had a clean box that could go from 3mm to RCA out. That would tell alot.
 

·
Registered
Joined
·
348 Posts
hmmmm, interesting. 2nd thing I'd do (this is my engineering thinking) disconnect one tweet and listen, then one tweet and one driver, ect. I would track that noise down by process of elimination and start with one driver then 2 and so on. Still maybe source noise though even if it hadn't done it before. Electricity just like sound is a lazy thing, always looking for the easiest path!

Just like coil packs in an Audi, its not if it will go bad but when...
 

·
Registered
Joined
·
168 Posts
Discussion Starter · #5 ·
Eek. I had one other idea, that may be less painful. I am going to get one of my older speakers from my brother, gave them to him recently. Since they are so similar in hardware, crossover, etc, I am going to see if the same problem is present. I also have a neat metering tool that could come in handy. Its made by beretta and has the model number 12gablowitup.
 

·
Registered
Joined
·
302 Posts
Go back to your speaker design. What is your crossover circuit? I've reviewed your build thread without seeing anything but a request for your XO simulation (post 3). I'm looking for a schematic with all connections back to the amp. You mention a PR shelf filter, a useful component but not something that would act as a crossover.

What you describe sounds a lot like a tweeter running full range, making tons of harmonic distortion below where it should reasonably cross. Being harmonics, distortion ends up as highs; 2K HD occurs at 4K, 6K, 8K,10K, etc. and the odd harmonics are typically judged to be grating on the ear.

Have fun,
Frank
 

·
Premium Member
Joined
·
1,923 Posts
Is one of the tweets blown? I've had tweets give up the ghost and go silent, but I also had a tweet (was a fabric dome) that screeched and sounded terrible after it went.

I agree with the procedure, disconnect them one by one and see if the sound changes. Maybe you had one hooked up in full range on accident and that's killing them slowly...

It would be helpfull to see how you have everything wired. With the textbook Xover intended for "8 ohms" there's a good chance that your tweets as they are wired are more like a 3 or 4 ohm impedance, which would correspond to a completely different (if not mystery) Xover point.
 

·
Registered
Joined
·
168 Posts
Discussion Starter · #8 ·
Both of you guys are asking good questions. Fbov, if you look back at the original thread you will see I said I have a pre assembled parts express crossover. Dont know if I stated it or not, but its crossed at 3500hz. The sound I am hearing could probably be described as sound well below the intended operating range of the tweet, but with the fact that both speakers have the identical issue and a pre-assembled and well labeled crossover I dont think that is it. Terminal>crossover in>tweeter/woofer out>drivers

fusseli, I have isolated atleast 3 of the 4 tweeters by themselves and seems to change nothing. I am really a bit baffled. I ordered 2 ART clean boxes that will be here tomorrow (needed them anyway) and am going to snatch one of my last speakers I built from my brother tomorrow to compare. One of the few variable left in my mind is this APA150 amp. Hopefully I can isolate it further tomorrow. Thanks for the input!
 

·
Registered
Joined
·
302 Posts
My apologies, I misread your build thread; no mention of a shelf filter, just the 3.5KHz PE XO. Pre-built XO's have significant performance issues (not rational to design the peg before you've seen the hole), but not as severe as the problems you describe.

Any chance there's a level mismatch between your Mac output and the APA's input levels? Are you using a line-level output from you Mac? Headphone jacks are not low-level outputs; you could be overdriving the amp inputs.

Hang on... shelf filter comments are in this thread, post 1:
"PE shelf 3500hz crossover between the amp and tweets/mids." or should this read PE "off the shelf" 3500Hz XO?

Do you know what driver impendence this XO assumes? If it's for 8 ohm and you're using 4 ohm, your XO frequency is half what you think it is and the tweets are screaming with HD like I thought they might. How are your multiple drivers wired together? Drivers have a DC resistance you can measure with am ohm-meter. The XO sees a highly variable load with frequency (4-50 ohms) that's only grossly related to the DCR. How you arrange them is critical to the final impendence seen by the XO, and thus critical to the XO frequency.

Let me ask again, please show us a circuit diagram for your connections. I don't care if you scan a pencil sketch, give us a picture so the words aren't in the way.

Have fun,
Frank
 
1 - 9 of 9 Posts
Top