L-pad for tweeter level control?

[Shakeydeal]

17 minutes ago, babadono said:

different strokes for different folks. If a piece of music needs more bass(to my ears on my speakers in my room) I turn up the bass. How many op amps do you think were in between the microphone at recording time and your speakers at playback time?

I hear this all the time, not a very good argument IMHO.

 

That's like saying, "my car is wrecked on the left side already, I don't care if someone hits it on the other side".......

 

Why add unnecessary crap to the playback chain just because it is already in the recording chain?

 

I once heard a system with a fancy 5K room correction device in the signal chain as a preamp. We took it out and plopped in an old Audio Research SP3 tube preamplifier and the improvement was amazing. Less was certainly more that day.........

 

Shakey

[CECAA850]

2 hours ago, Shakeydeal said:

I once heard a system with a fancy 5K room correction device in the signal chain as a preamp.

I've seen that as well.  The problem is normally the lack of proper set up and implementation.  Room correction devices can actually make a system sound worse if care isn't taken during installation.  The device is normally only as good as the person installing it and the time they put in it during the initial install.  I'm not aware of any analog devices that can fix issues the way a room correction device can however. 

[SWL]

Whatever it takes to make it sound better to YOU. Who cares what anyone else thinks.

Been using a DEQ2496 strictly digital with an outboard DAC for a few years now and haven't looked back.

I'll be back..... someone's knocking at my back door.[emoji6]

Sent from my SM-G950U using Tapatalk

[babadono]

Was not trying to start an argument. That's why I said DSDF. It's all good.

[Guest wdecho]

6 hours ago, rjp said:

Interesting you should mention this. I just measured the horns and I get 16 ohms at 1KHz and 8 ohms at 10KHz. There are no other frequency options in between on my LCR meter.

The pads I have are 8 Ohm.

 

Just go by what is printed on the driver. You start measuring and things get too complicated.  

[rjp]

Update: I installed the L-pads and they seem to work as expected. That is, the level of the tweeter goes down smoothly as I turn them. To my ears they still sound just as good so I don't think the pad is reducing fidelity at all. I seem to like them with just a very small amount of attenuation. I would estimate less than 1/8th of a turn from full on. Not sure if these are linear or logarithmic so no way of knowing the attenuation at this point. I had to go into work. Will listen more tonight.

 

So In summary, I would say adding L-pads to the RP-260F tweeters is a viable solution. If I decide to make it permanent  though, I'll have to decide where to mount them. I don't want to drill holes.

 

With further listening and measuring I hope to try to quantify the effects on the frequency response better using my "trusty iPhone spectrum analyzer" and white noise generator. Actually, I also have a calibrated signal generator that I could use to sweep the spectrum with CW tones. Then I could just measure SPL. That would be more accurate, but it's also more work to set up. I have all this stuff from radio electronics work but this is my first foray into loudspeakers so I'm trying to learn by asking questions. Thanks for all the great discussion today. Looking forward to more.

 

 

[Marvel]

If you find a place where it always sounds right for you, you can measure the values on the l-pad and make one with fixed resistors, and put them on the crossover.

 

Bruce

[pzannucci]

16 hours ago, wdecho said:

L-pads work fine for what you want to do. Not really hard to hook up if you just remember the center lug goes to the driver. If the outside ones are inverted and not  hooked correctly the pot will be reversed. Not really that big a deal. Outside lug input, output lug ground, center to driver. Using just a resistor will change the frequency of the X-over point of the tweeter, the reason for L-pad. 

Not quite correct.  Several things:

1.  any variation in level changes the crossover over point

2.  the lpad is not typically concise and how do you know the impedance at crossover to have a properly designed lpad,  so it will likely not provide exact impedance at the the frequencies required to match the components

3. one or two ohms worth of resistor is within margin of error and since it is not the loaded side of the crossover (particularly with 12db), will have minimal effect on the crossover point outside of my point 1.

Or just put an l-pad on the input of the high frequency and let the amplifier deal with most of the impedance change.  You don't have to worry about the impact on the components thinking they are acting on a 12ohm impedance vs a 8 ohm presented by the l-pad (50% off wouldn't make it happy).

 

13 hours ago, wdecho said:

Leave in the zobel. It has nothing to do with the crossover frequency. It tames a too hot driver or horn. Commonly used in quality speakers to refine the sound. Too much is not a good thing though. For instance someone went entirely crazy when designing the AL crossover using zobel networks in all the drivers for a more flat frequency response. Being more flat is not always best but I expect Klipsch has learned from their mistakes of the past and the zobel is needed with your speakers. If you want to experiment then certainly take it out and judge for yourself. Below is a calculator I use to determine the proper attenuation instead of a chart. Either one works just fine.

 

https://www.diyaudioandvideo.com/Calculator/DriverAttenuationLPadCircuit/

If you are using a zobel, the zobel's intent is to flatten the impedance to allow for the crossover to work more efficiently and not rely on the rising impedance of the driver which totally messes with the crossover point.  Keep that in mind when using a zobel.  If you think of it logically, using a zobel with a cap and a resistor, as the frequency goes up across the cap, the cap will allow more power through the resistor, countering the rising impedance of the driver.  It's not typically used directly for taming a driver unless you are trying to split the power between the resistor and driver.  On a lot of SS amps, they almost double their power into the low impedance loads so there really isn't a gain.  On tube amps, maybe some other trade offs on the frequency response.

[pzannucci]

6 hours ago, rjp said:

Update: I installed the L-pads and they seem to work as expected. That is, the level of the tweeter goes down smoothly as I turn them. To my ears they still sound just as good so I don't think the pad is reducing fidelity at all. I seem to like them with just a very small amount of attenuation. I would estimate less than 1/8th of a turn from full on. Not sure if these are linear or logarithmic so no way of knowing the attenuation at this point. I had to go into work. Will listen more tonight.

 

So In summary, I would say adding L-pads to the RP-260F tweeters is a viable solution. If I decide to make it permanent  though, I'll have to decide where to mount them. I don't want to drill holes.

 

With further listening and measuring I hope to try to quantify the effects on the frequency response better using my "trusty iPhone spectrum analyzer" and white noise generator. Actually, I also have a calibrated signal generator that I could use to sweep the spectrum with CW tones. Then I could just measure SPL. That would be more accurate, but it's also more work to set up. I have all this stuff from radio electronics work but this is my first foray into loudspeakers so I'm trying to learn by asking questions. Thanks for all the great discussion today. Looking forward to more.

 

 

Glad you are happy with the way you did them.

[rjp]

25 minutes ago, pzannucci said:

Not quite correct.  Several things:

1.  any variation in level changes the crossover over point

2.  the lpad is not typically concise and how do you know the impedance at crossover to have a properly designed lpad,  so it will likely not provide exact impedance at the the frequencies required to match the components

3. one or two ohms worth of resistor is within margin of error and since it is not the loaded side of the crossover (particularly with 12db), will have minimal effect on the crossover point outside of my point 1.

 

Not sure I follow your point here. Are you saying a single series resistor is a better solution than an L-pad at the tweeter? If so, would you add this additional resistance to the existing 4 Ohm resistor shown in the schematic?

[Guest wdecho]

Just add the L-pad. One resistor will change the crossover point. In other words screw up the crossover. I am not saying that Pzan is right or wrong but it just seems like complicating the matter too much. Sorry Pzan. When one is dealing with audio frequencies, speakers and crossovers there are too many variables to consider with a perfect answer. L-pads have been used by the heavy hitters in speaker manufacturers for years as being good enough so lets not try to complicate it too much for a listener just want to attenuate his driver some. If I did not have a tiny bit of electronic knowledge and experience I would have just added L-pads to my speakers for attenuation and be done. 

 

If you want some idea how complicated just speakesr and frequencies can be just look at an impedance curve of a speaker or driver. It may be rated at say 8 ohms but in actuality it can vary with the frequency a lot. Speaker manufacturers just take the nominal impedance of the speaker or driver and rate it at that and say good enough for use with a crossover. 

[pzannucci]

15 minutes ago, wdecho said:

Just add the L-pad. One resistor will change the crossover point. In other words screw up the crossover. I am not saying that Pzan is right or wrong but it just seems like complicating the matter too much. Sorry Pzan. When one is dealing with audio frequencies, speakers and crossovers there are too many variables to consider with a perfect answer. L-pads have been used by the heavy hitters in speaker manufacturers for years as being good enough so lets not try to complicate it too much for a listener just want to attenuate his driver some. If I did not have a tiny bit of electronic knowledge and experience I would have just added L-pads to my speakers for attenuation and be done. 

 

If you want some idea how complicated just speakesr and frequencies can be just look at an impedance curve of a speaker or driver. It may be rated at say 8 ohms but in actuality it can vary with the frequency a lot. Speaker manufacturers just take the nominal impedance of the speaker or driver and rate it at that and say good enough for use with a crossover. 

You have missed the points about variations due to driver output and such changing the crossover point also due to the reference of tweeter output to woofers.  As you lower the tweeter output, due to the slope of the woofer, your crossover point automatically goes up. Don't rely on the "old manufacturers used them" though I have I have built a fair number of speakers with crossovers from scratch and also used the single resistor technique on a set of RF82II with good affect.  Single resistor for small changes (not large) can work.  An yes I have run things like WooferTester against my built networks to measure the changes due to impedance compensation like zobels and changes due to introducing some type of frequency or impedance modification.  

My Maggie 1.7i's use a single resistor in line and so do other manufacturers today for small changes.  A single 1 ohm or so resistor is not a complicated change, tied between the posts on the back of the speaker and can be used for minor changes, say up to a couple of db, not 6 to 9 db like l-pads.  For that amount, use L-Pads, get the values, and then do a static lpad that works better using high quality resistors if you don't want to use an autoformer.  Additional information is that I have measured P-E's high quality 100watt l-pads and found them to be a couple of ohms off at particular locations at the knob from theoretical l-pad values so I wouldn't be so worried about the one ohm resistor trashing stuff.

 

BTW, autoformers are very good at allowing you to use smaller capacitors in front, as Klipsch does, significantly cutting the costs vs using an l-pad (if you ever want to do a crossover from scratch) because of the much larger capacitor values required.

 

Choose your poison carefully as there are different ways of doing it with varying outcome..

 

7 hours ago, rjp said:

Not sure I follow your point here. Are you saying a single series resistor is a better solution than an L-pad at the tweeter? If so, would you add this additional resistance to the existing 4 Ohm resistor shown in the schematic?

As long as you are happy, that's what counts though to answer your question, "is it better", it depends on if you are trying to accomplish a large change or not.  I found a good 1 ohm resistor tied between the terminals did take some of the edge off.  It will also depend a little on your choice of amplifier.  The speaker and amp are a system.

[Guest wdecho]

I agree that a one ohm resistor is not going to change anything that much but also it is not going to attenuate the speaker that much either, maybe a db or so. It takes a 3db change for most to notice a difference. L-pad, cheap, simple and it works quite well for it's purpose to the average consumer. One can certainly play many ways if one chooses and if what one does sounds good to them stick with it. Nelson Pass of Passlabs and Firstwatt whose is considered one of the giants of audio circuitry rebuilt the crossovers in a pair of JBL L300's and considered L-pads on his mids and highs good enough. His thought on L-pads vs resistors in his rebuild.

 

Many people look askance at L-pads. This is understandable, and the wipers are certain to get noisy over the years. Some people think they just sound lousy anyway, but I did in fact use them satisfactorily. If you want to do better, here is an alternative network using resistors tapped to discrete levels:

This network was tested, and the performance differences between it and the L pad were minimal, so you can use it without compromise. Just attached the driver's wire to tap points from Low to High as you like. The resistors can be 3 watts or more, and once again, 5% is plenty of precision. The sound didn't suffer for using wire-wound types, but if you want to pay more for non-inductive resistors you should feel free.

 

Page 7 of the article. Fantastic read from the master of audio and what it entails designing a crossover network with attenuation.   http://www.firstwatt.com/pdf/art_l300.pdf  And this was not old school philosophy.  

[rjp]

1 hour ago, pzannucci said:

 A single 1 ohm or so resistor is not a complicated change, tied between the posts on the back of the speaker and can be used for minor changes...

 

...As long as you are happy, that's what counts though to answer your question, "is it better", it depends on if you are trying to accomplish a large change or not.  I found a good 1 ohm resistor tied between the terminals did take some of the edge off.  

 

True, it already works fine with the Lpad, but I am eager to learn more about your idea none the less.

 

Clearly Klipsch feels that a single resistor is the way to reduce the tweeter volume as I believe that is the purpose of the 4 Ohm resistor in the schematic, correct? I think that resistor is there for no other purpose than to attenuate the tweeter to better match the woofer volume. Or am I missing something here?

 

I thought you were suggesting to add an additional 1 Ohm to this existing 4 Ohm resistor (which seems like what Klipsch would have done), but from your description it sounds like you are saying to put the 1 Ohm resistor across the tweeter the input posts at the back of the speaker cabinet, or are you saying on the back of the tweeter itself? In any even it is a parallel resistor not series?

 

To make it clearer could you please tell me where in the posted schematic you would put the resistor?

 

Thanks.

[pzannucci]

Just simply take off the strap from the woofer and tweeter and add the resistor instead of the strap.  No, it isn't variable like a knob type lpad and will just do small changes, not large sweeping ones.

[rjp]

Oh, ok. Now I follow. You put the resistor between the woofer post and tweeter post instead of the jumper. Got it.

 

In that case, it would indeed be identical to replacing the existing 4 Ohm resistor with a 5 Ohm resistor in the schematic.

 

There are two things I like about this:

 

(1) It seems to be in keeping with the way Klipsh would have done it.

(2) It seems it would have less impact on the crossover frequency since as you mentioned there is no longer any possibility of mismatching the tweeter impedance at the output of the crossover as in the L-pad case.

 

 

[rjp]

I tried the series resistor method. I measure approximately 1 dB decrease in tweeter output for each 2 Ohms of added resistance.

 

Form my testing with the L-pad I like the level about 2-3 dB down, so I would need 4 to 6 ohms resistance.

[pzannucci]

I'm not sure if that much resistance will cause some variations but let us know the results and if they work well.  A lot of folks could use this tweek if they are not using some type of eq in an AVR.

[Guest wdecho]

Adding more resistance than 1 ohm is going to upset what Klipsch designed in the crossover but then you may like the result. Not so with an L-pad whether discrete resistors or pots, simple ohms law. In a L-pad circuit the resistor in series attenuates the signal and the one in parallel is there to retain the correct impedance with the driver the crossover sees. Research resistors in parallel for a better understanding of how the resistor in parallel with the driver lowers the impedance the crossover sees. With the resistance the same using the parallel resistor with the driver using the resistor in series from the crossover  to attenuate the signal the crossover network circuit designed by Klisch is not going to be changed. What you are doing is designing a different crossover by adding resistance in the crossover circuit. Think of the L-pad circuit, whether discrete or with a pot, as being part of the driver and not part of the crossover. With the resistance changed by the resistor in parallel with the driver when you add the resistance of the series resistor with the new resistance of the driver and parallel resistor the nominal impedance of the driver the crossover sees is not changed. Adding just a resistor in series without adding the resistor in parallel with the driver increases the resistance the crossover sees. Instead of say 8 ohms you will be adding more resistance if you add say 4 ohms for total 12 ohms the crossover is seeing. The crossover was designed to see 8 ohms not 12. 

 

Nothing wrong with playing but understand what you are doing in the process. Who knows you may like the result. 

[rjp]

Hi wdecho,

As mentioned on Page 1, I presently have L-pads installed just before the tweeter. I understand how they work, and that's why I chose them as my first experiment here. I am very pleased with the results and my conclusion is that L-pads work well in this application. I guess this should come as no surprise since this is what they were designed for.

 

However, I think pzannucci brought up an interesting technical point. For an L-pad to present a constant impedance to the source at any dial position it is necessary that the design impedance of the L-pad match the impedance of the load. Since the load in this case is a speaker, it's impedance is not purely resistive and varies with frequency. This means the L-pad can never be a perfect match at all frequencies (though it may be close enough that it makes no difference that we can hear).

 

(Note: I measured the impedance of my RP-260 tweeters, and the equivalent series resistance (ESR) varied between 8 and 16 Ohms over the frequency range 1KHz to 10KHz. No single L-pad can match both extremes)

 

I'm not sure what the implications of this mismatch might be but it does seem that the introduction of an L-pad at the tweeter must change the impedance the crossover sees at least at some frequencies. So although it works great, I think it is fair to say that the crossover is not seeing the same load Klipsch designed it to see. Does it matter? I have no idea.

 

The series resistor in discussion is before the crossover, so it will not change the load the crossover sees, but will  change the HF response of the crossover itself (kind of like having higher resistance speaker cables). So it is also not a perfect solution but for a different reason. I'm not sure how to analyze this. Maybe someone good with PSpice could give it a try?

 

In any event, both methods seem to work and sound about the same to my ears, but I think single Kleenex tissue sounds even better as it seems to reduce mid treble and leave the high stuff unaltered. But tissues look bad in front of my speakers