Tweeter Help

[opus007]

I was rewiring my xovers today and needless to say dropped one and broke it.I have been experimenting with different tweeters in these huge old fisher cabs.I have a 3-way Eminence cross over and the drivers are Eminence 15" woofers ,Pioneer 5 1/2" mid and now I need new tweeters.I have put Infinity and a unknown brand in them and they are all too bright with a lot of sibilance.The tweeter i broke was a KLH from a model 9912 liqued cooled dome.This was the best match so far for this setup I had found.Any good suggestions for what to replace the tweeters with.I am looking for a good clean (not real bright) tweeter.Any help would be appreciated.the KLH gave good clean, crisp,detailed sound and that is what i am looking for.Thanks

[wintermute]

If you still have any of the old tweeters you have tried, it would be worth (If you haven't tried this allready) trying to attenuate them. The problem could simply be that the tweeters are way more efficient than the other drivers.

do a search on LPAD and you should get the idea. If you have the specs on the drivers all the better, for instance if you have a tweeter rated at 93db efficiency and a woofer rated at 89db as a starting point you could try attenuating the tweeter by 4db (may be too much, but somewhere to start).

You can make your own with a couple of resistors, or you can buy variable ones (not so cheap an option).... the variable ones have the advantage that you can keep trying different settings till you find something you like.

Regards,

Tony.

[opus007]

Thanks for the advice .I bought a pair of variable resistors from radio shack and installed them in the cabinets.I backed off the tweeters a half a turn (just enough to bring it down to the level of the mid) and now it is nice and smooth and the sibilance has disapeared.What a difference.Thanks again.

[wintermute]

No worries. Glad it's made an improvement, I may have given slightly the wrong impression when I said "variable ones" I was talking about variable LPADS, but if the variable resistors have done the trick then I guess that's ok

Regards,

Tony.

edit: I hope that they were pretty high powered variable resistors???? If not you better not crank the volume up too high! Might be best to measure the value and replace them with some 5 or 10W resistors.

[opus007]

Could you explaine variable LPADS ? Not sure what you mean .As for the variable resistors they are 3watts 20% tolerance 25 ohm.If cutting down the voltage to the tweeter at high volumes with these is not good what could be the outcome?Blowing the tweeters or serious clipping?So far i notice no cliping.Thanks.

[wintermute]

The only concern I had about the power rating was that you may smoke the variable resistor. I thought maybe you had bought just a pot (typically 1/4W) which would burn up, shouldn't be any risk to the tweeter (apart from maybe catching your speaker damping material on fire

I think 3W should be fine on a tweeter.

an LPAD is basically a resistor in series with the speaker with another in parallel with the speaker, the ratio between the two resistors is such that the impeadance the xover sees is the same as that of the speaker (if you only use one resistor in series you get the cut but you also change the impedance the xover sees which changes the xover freq (in your case this may or may not be desirable) but you are happy with the improvement so that's good

A variable LPAD is basically two variable resistors configured in such a way that one is in series and one in parallel with the speaker, and as you turn the knob, the value of the resistors change in such a way as to keep the impeadance constant (normally 8 ohms)......

This site http://www.lalena.com/audio/calculator/lpad/ has an lpad calculator I've attached a picture from there which shows the basic layout.


Didn't mean to get you concerned, I think that if you are now happy with the result that's the main thing!!!

Regards,

Tony.

[opus007]

Thanks .Let me see if I have this correct and please bear with me as I am new to this and appreciate the help.Lets say I was to remove the variable resistor and wanted to bring the tweeter down 4db.Would I get a 4ohm resistor and put it in line on the + lead to the speaker.Then I would get a ? ohm resistor and put it inline from the + and - speaker leads ?And this would keep the speakers ohm rating at what it is supposed to be?The tweeter is rated at 8ohms.Do i Need to use variable resistors?if so wouldnt I have a hard time getting the correct settings on both?It sounds ok now but I woulkd like to do it the correct way.Thanks

[opus007]

Also is this correct.By using the link you supplied and the speaker at 8 ohms and reducing it 4db the calculations the site gives me is R1=2.95ohms and R2=13.68.So would I round that off to the nearst number and it would be R1=3ohms and R2=14ohms.Do they make a 14 ohm resistor or could I combine serveral to make a 14 ohm resistor?I have no idea of how many dbs I have backed the tweeter off with the variable resistor and this is whay I ask about using regualar resistors.

[wintermute]

Ok I'm now getting out of my depth, If someone else wants to step in....... The way you have done the attenuation isn't strictly wrong, it's just that the freq response of the tweeter will likely be attenuated by different amounts at different frequencies...... If you are happy with the result you probably shouldn't worry too much.

here is a link to a cheap variable l-pad, buying a couple of these is probably the least hassle (most flexible) way of doing doing the attenuation, in a crossover freindly manner.

http://www.partsexpress.com/pe/showd..._ID=5647&DID=7 You definitely shouldn't try to make one yourself using two variable resistors

There are lots more there if you do a quick search on L-Pad on partsexpress web site.

I'm sure there must be a formula somewhere for calculating the "theoretical" db cut that your current series resistance is making, If you had that formula you could measure the resistance of the pot with a multimeter and then plug that into the formula to work out how many db's of cut you are getting, but I stress this is only theoretical, as the cut will be different amounts at different frequencies. If you then use db cut on the site I posted a link to before you could work out your R1 and R2 values. Be warned though that it will probably sound different to what it does now, as the proper L-Pad will pretty much cut the entire freq range by the same level.

With respect to resistor values, yes you can make a bigger (or smaller) value resistor by wiring multiple resistors in series or parallel. In series they simply add together, parallel is a bit more complex, for parallel Rtotal = 1/(1/R1 + 1/R2 + .....).

Honestly if you are happy with the way the speakers now sound I wouldn't worry about it too much, you may go to the trouble of changing it and find you don't like it as much, I guess on the flip side though you might like it even better...... But trust me, the variable L-pads are the easiest way of doing it And once you've set them to the level you want, you can allways disconnect the tweeter measure the two resistances, and replace with discreet resistors if you wish (purely optional, I've thought about it but still have my variable l-pads in the system).

Regards,

Tony.

PS. Sorry if I have subtracted from your initial happiness of getting a big improvement from your speakers, it wasn't the intention! It's a bit like something Circlotron said in another thread, where he said (paraphrasing here) he would measure something, find out about a problem he didn't know he had and then feel compelled to fix it!!!

[opus007]

Thanks again.One more question.If i was to go with the l-pad link you posted would I go with a mono or stereo one for each side.Also shouldnt its watts be the same as the tweeter.The tweeter is rated at 50w rms or 100w rms. I can not really remember?Not sure i will do this but wanting to know for future reference.Thanks.