Design calls for .7mh, is .68 OK?

Just finalizing my project with the Vifa P13 and the XT ring radiator. Found that I like the proven design of the Okara the best. My quick prototype revealed to me that this is a very neutral speaker , so it is this design I will end up with.

The low pass section calls for a 1mh and a .7mh inductor.
I will be ordering the parts from Solen and they have the 1 mh but not the .7 mh. Will a 6.8 mh be OK to use ?
Likewise the high pass calls for a 6.7 mf cap. Will 6.8 mf be good ?
I'd rather not bypass caps to get the value if possible so I can save precious space inside the cabinet. (plus I don't want to spend the extra money)
Also I noticed that all the inductors have resistors preceeding them. I assume it is the value of the DC resistance? Is this really necessary, or will I be alright without them?
The Okara uses 12 gauge foil inductors. As I am on a budget for this I was thinking of using 16 gauge air core inductors. Will all these substitutions seriously compromise the design? Sorry for the multiple questions, I want to get it right the first time and not have to reorder stuff again :rolleyes:
 
its all a matter of how picky you are :) If you like doing things right and want to use their standard inductors (over the hepa-litz, I think they make more sens for low frequency crossovers, having one large gauge conductor) then I suggest you give them a call and ask if they'll make one of the specific value for you. I've been to their store, and I do believe they roll them on the spot. If not, I suggest you get a 0.75 and unwind it to 0.7.

There's a spreadsheet somewhere that someone designed for calculating how to unwind an inductor, I remember zaph had a link to it on his webpage, but I can't seem to find it anymore. You definately want to find that again if your looking into getting the propper value.

Hope this project is going to turn out nice.
 
Passive components such as capacitors and inductors are all manufactured within certain tolerances, typically 5% or even 10%. Even when you use a LCR meter to measure these components, the meter itself has published tolerances. For instance, the Elenco LCM-1950 meter that I purchased from Parts Express has a resistance accuracy between 1 to 5%, and capacitance/inductance accuracies of 5%. So if your capacitor has a 5% tolerance, and your meter has another 5% tolerance, well you can do the math. Hope this helps.

Here's the link for winding your own inductor:

http://lalena.com/audio/calculator/inductor/

Capacitor codes for identifying tolerances:

http://www.montek.com/tutorials/cap_codes.html

Hong
 
Thanks for the replys, I think I'll end up using the standard 14 gauge inductors for the low pass and I will bypass caps with smaller film and foil to get the exact values. The picky part of me wouldn't have it any other way. Thanks sberube for mentioning it!

This always happens but,... just when I thought I had my design finalized I read info that makes me second guess it.
The Okara design calls for a .22 cubic ft. enclosure tuned to about 63 hz. That is approx. what my test enclosures are now but I read numerous suggestions stating that a good recommended size would be .38 tuned to 60 hz. (slightly larger)
Does this enclosure size call for deeper if not somewhat looser bass? I don't mind this so much as the P 13 lacks bass anyway.
Would the x-over values need to change if I went with the larger enclosures? or would the same x-over be good?
Any one have a thought on this?
 
Changing the volume of the box will definately make a big difference, but from what I understand (might be wrong here) only the low frenquency gain will change, your x-o point gain should stay the same. Although it might be possible that the entire range of your woofer's gain changes.

As for giving you a looser bass, I really don't think so.

BTW, if you haven't ordered your parts from solen yet, try this. Get a 47 Ohm resistor, and use it as a bypass on your inductor for your bass driver. Let me know what you think... It gave out some pretty interesting results on my part.

I also used a cascade (instead of just a bypass - eventhough bypass was already real cool) of caps for my tweeter (Vifa XT19 ring radiator) and was very impressed by the change. If you can/will afford very expensive caps, then it might not be worth it, but with solen chateauroux, its definately worth the extra $$ your going to pay.

Too bad your not in montreal, I could have lent you loads of components to do value testing in your x-o.

Good luck and let us know how it turns out.
 
...and adjust the phase change, impulse response. Salas, also said that for "critical" listening that it can change the speaker's tonal quality or, magic. I guess salas has a very keen ear, I'm still working on it.

But then, a passive x-over smears and glitters phase and power all over the place, so then an active xo must be used. And then, and then.....oh no! I'm caught in the loudspeaker designer vortex!!!!
 
sberube said:
Changing the volume of the box will definately make a big difference, but from what I understand (might be wrong here) only the low frenquency gain will change, your x-o point gain should stay the same. Although it might be possible that the entire range of your woofer's gain changes.

As for giving you a looser bass, I really don't think so.

BTW, if you haven't ordered your parts from solen yet, try this. Get a 47 Ohm resistor, and use it as a bypass on your inductor for your bass driver. Let me know what you think... It gave out some pretty interesting results on my part.

I also used a cascade (instead of just a bypass - eventhough bypass was already real cool) of caps for my tweeter (Vifa XT19 ring radiator) and was very impressed by the change. If you can/will afford very expensive caps, then it might not be worth it, but with solen chateauroux, its definately worth the extra $$ your going to pay.

Too bad your not in montreal, I could have lent you loads of components to do value testing in your x-o.

Good luck and let us know how it turns out.

sberube, A 47 ohm resistor bypassed on the inductor? I would assume this would be to sove any baffle step issues? There is already a 19 ohm resistor bypassed for this reason. Wouldn't a 47 ohm really exagerate the BSC effect? I'll give it a try anyway just to see but I don't know...
:confused:
When you say you cascaded the caps aswell as bypass them, do you mean to say that you put one after the other in series with each other? What exactly would be the reason for this?
Sorry for all the questions but I'm learning a lot from you guys (girls?)- don't want to be a sexist you know:D
 
no actually its not for BSC. Not that I know of... just a simple bypass (between both leads of the inductor). I think it'll make the lower frequencies take this path instead of the inductor path which will reduce the length of the path to your drivers which makes the bass more defined... clearer. Of course ever so slightly...

As for the caps, I bypassed at first (needed 11uF and used some combo to get there with a 0.47 film and foil) which sounded really good. But then what I did was to go with the following (in parallel):

6 uF 200Vdc
3.3 uF 330 Vdc
1.5 uF 400 Vdc
0.47 uF film and foil 630Vdc

It was really worth it.
 
It has been suggested to me by someone else on another forum (Madrok) to put the P13 in a .38 cubic ft box as suggested earlier , but rather than the standard bookshelf style, design a mini floor stander with an I.D. of 5"x4"x33",then insert the port at the bottom of the front or back? of the enclosure (tuned to 60 hz.).
I kind of like this suggestion because I don't have to worry myself about making a stand for them :) . The only thing that concerns me in this case is that the distance from the back of the woofer to the back of the enclosure is now 1.5" (the depth of the woofer is 2.5") Would this be enough breathing room for the woofer?
Ahh Man! decisions, decisions!
 
Bose(o) said:
...and adjust the phase change, impulse response. Salas, also said that for "critical" listening that it can change the speaker's tonal quality or, magic. I guess salas has a very keen ear, I'm still working on it.

But then, a passive x-over smears and glitters phase and power all over the place, so then an active xo must be used. And then, and then.....oh no! I'm caught in the loudspeaker designer vortex!!!!


eek don't down passives here :p

I would need quite extensive ABX testing to believe that a 5% tolerance .68 mh sounds different than a 5% tolerance .7mh

even 1% seems ridiculous to me

that .68mh might be .72mh... and that .7mh might be .67 mh

and if you can tell the difference between 2.1khz HPF and a 2.09khz HPF... then you need to patent your damn ears man
 
By the way I run 2 P13's with a Xt25 in an open baffle MTM arrangement. My XO design called for 0.40mH in series with the woofers... I used two 0.2mH in series to get 0.4mh. After listening, the sound was too laid back for me, I first bypassed one 0.2mH and now both... I like the sound better this way with 0.40mH completely bypassed with Alligator clips (yes I said it, ALLIGATOR CLIPS) till I finalize the voicing... so much for calculated values and real life listening.

I will add that the P13 is a soft (not forward) sounding speaker that is forgiving and easy to work with.... your 0.68mH will sound just fine. ;)

The issue now is that depending on some recordings, the sound is sometimes bright and sometimes laid back.... I am now in a position where bypassing the 0.4mh (i.e. the P13's decay /roll-off naturally) is in a middle ground between power handling and clarity....
 
Alright, I think we've agreed that the inductor is all nice and fine. And, if you want to, just slip in a special note asking for the "right" inductor. In my experience, I've rounded up and down and have found that only REALLY expensive test measurement will reveal your decision of such compromise.

Okay, bypassing inductors and capacitors-now this seems interesting. Damn you guys! I was planning to build my speakers and external xo and finally be done with my system until I finished University.

Can someone please explain/provide wiring diagram examples please? TIA
 

Jay

Banned
2003-02-11 9:02 am
Jakarta
Wait wait wait... :smash:

Why is it, people always use the tolerance issue against opinion on the effect of slight changes in components values? Why bother explaining the not-so-advance math? :D

People know the effect of changing 0.68 to 0.70mH is from experience, not from playing around with calculator. Measure your inductor, then unwind and measure again. The 0.02 is relative, has nothing to do with inductor and LCR-meter tolerance!

In your speaker, replace the 4.7uF with 5uF, and I believe the sound will equally 'good'. Or downgrade to 4.2uF, it will still be equally 'good' too. Am I wrong here? Or simply run your P13 fullrange, you will still love it. Am I right again?? Or now pull a series xo, do you now realize that your parallel designs sound worse?

In Okara design, the 0.7mH is not critical. The 6.7uF is more critical. The effect of slight change in component is real, you just don't get the critical network. High order parallel network is critical, series networks are 'less'. And it is not simply about crossover frquency shift or changes to SPL response.

Crossover network is a headache full of compromise. For example, if for a reason (eg. tweeter low fs) we HAVE TO cross our woofer high, we will face the resonance, if we use notch filter we will loose the dynamic, so we will cross the woofer as high as possible, high enough that we wont need a notch. What does it mean? It means we may find a position where we cannot go up or go down. OR, a position where we cannot use higher or lower component value.

Okay, I will patent my golden ears soon enough
:devilr: :devilr: :devilr: :smash: :whazzat:
 
Jay said:
Wait wait wait... :smash:

Why is it, people always use the tolerance issue against opinion on the effect of slight changes in components values? Why bother explaining the not-so-advance math? :D

People know the effect of changing 0.68 to 0.70mH is from experience, not from playing around with calculator. Measure your inductor, then unwind and measure again. The 0.02 is relative, has nothing to do with inductor and LCR-meter tolerance!

In your speaker, replace the 4.7uF with 5uF, and I believe the sound will equally 'good'. Or downgrade to 4.2uF, it will still be equally 'good' too. Am I wrong here? Or simply run your P13 fullrange, you will still love it. Am I right again?? Or now pull a series xo, do you now realize that your parallel designs sound worse?

In Okara design, the 0.7mH is not critical. The 6.7uF is more critical. The effect of slight change in component is real, you just don't get the critical network. High order parallel network is critical, series networks are 'less'. And it is not simply about crossover frquency shift or changes to SPL response.

Crossover network is a headache full of compromise. For example, if for a reason (eg. tweeter low fs) we HAVE TO cross our woofer high, we will face the resonance, if we use notch filter we will loose the dynamic, so we will cross the woofer as high as possible, high enough that we wont need a notch. What does it mean? It means we may find a position where we cannot go up or go down. OR, a position where we cannot use higher or lower component value.

Okay, I will patent my golden ears soon enough
:devilr: :devilr: :devilr: :smash: :whazzat:


I'll give you $500 if you passed an ABX test with with 6.7uf cap 1% tolerance and a 6.5uf cap 1% tolerance on a simple HPF

you simply would not do it

and many of the reasons you state is why I and many others refuse to go passive because of the infinite problems

it's very laughable for someone to state that they can hear differences that are within the tolerances of the components... it's quite laughable that to 99.999999% of people in this world even if you precision matched a pair of caps to less than .01% tolerance that you STILL couldn't pass an ABX test with a 6.5uf and 6.7uf cap
 

Jay

Banned
2003-02-11 9:02 am
Jakarta
Forget the tolerance and the E=mc^2, I guess my message didn't delivered.

6.5uF + 100nF + 100nF, that's about 6.7uF. Find a proven design and add a 100n to any 6.5uF in the network, listen and go back after a month. Capacitors (or the join?) need to 'burn-in'. Wait, I hear you laughing. You need more time to perceive the change in capacitance than in inductance.

If you are close, I would be very very happy to accept the $500 challenge. But of course, I should have the right to choose or to create the crossover network (and the drivers too!). This is where the 'critical' word plays its role! :smash:

I was describing the infinite problems with passive parallel networks (not so good example tho) was to explain that you cannot just plug any components and there you have a good speaker.