KLF-10 HEIGHT, STANDS OR TILT?

[MechEngVic]

Well it's been a couple of months but I finally got around to building stands.

 

I tried tilting the speakers and it did help some but it was a trade-off, mid-range improved but it high frequencies were off in the distance and not as present as they should be.  Plus tilting adds another dimension to finding the sweet spot. It's bad enough having to adjust toe-in, distance apart, and distance away to also have to get the distance where the tilt angle height is right. Too close and the sweet spot's still below you, too far and it's above your head.

 

I bought a 4x4 and a 2x6 and built 6" stands that incorporated the "corner post" style of the stands moray james posted, along with the open box style that others suggested would help maintain bass coupling. I am pleased to report I got the sound I was looking for. 

 

If there was a loss or change of bass output it was completely eclipsed by the amazing improvement in imaging and mid-range punch that putting the speakers closer to ear level made. Bass is still strong and solid and appears to have tightened up but I think that's due to the woofers being higher. I can feel it in my chest more now.

 

Keep in mind that the woofers of the KLF-10's are responsible for frequencies below 2300Hz, that's a lot of direction-sensitive sound sitting around your knees, so getting them closer to ear height makes the mids and upper bass bloom. If you're trying to tame bright horns, getting the horn just above ear level, and bringing the lower frequency producing parts of the speaker closer to the ear may help.

Edited August 30, 2019 by MechEngVic

[glens]

11 minutes ago, MechEngVic said:

I bought a 4x4 and a 2x6 and built 6" stands that incorporated the "corner post" style of the stands moray james posted, along with the open box style that others suggested would help maintain bass coupling. I am pleased to report I got the sound I was looking for. 

 

The last part of that quote is unambiguous (congrats!) but I'm having a little trouble correlating the "corner post style" with the "open box style."  Could you elaborate, or maybe even prove you actually did it™ with a pic or two?  :^)

[Maz4bz]

Nice stands. They turned out really great. You have some wood working skills for sure.  

 

I would imagine that your KLF10 sound wonderful as I adore my KG5.5's which are the KLF10's predecessor if I'm not mistaken, they do look similar with their dual 10" woofers and tweeter. As I sit in my lounge with my ears at tweeter height I've not lifted mine.

 

I think you may have found that the double woofer, two way arrangement of these speakers has a very narrow vertical forward lobe, like I have.  This means that when I move above or below the tweeter much, the midrange suffers.

 

When I measure the KG's in two way configuration vs an all new 2.5 way crossover I've designed for this system there is a big difference in their off axis performance around the crossover region. Measured upwards 25 degrees above for 2.5 way (black) vs 20 degrees above for the 2 way crossover (red) ....

 

 

Here is the downward measurement both at 20 degrees (again black = 2.5 way, red = 2 way)….

 

 

So I think you hit the nail on the head when you say your speakers bloom when you're on axis with the tweeter. This is also why I think our speakers deserve a 2.5 way crossover, rather than the 2way that they have.  

 

I thought I'd plug the KLF10 crossover that Moray posted a while back into XSim and load it up using the individual KG 5.5 driver frequency response measurements to see where your crossover is. I'd call this a 1500hz crossover where (green) both woofers wired in parallel and (red) tweeter frequency responses cross over:

 

Now this is a bit academic as we are not using the individual measured frequency response of your KLF10 drivers here, but no matter, it'll likely be close but certainly not perfect!  The KLF 10 crossover schematic is almost identical to the KG 5.5. The KLF woofers are a different model number, however the tweeter seems to use the same diaphragm.  

 

Notwithstanding if your drivers have similar sensitivity to the KG's then your tweeter will be around 5db or so hot like it is in the graph above. If this is the case then your KLF's can be sooooo much better!

 

So I notice you mention the tweeter is a bit hot, same for the KG 5.5's. Here is a little suggestion I've modelled for you. Similar to what I did on my KG journey.  

 

Here is the original KLF10 crossover:

I would like to propose a single change to the tweeter (S1) circuit with the addition of a resistor R2 of 10 ohms:

 

This will result in the following change in output to the tweeter, grey is original and blue is new:

 

 

This flattens the tweeter to be more in balance with the woofers and should be a BIG improvement. It was for the KG's.  

 

Of course we are not working with the measured response of your KLF10 drivers so you can adjust R2 to taste, there is no right or wrong here, as Troels Gravesen says:

 

"Some call the voicing of speakers an art. I don't think so. Voicing a speaker is a matter of taste like adding spices to a stew. Some like it hot, creamy or crunchy - some don't."

 

So here is the same crossover with a 6.2 ohms R2 for a more subdued top end:

 

 

Or here with a 13 ohms R2 if either of the above took out too much sparkle:

 

 

What I am proposing here is to experiment with the value of R2 until you enjoy what you hear.  Resistors are really cheap. Buy a handful and have some fun experimenting. This change is completely reversible if you don't like it.

 

It will take some time to acclimate to the changes, give it time and if you start to think too dull or too bright, tweak R2 some more. It took me some weeks of listening to adjust to the Lpads that I originally added to the KG's. I incrementally changed these until my speakers got too dark/dull. Then I retreated a little to regain some top end sparkle. What I settled on ended up being just about ruler flat with no measuring equipment. Trust your ears!

 

If you are the scientific kind however, then a cheap USB mic and some free software is like being able to see for the first time. I highly recommend the investment. Wish I'd done it earlier.  I use the Dayton Audio Omnimic system, it is a terrific tool and so easy to use.

 

And don't worry about how low you can go in terms of ohms with R2. As an example if you set R2 to just 1 ohms, which I really doubt you'd ever do as it cuts about 15db output from the tweeter and would sound very dark,  the system impedance never dips below 4 ohms.

 

Here is the original crossover in grey compared to R2 @ 1 ohms in blue:

 

 

Hope this helps and may your terrific Klipsch provide you many years of enjoyment - they are undoubtedly very fine speakers.

 

Cheers  

 

 

[MechEngVic]

14 hours ago, Maz4bz said:

Nice stands. They turned out really great. You have some wood working skills for sure.  

 

I would imagine that your KLF10 sound wonderful as I adore my KG5.5's which are the KLF10's predecessor if I'm not mistaken, they do look similar with their dual 10" woofers and tweeter. As I sit in my lounge with my ears at tweeter height I've not lifted mine.

 

I think you may have found that the double woofer, two way arrangement of these speakers has a very narrow vertical forward lobe, like I have.  This means that when I move above or below the tweeter much, the midrange suffers.

 

When I measure the KG's in two way configuration vs an all new 2.5 way crossover I've designed for this system there is a big difference in their off axis performance around the crossover region. Measured upwards 25 degrees above for 2.5 way (black) vs 20 degrees above for the 2 way crossover (red) ....

 

 

Here is the downward measurement both at 20 degrees (again black = 2.5 way, red = 2 way)….

 

 

So I think you hit the nail on the head when you say your speakers bloom when you're on axis with the tweeter. This is also why I think our speakers deserve a 2.5 way crossover, rather than the 2way that they have.  

 

I thought I'd plug the KLF10 crossover that Moray posted a while back into XSim and load it up using the individual KG 5.5 driver frequency response measurements to see where your crossover is. I'd call this a 1500hz crossover where (green) both woofers wired in parallel and (red) tweeter frequency responses cross over:

 

Now this is a bit academic as we are not using the individual measured frequency response of your KLF10 drivers here, but no matter, it'll likely be close but certainly not perfect!  The KLF 10 crossover schematic is almost identical to the KG 5.5. The KLF woofers are a different model number, however the tweeter seems to use the same diaphragm.  

 

Notwithstanding if your drivers have similar sensitivity to the KG's then your tweeter will be around 5db or so hot like it is in the graph above. If this is the case then your KLF's can be sooooo much better!

 

So I notice you mention the tweeter is a bit hot, same for the KG 5.5's. Here is a little suggestion I've modelled for you. Similar to what I did on my KG journey.  

 

Here is the original KLF10 crossover:

I would like to propose a single change to the tweeter (S1) circuit with the addition of a resistor R2 of 10 ohms:

 

This will result in the following change in output to the tweeter, grey is original and blue is new:

 

 

This flattens the tweeter to be more in balance with the woofers and should be a BIG improvement. It was for the KG's.  

 

Of course we are not working with the measured response of your KLF10 drivers so you can adjust R2 to taste, there is no right or wrong here, as Troels Gravesen says:

 

"Some call the voicing of speakers an art. I don't think so. Voicing a speaker is a matter of taste like adding spices to a stew. Some like it hot, creamy or crunchy - some don't."

 

So here is the same crossover with a 6.2 ohms R2 for a more subdued top end:

 

 

Or here with a 13 ohms R2 if either of the above took out too much sparkle:

 

 

What I am proposing here is to experiment with the value of R2 until you enjoy what you hear.  Resistors are really cheap. Buy a handful and have some fun experimenting. This change is completely reversible if you don't like it.

 

It will take some time to acclimate to the changes, give it time and if you start to think too dull or too bright, tweak R2 some more. It took me some weeks of listening to adjust to the Lpads that I originally added to the KG's. I incrementally changed these until my speakers got too dark/dull. Then I retreated a little to regain some top end sparkle. What I settled on ended up being just about ruler flat with no measuring equipment. Trust your ears!

 

If you are the scientific kind however, then a cheap USB mic and some free software is like being able to see for the first time. I highly recommend the investment. Wish I'd done it earlier.  I use the Dayton Audio Omnimic system, it is a terrific tool and so easy to use.

 

And don't worry about how low you can go in terms of ohms with R2. As an example if you set R2 to just 1 ohms, which I really doubt you'd ever do as it cuts about 15db output from the tweeter and would sound very dark,  the system impedance never dips below 4 ohms.

 

Here is the original crossover in grey compared to R2 @ 1 ohms in blue:

 

 

Hope this helps and may your terrific Klipsch provide you many years of enjoyment - they are undoubtedly very fine speakers.

 

Cheers  

 

 

WOW! I'm gonna have to spend some time going over your post. 

 

I do have the Crite's titanium diaphragms and custom upgraded crossovers, but using warm tubes & PIO caps in addition to height and placement adjustments really helped tame the high end. It's way more "S", "T", "P" and "H/CH" articulate. But I am certainly not opposed to revisiting my crossovers...

 

[MechEngVic]

On 8/30/2019 at 12:53 PM, glens said:

 

The last part of that quote is unambiguous (congrats!) but I'm having a little trouble correlating the "corner post style" with the "open box style."  Could you elaborate, or maybe even prove you actually did it™ with a pic or two?  :^)

I hope you can see the pics now.

 

In earlier comments on this thread, moray james posted an image of stands with a corner post design which I thought looked good but others commented that without solid sides around the stands, they would lose bass coupling to the floor. By "open box" style I meant the four sided base with open top and bottom like the bases on Chorus and forte speakers. I sort of combined the two.

[glens]

Yeah, I see that you've added the pics to your post, which weren't there when I'd posed the question.  Makes me look rather foolish now - thanks!

 

I use an 1/8" radius round-over bit, too, on a lot of stuff like that.

 

 

[Maz4bz]

I should have added the XSim .dxo file I used to do the sims so you can have a play for yourself.

 

Just head over to the XSim thread at diyaudio.com and grab the latest version from there.

 

Please do let us know how you get on. If you need any help just sing out! 

KLF10.dxo

[MechEngVic]

On 9/2/2019 at 6:46 AM, Maz4bz said:

I should have added the XSim .dxo file I used to do the sims so you can have a play for yourself.

 

Just head over to the XSim thread at diyaudio.com and grab the latest version from there.

 

Please do let us know how you get on. If you need any help just sing out! 

KLF10.dxo 113.59 kB · 2 downloads

A million thanks!! I've been needing this. I wish I would've had it before I built my custom crossovers. Since the 1.75uf cap is not commercially available I went with a 1.8uf, and now with the Xsim I realize I should have gone with a smaller not larger value. I will make the adjustment. Also, I have 10ohm resistors on the way. I think your resistor fix is just what I've been looking for. I'll let you know how it goes.

[Denman53]

Hello...my name is Denman....I'm new to this web site and have been struggling to navigate and learn how to communicate....I did find the talk about the KLF-10s....and i am needing one of the plastic carpet spike covers....is there a possibility someone my know how to find one of these....THANKS denman

[MechEngVic]

50 minutes ago, Denman53 said:

Hello...my name is Denman....I'm new to this web site and have been struggling to navigate and learn how to communicate....I did find the talk about the KLF-10s....and i am needing one of the plastic carpet spike covers....is there a possibility someone my know how to find one of these....THANKS denman

Hello Denman,

Those covers will be hard to come by. A possible replacement is a set of 1" spike feet and 1" felt pads, they will do the trick.

 

https://smile.amazon.com/gp/product/B075WSC4HQ/ref=ppx_yo_dt_b_asin_title_o06_s00?ie=UTF8&psc=1

 

https://smile.amazon.com/gp/product/B07C2KK96B/ref=ppx_yo_dt_b_asin_title_o03_s01?ie=UTF8&psc=1

[Maz4bz]

Hi MechEngVic,

Glad you were able to have a play with the XSim .dxo it is a great tool.

 

11 hours ago, MechEngVic said:

Since the 1.75uf cap is not commercially available I went with a 1.8uf

 

I think you did the right thing here. 0.05uF is a very small difference in value and I would be highly surprised if that was audible.

 

11 hours ago, MechEngVic said:

I will make the adjustment.

 

I would hate it if you replaced that lovely looking Clarity Cap on the basis of what you have seen in the XSim .dxo I provided. The reason for this is that we are not using the KLF10 driver measurements for the combined system response. We would need these measurements to do any kind of granular change to the crossover, such as changing a cap by any value, especially a value of only 0.05uF, for example.

 

My intent in sharing the XSim model was simply to show how we might address the  gross difference in output between tweeter and woofers, if these have similar efficiency to the KG 5.5 drivers. The beauty of the R2 tweak I have suggested is that it works like a simple volume control on the tweeter level that you can absolutely hear the difference of when implemented into a system like the KG 5.5's which are approximately 5db hot on top. 5db is a big/gross change to the level of the tweeter that will definitely be audible. 

 

Notwithstanding I encourage you to change up your crossover however you like, its completely reversible and a fun way to learn. I would like to suggest that you can do that with nice cheap parts, like I have done in my KG thread, first. Then if you settle on a design as final, you can go nuts and spend big on parts then. Of course if money is no object please completely ignore me!  

 

If money is no object then you really owe it to yourself to get a measurement mic. Then you will know precisely the impact of any changes you make.

 

11 hours ago, MechEngVic said:

I have 10ohm resistors on the way. I think your resistor fix is just what I've been looking for. I'll let you know how it goes.

 

I will very much look forward to what you find.

 

I also highly recommend trying other values for R2, mainly as a learning opportunity, and because the 10ohms value was just a stab in the dark on my behalf - we are not using your KLF drivers in the model after all. If you try a range of values and give each change a while for you to acclimate to, I'm sure you will find a balance that suits your taste. If you then give us a "review" of each value it may help others to zero in on a value that works for them too. 

 

Have fun and good luck! 

[glens]

4 hours ago, Maz4bz said:

Notwithstanding I encourage you to change up your crossover however you like, its completely reversible and a fun way to learn. I would like to suggest that you can do that with nice cheap parts, like I have done in my KG thread, first. Then if you settle on a design as final, you can go nuts and spend big on parts then. Of course if money is no object please completely ignore me!  

 

Just be careful that you don't spend as much playing with passive stuff as you would getting a DSP and another amp.  No contest.

[MechEngVic]

17 hours ago, Maz4bz said:

Hi MechEngVic,

Glad you were able to have a play with the XSim .dxo it is a great tool.

 

 

I think you did the right thing here. 0.05uF is a very small difference in value and I would be highly surprised if that was audible.

 

 

I would hate it if you replaced that lovely looking Clarity Cap on the basis of what you have seen in the XSim .dxo I provided. The reason for this is that we are not using the KLF10 driver measurements for the combined system response. We would need these measurements to do any kind of granular change to the crossover, such as changing a cap by any value, especially a value of only 0.05uF, for example.

 

My intent in sharing the XSim model was simply to show how we might address the  gross difference in output between tweeter and woofers, if these have similar efficiency to the KG 5.5 drivers. The beauty of the R2 tweak I have suggested is that it works like a simple volume control on the tweeter level that you can absolutely hear the difference of when implemented into a system like the KG 5.5's which are approximately 5db hot on top. 5db is a big/gross change to the level of the tweeter that will definitely be audible. 

 

Notwithstanding I encourage you to change up your crossover however you like, its completely reversible and a fun way to learn. I would like to suggest that you can do that with nice cheap parts, like I have done in my KG thread, first. Then if you settle on a design as final, you can go nuts and spend big on parts then. Of course if money is no object please completely ignore me!  

 

If money is no object then you really owe it to yourself to get a measurement mic. Then you will know precisely the impact of any changes you make.

 

 

I will very much look forward to what you find.

 

I also highly recommend trying other values for R2, mainly as a learning opportunity, and because the 10ohms value was just a stab in the dark on my behalf - we are not using your KLF drivers in the model after all. If you try a range of values and give each change a while for you to acclimate to, I'm sure you will find a balance that suits your taste. If you then give us a "review" of each value it may help others to zero in on a value that works for them too. 

 

Have fun and good luck! 

 

You did it.

 

Your 10 ohm resistor fixes the KLF-10/KG5.5.

 

I can't believe a pair of 3.00$ resistors made more improvement than the Crite's titanium tweeter diaphragms, the Clarity CSA and Russian PIO caps, and the silver plated pure copper 14 gauge wiring. Proof positive that the greatest improvements in sound come from a properly designed circuit, and not from fancy boutique parts or exotic materials.

 

The resistors actually measured 12 ohms so it's not such a drastic change but boy does it make a difference! I don't hear the need to make any other changes to the crossover at this time. But I will definitely be putting in some time with the XSim.

 

I owe you big. Next time you're in Los Angeles I'm buying you a steak dinner! 

[DirtyErnie]

Doesn't that first 1.75uf cap set a 6db slope to compensate for the constant-directivity of the horn?  I see the same set-up on my CF2's, same horn I think.  Would it be possible to both decrease the volume a bit (3db) and extend the treble by moving that cap's break-point up a half octave?

[MechEngVic]

18 hours ago, DirtyErnie said:

Doesn't that first 1.75uf cap set a 6db slope to compensate for the constant-directivity of the horn?  I see the same set-up on my CF2's, same horn I think.  Would it be possible to both decrease the volume a bit (3db) and extend the treble by moving that cap's break-point up a half octave?

Your question is beyond me. Hopefully Maz4bz can chime in or maybe you can message him. How would you move a cap's break point? By changing the value? And if so, how do the values correspond to octaves? And please tell me about this constant-directivity of the horn compensation. I'm afraid I'm making you give knowledge instead of receiving it, hope you don't mind. I've been wondering what specific purposes the individual components serve.

[glens]

Caps don't really have "break points."

 

What they have is a decreasing reactance (impedance) as frequency rises.  They all follow a curve starting at "infinity" (or thereabouts) at 0 Hz (DC).  What one might call the "break point" is the place in the curve for any value of capacitance where the impedance of the cap matches that of the driver itself or as seen through a transformer if there is one (or any other resistance between the cap and driver), thus each consuming half the signal (the cap and whatever makes up the load as seen after the cap).

[glens]

22 hours ago, DirtyErnie said:

Doesn't that first 1.75uf cap set a 6db slope to compensate for the constant-directivity of the horn?  I see the same set-up on my CF2's, same horn I think.  Would it be possible to both decrease the volume a bit (3db) and extend the treble by moving that cap's break-point up a half octave?

 

No.

 

"Constant Directivity" means that energy at different frequencies throughout the working range stays within the confines of the thrown pattern.  In that case the balance between the lower and higher ends of the range is different than if the lower frequencies were allowed to spread our more (as they'd be inclined to do otherwise) and the higher frequencies were allowed to "beam" more (as they'd be inclined to do otherwise).  If a driver is not made in such a way as to compensate for that itself then "exterior" means must be employed.  If done passively (as being discussed) then one or more combination of further elements are employed to "equalize" (bleed off unwanted spectrum energy) the signal going to the driver.  This will be a block of specific-value L/C/R components separate from those of the "crossover filter elements."  I see none of that in any of the schematics above based on a cursory glance scrolling up the page.

[DirtyErnie]

You're right about what a constant directivity Horn does, but all the (manufacturers) literature and theory I've seen about the beasts says they need a 6db/octave power increase above about 5KHz to keep a flat frequency response across their wider directivity. I'm pretty sure (granted, it's from about an hour playing with Xsim) that the first 1.75uf cap sets that 6db/octave slope, as it's -3db break point with an 8-ohm driver is around 12khz. If that's the case, one could theoretically be able to take a -3db attenuation from the whole Tweeter Horn by moving that point a half octave higher, to around 18KHz.

[Maz4bz]

On ‎9‎/‎17‎/‎2019 at 7:01 AM, DirtyErnie said:

Doesn't that first 1.75uf cap set a 6db slope to compensate for the constant-directivity of the horn?

 

The tweeter has a third order or 18db high pass filter which is created by C1, C2 and L1 in combination.

 

On ‎8‎/‎31‎/‎2019 at 6:37 PM, Maz4bz said:

 

 

Horn compensation when a compression driver is used is a little different and is usually formed by a capacitor and resistor in parallel.  See here the Econowave crossover design by Zilch. In this design the compensation is formed by C3 & R1 for the Selenium D220 on the Econowave guide...

 

 

Wayne Parham's white paper "Speaker motors and passive crossover filters" explains this well, certainly better than I can. 

 

In this case the KLF/KG and many other Klipsch horns front a dome tweeter which may explain why these don't require compensation - but I could be wrong in saying this. As I understand horns it is the constant directivity that causes the 6db decline in output which is why I was surprised the KG horn performed so well without compensation.

 

On ‎9‎/‎17‎/‎2019 at 7:01 AM, DirtyErnie said:

Would it be possible to both decrease the volume a bit (3db) and extend the treble by moving that cap's break-point up a half octave?

 

The breakpoint or cutoff frequency or crossover frequency (Fc), if I'm correct in understanding your use of breakpoint can't really be changed without  requiring a complete redesign of the entire crossover. Altering C1 affects the filter Fc on the tweeter, which will impact on the summed response of the tweeter and woofer, thereby requiring a subsequent tweaking of the woofer circuit. As I mentioned previously in this thread any changes to the crossover filter design is unlikely to yield satisfactory results without the actual KLF-10 driver frequency and impedance measurements.

 

2 hours ago, DirtyErnie said:

they need a 6db/octave power increase above about 5KHz to keep a flat frequency response across their wider directivity

 

"Can anyone explain HF Compensation and the need for it?"

 

Why not have a play with your CF-3's and see if you like what you hear? You can reverse it if you don't like it.

 

Better yet get a measurement mic and see what you hear! 

 

[DirtyErnie]

There are many ways to skin a cat. Klipsch, being a large producer, is likely to aim for the one that costs less. Quite a complex crossover posted for the Econowave, CF2 is much simpler.