A55g and need for an HF crossover point mod?

[Alexander]

 

I have klf-30's with the a55g mids. Having read that these are done by ~6300Hz.  But the crossover goes to 7000Hz on these klf-30's. It has been said that the k75/79k can go down to 6000Hz and the DE10/MAHL 2000Hz so the question is about the 700Hz “hole” that exists. This leads me to ask what component values can we change in the oem HF/MF xo to bring down the HF crossover point to say 6000Hz? The idea should also apply to models that use the k75/79k tweeter when adding the a55g driver.

 

 

 

[CANT]

7kHz is the approximate acoustical crossover point between the original midrange driver and the tweeter... neither came to a dead stop at that frequency? So despite the fact that the A55G doesn't go as high as the K52 there likely isn't a gap in response but there is no way to tell without taking a measurement. The K75/79 is generally spec'd to cross between 5-7kHz with little to no change in the tweeter circuit (i.e. Chorus I vs Chorus II). Also, the K75/79 should be good down to at least 2kHz if you wanted to change the crossover point for some reason? You would need add a low pass to the midrange as well as adjust the high pass for the tweeter though.

 

If it were me I would just put the K52 back in?

 

[jjptkd]

8 hours ago, Alexander said:

 so the question is about the 700Hz “hole” that exists. 

There really isn't much of if at all a hole in response as the tweeter has a very gradual roll off 7khz and down also the A55-G doesn't stop completely at 6300hz it just starts to drop off there pretty fast but still has output so there is overlap between the two drivers you kind of want them both to drop to the point when they intersect they're roughly half volume each or so to keep the response even across the band if that makes sense. 

[jjptkd]

1 hour ago, CANT said:

 

If it were me I would just put the K52 back in?

 

I have not tried the A55-g in the klf30 but that driver works very well in the Chorus II and is a far superior driver than the stock k52. 

 

 

[SWL]

I have that setup in my 30's and the hole sounds pretty damn good.

Sent from my SM-G950U using Tapatalk

[CANT]

2 hours ago, jjptkd said:

I have not tried the A55-g in the klf30 but that driver works very well in the Chorus II and is a far superior driver than the stock k52. 

 

 

The A55G didn't exist when I owned my KLF20's but I did throw a pair K55's in there just for sh!ts and giggles... it was different but I would not say it sounded or looked better.

 

The thing to me is that the 55 and 52 excel at completely different things and I would not consider them interchangeable. For a Khorn, La Scala, Belle or maybe older Cornwalls I get it. There is a reason the K55 is used and I could understand the argument that an A55 might be better, though I don't buy it. For models that naturally come with the K52 I think the benefits of either 55 are lost on that model due to their inherent differences. 

 

And I can completely understand someone saying the A55 sounds better to them but adjusting the bass/treble might also sound better to that same person... so what does better mean? 

[jjptkd]

15 hours ago, CANT said:

And I can completely understand someone saying the A55 sounds better to them but adjusting the bass/treble might also sound better to that same person... so what does better mean? 

 

Yes, I agree and admit I'm really no expert here I like to tinker and try new things mostly OEM Klipsch parts swapping from one model to the next as I like to keep things close to authentic as possible in most cases. I share my experiences here not based on math or measurements but by ear from my easy chair and I fully understand that things like my room, things in my room, the equipment I use to power my speakers, the source and quality of my music, type of music I'm listening to, how loud I have it turned up or down how many beers I've had or even the particular mood I'm in at the time all can have a dramatic influence on how I perceive the sum of what is in front of me and how much I enjoy it and even then I know its only for that distinct period in time. 

 

When it comes to the k--52 vs the k-55 from what I've been told, from what I've read and just knowing how when and where these were used I believe them  to be somewhat interchangeable, look no further than the Cornwall & Heresy 1.5. The k-52 was produced to be a drop in replacement for the k-55 in these two speaker models and then were adapted for use in later models. There are some slight differences I believe due to the size of the drivers in frequency response and sensitivity but they aren't that far off, impedance is roughly the same.

 

Now I know the experts will step in and tell me why I'm wrong and I encourage that thanks..

[Alexander]

 

Yes the a55g most certainly sounds better than the k52h. But it begs the question with the known crossover point of 7000Hz and also known that the a55g signs off around 6300Hz that there might be something to be gained by adjusting the crossover point?

 

I would venture to say if this scenario were with say a set of LaScala's or Khorns the reply would not be to put the old driver back in. Or would it?

 

[MechEngVic]

On 3/22/2020 at 2:41 PM, Alexander said:

 

Yes the a55g most certainly sounds better than the k52h. But it begs the question with the known crossover point of 7000Hz and also known that the a55g signs off around 6300Hz that there might be something to be gained by adjusting the crossover point?

 

I would venture to say if this scenario were with say a set of LaScala's or Khorns the reply would not be to put the old driver back in. Or would it?

 

A driver is going to respond to a full bandwidth signal in a particular way based on how it is built. Tuning the signal doesn't really change the "sound signature" of the driver, rather, a tuned signal changes the "volume" of specific parts of the signal that the driver will see. The driver will tend to keep its sound signature, that's why the a55g will sound different (many say better) than the k52 no matter what changes you make. Now, if the a55g begins to roll off at 6300hz, you'd still have a lot of information coming through at 7000hz with a full signal. What you need to find out is how much "volume" the KLF-30 crossover is giving the driver at 7000hz. It may be less or more than you want since the KLF-30 crossover was meant to tune the k52. I would model the drivers and crossover in X-Sim (or similar software) and compare the crossover's frequency response curves using the same crossover values but switching the mid driver FR and Z between the a55 and k52. It won't give you precision but it will show you how the crossover treats each mid driver. If the graph shows more dip at 7000hz with the a55 then with the k52, you may be able to make changes to either the tweeter or the mid part of the crossover.

[Alexander]

9 hours ago, MechEngVic said:

A driver is going to respond to a full bandwidth signal in a particular way based on how it is built. Tuning the signal doesn't really change the "sound signature" of the driver, rather, a tuned signal changes the "volume" of specific parts of the signal that the driver will see. The driver will tend to keep its sound signature, that's why the a55g will sound different (many say better) than the k52 no matter what changes you make. Now, if the a55g begins to roll off at 6300hz, you'd still have a lot of information coming through at 7000hz with a full signal. What you need to find out is how much "volume" the KLF-30 crossover is giving the driver at 7000hz. It may be less or more than you want since the KLF-30 crossover was meant to tune the k52. I would model the drivers and crossover in X-Sim (or similar software) and compare the crossover's frequency response curves using the same crossover values but switching the mid driver FR and Z between the a55 and k52. It won't give you precision but it will show you how the crossover treats each mid driver. If the graph shows more dip at 7000hz with the a55 then with the k52, you may be able to make changes to either the tweeter or the mid part of the crossover.

 

 

Yes, that is were I am headed. On the chance that you have read the recent past of posts I have made about modeling autotransformers, changing taps, asking for .xma and .frd files & calibrated microphones recommendations. And lets not forget the measuring .xma mass method on closed back drivers

 

At this point in time I have not figured out how to (properly & accurately) have an auto transformer in either LT Spice or x Sim. circuits I am sure there will be a way I just haven't found out how yet but admittedly I have barely used x Sim. BTW can x Sim change the backgrounds to black? It is very hard for me to see things on a white background.

 

 

 

 

 

[mboxler]

2 hours ago, Alexander said:

 

 

 

At this point in time I have not figured out how to (properly & accurately) have an auto transformer in either LT Spice..

 

 

 

 

 

 

I would be happy to send you my version of a 12 tap auto transformer.  It seems to work pretty well.  It's currently part of a Type A simulated circuit.  I assume I can attach an .asc file in a private message.

 

Mike

[MechEngVic]

14 hours ago, Alexander said:

 

 

Yes, that is were I am headed. On the chance that you have read the recent past of posts I have made about modeling autotransformers, changing taps, asking for .xma and .frd files & calibrated microphones recommendations. And lets not forget the measuring .xma mass method on closed back drivers

 

At this point in time I have not figured out how to (properly & accurately) have an auto transformer in either LT Spice or x Sim. circuits I am sure there will be a way I just haven't found out how yet but admittedly I have barely used x Sim. BTW can x Sim change the backgrounds to black? It is very hard for me to see things on a white background.

 

 

 

 

 

I don't know about changing the color theme of X-Sim, I'd like it dark too! I'll check next time I get on. As far as the autotransformer goes, they are similar to an inductance divider but with much different impedance behavior. As of now I don't know how to model one. 

[glens]

A transformer is a transformer, whether it has isolated windings or not, I say; and that there's no need to model them as separate entities in a crossover.  Simply use the impedance as seen through them as / when necessary and you'll be golden.

[jjptkd]

On ‎3‎/‎22‎/‎2020 at 2:41 PM, Alexander said:

Yes the a55g most certainly sounds better than the k52h. But it begs the question with the known crossover point of 7000Hz and also known that the a55g signs off around 6300Hz that there might be something to be gained by adjusting the crossover point?

 

Charts of the output response of the K-55's vs the k-52 done some time back

 

[pzannucci]

1 hour ago, glens said:

A transformer is a transformer, whether it has isolated windings or not, I say; and that there's no need to model them as separate entities in a crossover.  Simply use the impedance as seen through them as / when necessary and you'll be golden.

Exactly.

 

Also the KLF does not have an upper crossover cut off on the mid.  It naturally rolls off.  Two things to make sure of, match the output of the A55 to the original KLF 30 mid (I have a feeling they are pretty close though don't know the impedance matches between the two which imposes changes on the mid components).  The tweeter can have  a slightly larger first cap if you can to add a slight lowering of the crossover point.  Likely though, you won't likely hear the hole though may be shown in measurements.

[Alexander]

6 hours ago, glens said:

A transformer is a transformer, whether it has isolated windings or not, I say; and that there's no need to model them as separate entities in a crossover.  Simply use the impedance as seen through them as / when necessary and you'll be golden.

OK, figured out how to create a transformer in LT Splice and I think we would need to look at it as a step down since we are cutting voltage to the driver. Here is the YouTube video I followed: https://www.youtube.com/watch?v=mzwuJ0y8vuw&t=253s

 

So if I am correct then we need to place values in henrys (mH in our case) and the primary (mH) to secondary windening ratio is squared and that value will be used on the secondary side of the model. If I am right then where can we find the induction values of the primary side? Would we actually measure that?
 
I have measured an unmarked auto transformer that I was told was a t7a (-10db) and here is what I got. 0-5 (primary yes?) is 105.3mH/2.2ohm & 2-5 (secondary yes?) is 49.5mH/.89ohm. I did a third measurement of 0-2 @ 8.2mH/.22ohm but did not think that info is used. Pure DC resistance values were 131.3, 61.4 & 10.6ohms. All measurements were done using DATS v3.

[Alexander]

On 3/26/2020 at 7:49 AM, jjptkd said:

 

Charts of the output response of the K-55's vs the k-52 done some time back

 

Thanks for the link, I d/l'd the images to give them a look see.

 

[Alexander]

4 hours ago, pzannucci said:

Exactly.

 

Also the KLF does not have an upper crossover cut off on the mid.  It naturally rolls off.  Two things to make sure of, match the output of the A55 to the original KLF 30 mid (I have a feeling they are pretty close though don't know the impedance matches between the two which imposes changes on the mid components).  The tweeter can have  a slightly larger first cap if you can to add a slight lowering of the crossover point.  Likely though, you won't likely hear the hole though may be shown in measurements.

 

The plan is to do actual measurements with a to-be-ordered calibrated mic via REW and then we will see if there is any need to consider changing anything.

 

As I understand it the a55g was made as an optional k55 replacement and the impedance is similar, also the k52h is also, we will measure both drivers so we can compare but we do not have any k55 sames on hand.

[pzannucci]

4 minutes ago, Alexander said:

 

The plan is to do actual measurements with a to-be-ordered calibrated mic via REW and then we will see if there is any need to consider changing anything.

 

As I understand it the a55g was made as an optional k55 replacement and the impedance is similar, also the k52h is also, we will measure both drivers so we can compare but we do not have any k55 sames on hand.

Good, you have me interested.  I've been thinking about the mid in my KLF-30s.  Not sure my old ears would be able to tell too much but none the less, still debating the project.

[Alexander]

28 minutes ago, pzannucci said:

Good, you have me interested.  I've been thinking about the mid in my KLF-30s.  Not sure my old ears would be able to tell too much but none the less, still debating the project.

 

I will tell you when I changed out the k52h to the a55g's it was immediately apparent of the change for the better. I would not hesitate to recommend them to any klf-20/30 owners if they could afford them.

I am 62 so I am sure I have lost some hearing too. I use to hear the 17.5Hz from the tripler on CRT TV's, that would drive me nuts.