First DIY !

Status
Not open for further replies.
Jump to Latest
#61
Hi Jay_W3;

That's an impressive amount of work you did for Gareth, nice job.

My experience is that DCR for a woofer low pass is quite important, I know Gareth is hardly building a state of the art design, but I've always believed in picking the "low hanging" fruit.

An example: Say the woofer is 8 ohms and the amp has a damping factor of 50 (or ~0.16 Ohms output Z) by adding a 0.4 Ohm DCR coil damping factor is now 14.2. In my experience this change is quite audible. Lower DCR will generally cause a taughter bass with less hang and a little more punch. This is part of the reason why absurdly over priced and over sized speaker cables do tend to sound better.
 
#64
#65
MaVo said:
isnt that unimportant, considering that the voice coil will kill every damping factor there ever was with its many ohms? i thought damping factor was just marketing mumbojumbo.
Click to expand...
I base my observation on listening not marketing, to me the difference is audible. Your mileage may vary.

This might help to explain why vacuum tube amps (except the most powerful very expensive) typicaly don't do as well in the bass as a solid state amp.
 
#66
Jay_WJ said:
I use FRDC tools for in-box response modeling:

http://www.pvconsultants.com/audio/frdgroup.htm

Unibox for bass box modeling and BDS for baffle diffraction/loss sims. And Speaker Workshop for crossover modeling.


Again many thanks Jay, the links to software are excellent and I feel I will be using these in the future. Whilst waiting for the crossover components to arrive I am thinking about building another pair. Not yet though, this will be in the future. I don't want to be misconstrued as blasé with regards to the present project or ungrateful for your help.

One thing I have noticed in your link to your speaker design tips is the use of L-Pad attenuators for the tweeter. Why is an L-Pad used as opposed to a pi-filter?

Thanks
Gareth
Click to expand...
 
#67
gareth said:
One thing I have noticed in your link to your speaker design tips is the use of L-Pad attenuators for the tweeter. Why is an L-Pad used as opposed to a pi-filter?
Click to expand...

You mean a pi filter used for AC filtering?

No, an L-pad used in speaker crossovers should give the same level of attenuation at all frequencies. It consists of only resistors. It's different from "L-filter."
 
#68
Jay_WJ said:


You mean a pi filter used for AC filtering?

No, an L-pad used in speaker crossovers should give the same level of attenuation at all frequencies. It consists of only resistors. It's different from "L-filter."
Click to expand...


I believe you can use pi-filters ( t-filters and h-filters) connected across the tweeters terminals to provide a more balanced attenuation. These networks are resistor based.

I have just found a good pocket electronics book and they seem to be pretty good. There are also neat little tables that you can use to select values.

Have you any experience of these?

Gareth
 
#69
gareth said:



I believe you can use pi-filters ( t-filters and h-filters) connected across the tweeters terminals to provide a more balanced attenuation. These networks are resistor based.

I have just found a good pocket electronics book and they seem to be pretty good. There are also neat little tables that you can use to select values.

Have you any experience of these?

Gareth
Click to expand...

As far as I know, T filter, Pi filter, etc use shunt capacitor(s) and series inductor(s) to remove the AC component.

I don't know what you mean by "more balanced attenuation." In passive crossovers for loudspeakers, tweeter attenuation (not filtering) is done by several different topologies of resistors. The traditional L-pad is used to attenuate the tweeter level while keeping the tweeter's impedance constant (e.g., 8 ohms). The reason is that old, textbook HP filters works only with a certain tweeter impedance assumed.

But these days with CAD available, we no longer need such tables or formulas. Sometimes, a single series resistor is used either before the HP filter or after it. Other times, different topologies are used (e.g., traditional L-pad type, a series resistor before the filter and a shunt resistor before the driver, and so on).

The reason why these different topologies are used is that they exhibit a bit different behaviors in attenuation across frequencies because the tweeter's impedance profile is not perfectly flat.

One advantange of using the traditional L-pad topology is that it can help a bit to flatten the resonance peak.
 
#70
Jay_W3 and gareth;

Both T and O pad purely resistive attenuators exist. The O pad is very popular for balanced circuits. I don't believe I've ever seen a balanced crossover and I'm not clear on any advantage.

The T pad allows you to maintain input impedance independently from attenuation, usually not important if you designed your own crossover. The L pad uses the fewest parts while maintaining a low impedance for the crossover reactive components to work their best. Because the parts count is low adjusting the attenuator for speaker "voicing" is simple and inexpensive. It's not unusual to see a wire wound potentiometer used in this location, although IMHO they don't sound very good.
 
#71
hermanv said:
Jay_W3 and gareth;

Both T and O pad purely resistive attenuators exist. The O pad is very popular for balanced circuits. I don't believe I've ever seen a balanced crossover and I'm not clear on any advantage.

The T pad allows you to maintain input impedance independently from attenuation, usually not important if you designed your own crossover. The L pad uses the fewest parts while maintaining a low impedance for the crossover reactive components to work their best. Because the parts count is low adjusting the attenuator for speaker "voicing" is simple and inexpensive. It's not unusual to see a wire wound potentiometer used in this location, although IMHO they don't sound very good.
Click to expand...


So are there balanced crossover designs out there? It sounds as if there are based on what your saying. Wouldn't it be better to use a balanced configuration, afterall speakers are AC, or am I getting a bit confused?

Gareth
 
#73
gareth said:
So are there balanced crossover designs out there? It sounds as if there are based on what your saying.
Gareth
Click to expand...
I said "I've never seen a balanced crossover".

Since most crossover designs are series circuits, it matters not at all if you put 1/2 the inductor in series with each side of the driver or the full value on one side only. Parts count and cost is lower for an "unbalanced" design. Almost all solid state amplifiers and most vacuum tube amplifiers tie one side of the speaker output terminals to chassis ground, for these designs a balanced crossover makes no sense.

Keep it simple, do the conventional thing until you are expert enough to explore new or fresh topologies with a thorough understanding of why things work.
 
#74
MJL21193 said:



I'd use the one designed by Jay. Nicely done, you should appreciate the effort put into designing it.

Forget about all of that other BS until you understand filter operation better.
Then you can discount it.
Click to expand...


I completely and totally appreciate what Jay_WJ has done for me, really I cannot express this enough. Whilst waiting for the parts to arrive I am looking around and reading different things, passing time so-to-speak. Probably I am filling my head full of nonsense that I have no need of, but after modding my CD63se (elsewhere on diyAudio) I am looking to learn more if you know what I mean and I am questioning you guys because you know more than I do.

I have no intention of deviating from what Jay has suggested either, honestly I can't wait to get these speakers up and running!

Thanks
Gareth
 
Status
Not open for further replies.
Top Bottom