Could I get some help here?

[Ken Pachkowsky]

I just picked up a Klark Tekniks DN800 stereo 4 way xover. This has 6 xover cards that can be modified by the user in several different ways. Its pretty high quality like all Klark Tekniks products and has some great features.

Frequency card component table

Fr Cap Res

50Hz 330nF 6.8k
63Hz 220nF 8.2k
80Hz 330nF 4.3k
100Hz 100nF 11k
125Hz 100nF 9.1k
160Hz 47nF 15k
200Hz 100nF 5.6k
250Hz 22nF 20k
315Hz 33nF 11k
400Hz 68nF 4.3k
500Hz 33nF 6.8k
630Hz 22nF 8.2k
800Hz 33nF 4.3k
1000Hz 10nF 11k
1250Hz 10nF 9.1k
1600Hz 4.7nF 15k
2000Hz 10nF 5.6k
2500Hz 2.2nF 20k
3150Hz 3.3nF 11k
4000Hz 6.8nF 4.3k
5000Hz 3.3nF 6.8k
6300Hz 2.2nF 8.2k
8000Hz 3.3nF 4.3k
10000Hz 1nF 11k
12500Hz 1nF 9.1k
16000Hz 1nF 6.8k
20000Hz 1nF 5.6k

Having this as a reference, can one calculate whats required for 4.5K?

I want the crossover points to be 250/1K/4.5K

Can someone direct me to the right make and source for the electronics?

Thanks all

[Ken Pachkowsky]

Very nice features...

[Robh3606]

Hello Ken

Looks to me like you need to drop the manufacturer an email. There is nothing in the manual to show you how to derive the component values. I don't see a key on what components to change on the cards or how to select the crossover slopes/configure the cards. Do you need cards for each slope??There has to be more documentation somewhere on their site.

http://www.klarkteknik.com/index.htm

Hope this helps

Rob:)

[Oldmics]

Strictly a fast guess , 4.7 cap and 5.7 resistor for a 4.6K crossover point.

Most of those crossover points are determined from the availiability of hardware supporting the math.In other words some of those points do not corrolate to the hardware used to achieve those documented points.They are close enough to consider for the fiters used in equalisation.

Also try E.V. support,who now owns K.T. at 1-800-392-3497

Oldmics

[GordonW]

Yeah, if you plug and chug the math, it looks like a simple product:

cap value * resistor value * frequency = constant (in this case, about 1150, it varies a bit between 1100 and 1190, with most values between 1130 and 1175 for the pairs given that I checked).

The values Oldmics gave should work, though by the math above is indicated for a crossover point of about 4400 Hz (only about a 2.2 percent error, can't imagine that causing a problem!). Here's some other value pairs, that should also get right at 4500 Hz:

3.9 nf cap and 6.8K ohm resistor
6.0 nf cap and 4.3K ohm resistor

Regards,
Gordon.

[Ken Pachkowsky]

Thanks to those that have replied. I heard from KT this am. and was informed that they are sending a software program that calclates cap and resistor values depending on desired xover points.

Will keep you informed.

[boputnam]

Hey, Ken...

Looks like oldmics and Gordo have you sorted.

But, I'm wondering why the switch from the Westlake crossover - is this another "taste-test" you're so oft doing? Dammit - I think you get to play more than any of us!!

If you're really in the mood for something, and can get one, I'd urge you toss an Ashly into the mix - their XR4001 is just the ticket for your application, and it has those winkie "Response" curve adjustments both Gordon and I have posted (elsewhere...) about - where you can subtly change the slope at the crossover point(s) (without changing the 24dB/oct slope). They give some spectacular results.

I know the Widget, Ian and myself were mighty impressed with the XR1001 two-way model we played with - Widget went-out and got one! :yes:

[Ken Pachkowsky]

Originally posted by boputnam
Hey, Ken...

But, I'm wondering why the switch from the Westlake crossover - is this another "taste-test" you're so oft doing? Dammit - I think you get to play more than any of us!!


If you're really in the mood for something, and can get one, I'd urge you toss an Ashly into the mix - their XR4001 is just the ticket for your application.



Hi Bo

The Westlake crossover works well. I picked this up to do exactly that....experiment. If I don't like it, thats what Ebay is for.

I have looked at the Ashly. I have also heard some great reports on it. Perhaps I should try one of those as well.

Here is a drawing of the monitor stands I am building for the Westlakes. All my equipment will be rack mounted underneath. The wheels will make moving these 430lb beasts much easier.

Later Bud

[boputnam]

Ha! I guessed as much... :rotfl:

That double-wide rack looks sweet. A few houses with built-in consoles use that to stack EQ's and FX's beneath the desk. There might be something commercially available, but not to hold 430lbs...!!

[boputnam]

Originally posted by Giskard
Quad amp those 4345's! Hmm... I think our little road rack could do just that... :hmm:

[4313B]

Yep. There you go!

[Oldmics]

Hello All

That K.T. piece allows bandpass overlap.Thats a mighty nice feature.Overlaping points and choosing individual slope patterns per bandwidth is one of the procedures used for hi power pro applications.
I know that it also allows different slope combinations along with a choice of Bessel, L/R or Mrs.Butterworths responses.

These parameter changes became instantly accessable thru digital tecnology as opposed to replacing the required hardware to do these chores in anologue circuit designs.

The drawback of digital equipment is limitations or sonic charactoristics induced by the converter mechanisms.Some converters just sound better than others.It depends on your preference for the application.

Oldmics

[4313B]

"The drawback of digital equipment is limitations or sonic charactoristics induced by the converter mechanisms.Some converters just sound better than others."

Well, I've found that if you put the treadmill rats running the D converters facing the treadmill rats running the A converters instead of their backs to each other, or even side by side, facing together or away, the sound is much better! I wish I had a bar napkin to draw it on but it's not 4 pm yet. :(

[Oldmics]

Giskard

I gotta disagree with you here!!!!!!!

Did you forget about the opposing electro magnet factor generated from rat fur and being proprogated thru their tails?

Side by side or backs to each other would be a shorter transmission path of the distortion of rat fur EMI as opposed to your suggestion.

How about we meet at the lounge around 4:00 for MEK shooters and a more in depth discussion?

Best regards
Oldmics

[4313B]

I never thought not to shave my rats!

Okie Dokie then! MEK shooters at 4 pm it is! :p

[Ken Pachkowsky]

For those interested here are some photo's. Should be an interesting journey.

This is the Main Board with 4 Filter Boards and 2 Limiter Boards already installed.

[Ken Pachkowsky]

This is one of the filter boards removed.

[Ken Pachkowsky]

Last post. One side close up.

[Ken Pachkowsky]

Filter Board

[Ken Pachkowsky]

I just sent you the other files I received from KT

Its what we needed. Oldmics is workin on it as well. Be interesting to see if you both come up with the same solution. In any case I appreciate the help. I am useless with schematics and although I have a very basic understanding of slopes etc. this stuff is way beyond my comfort level.

[Ken Pachkowsky]

Well, ya can teach an old dog new tricks. I spent the evening learning how to calculate resistor values and identify them by there colored bands.

[Ken Pachkowsky]

FYI

I purchased a complete soldering/desoldering system (Pace) on Ebay and have also gathered a PCB vice and third hand for soldering components onto PCB's.

I spent a couple of evenings practicing on old computer circuit boards. I removed and re installed all types of caps and resistors until I was confident I would not render the KT a useless piece of junk.

The changes required removing and replacing 116 resistors and caps on the 4 circuit boards. Crossover points are now 250/1k/4.5k

The input stages on the main PCB have input filters that limit the input to 30Hz. I changed these to 15Hz by replacing the .33uf caps with .68uf musicaps. I expected the difference to be negligible. I was wrong. It’s really quite astounding, especially at low volumes. These 2235H’s are really something.

One difficult lesson was learned. After re-populating the boards and doing the first test The R channel board failed in the mid bass region. When the gain on the amp was turned up it caused a sound similar to the sound from a dirty pot. The scary part was the clipping indicators on that channel would light up if the knob was turned quickly (strange). After comparing the 2 boards for hours looking for broken traces, I came across an article warning that bending the leads to close to the body on Holco resistors caused the leads to separate from the end cap causing arcing. This is hard to see because of the plastic coating over the capacitor and end cap. I used ½ watt resistors and had to squeeze them into place to get them flat on the board.

Hmmmm! could this be my problem? A quick stripping and re populating of that section solved the problem. A good lesson learned!

Overall First Impressions of the DN800

It is extremely quiet.

Has a somewhat brighter sounding midrange than the Westlake HRX. (could be the caps need breaking in)

The imaging is not quite as good as the HRX (perhaps it will improve)

Don’t get me wrong, this is a fine sounding crossover but I will endeavor to continue my quest for nirvana. Could be nirvana is currently being designed in the land down under?

In closing let me say I was very nervous about taking a soldering iron to a high end piece of gear. Get the proper tools and start smoking! Making those shiny little pin drop puddles is great fun and quite relaxing.

Ken

[Robh3606]

Thanks for the update Ken. The tools make all the difference especially when soldering. How are those 2235 set up in the Westerlakes?? Then in a common enclosure or seperate?? When I went to listen to the 4350's they could rattle your fillings loose!!

Rob :)