KLS3 Gold crossover - MkI and MkII - World-Designs-Forum

Brian · #1 24th July 2006, 10:33 PM

I've posted before about Greg and James having helped me in recent times with my KLS3 Gold MkII speakers, they are now better than they've ever been. The problems of the pasta of excessive bass has gone by changing the crossover to the standard Gold version, but after extensive listening now over a period of some weeks I'm still not happy. I think there is some bass power gone from the music that was good to have, I'm trying to get to something that is between the performance of the KLS3 as achieved with these 2 crossover designs. I want that extra bit of bass power, but without the boom.

Looking on the web' at almost every crossover I can find for a 3-way, I notice that the high pass section of the midrange has a capacitor on the positive side and usually an inductor in parallel. (Note: It's true to say that the KLS3 Gold Standard crossover doesn't have this cap/inductor, but according to the supplement #32, the design allows for the mid unit to just run out of steam at ~100Hz), so no components. The LF filter values are 12mH on the positive side and 120uF cap in parallel, which I assume sets the bass to stop at ~100Hz to fit in with the mid unit.

The KLS3 Gold MK II is supposed to crossover between LF/MF at ~400Hz, so the LF section has an inductor of 2.7mH on the positive side, and a 50uF cap in parallel to run the bass unit up to ~400Hz. The mid unit is not being allowed to run out of steam in this version and so is rolled off, BUT there is no inductor in the filter of the high pass section on the midrange, just 2 x 50uF caps in series, with a resistor across one of them for impedance correction. Is this all it takes to cut it off ~400Hz? Why no inductor in parallel? Might this be a cause of the excess bass I get from the KLS3 using this crossover design? Could I rebuild this crossover and add a resistor somewhere to try to get rid of the extra bass?

I've considered buying the parts for the KLS11 design, another 3-way but one that crosses over LF/MF at 800Hz. This means 2.75mH and 33uF on the bass unit and 12uF and 2.5mH on the midrange high pass section. I have some of these components already and could get the others for not a lot of cash, but is this idea barmy? I can take it on the chin.....

KLS3 MK II Gold

Any advice will be appreciated as usual.

petercom · #2 25th July 2006, 07:59 AM

First of all don't worry about what you see in other people's designs and what your speakers are actually doing. Crossover 'rules' are there to be broken and what works in theory rarely works in practice due to the non linear nature of drive units.

Having said that you are quite correct that an inductor in parallel with the midrange unit is used to increase the slope of the bass to midrange crossover. A series capacitor, as used in the Mk2, provides a first order electrical crossover, and the addition of the inductor takes it to second order electrical.

Notice I said electrical. That is because the midrange will have a mechanical roll off towards the bass end due to the restricted volume of the enclosure. This, in itself, usually approximates to a first order slope to start with leading to a second order slope below resonance. If we then add a first order electrical crossover as in the Mk2 the electrical and mechanical slopes add to provide an approximation to a second order slope, leading to third order below resonance. This is what was derived in the Mk2 to match the second order slope of the bass unit crossover.

The important point here is that the designer(s) adjusted the electrical crossover to give a phase integrated combined crossover. If you go and add an inductor to force a change to second order electrical you will almost certainly mess up the phase integration.

I suspect that your bass problems have little to do with the midrange unit, but you can check for yourself. Just disconnect the bass unit and listen to the amount of bass coming from the midrange. If it isn't doing anything in the frequency range that you are having problems with then it is entirely innocent.

The answer is probably more mundane. A reduction of series inductor with the bass unit from 12mH to 2.7mH has two effects. One is, obviously, to increase the amount of upper bass and midrange output. The other is a consequence of the amount of copper in the inductor - the series resistance is lowered. This resistance is constant irrespective of frequency and so the 12mH coil actually reduces the overall bass energy.

In the both crossovers the value of the coil can be adjusted to give the bass and midrange performance that you like. If you don't like the bass of the mk1 but like everything else about it then you can either:

a) put in a coil with a lower series resistance. This will increase the overall bass level, probably mostly noticeable at the lowest frequencies, but is very expensive.

b) reduce the coil value. You will need to experiment with this as it will require you to listen and balance the change in upper bass and midrange output with the 'improvement' in bass. I would start off unwinding in steps of 6 turns, reducing this to steps of 3 turns as you get closer to the sound you want. Bear in mind that, as the inductor value reduces, the midrange will start to become more coloured and blurred/less distinct as the bass and mid overlap more. You can probably increase the capacitance in parallel to compensate for this - try steps of 10uF.

Hope that helps and contact me if you need any special components.

Brian · #3 25th July 2006, 07:48 PM

Peter

Thanks for the reply.

Right, there is some intrigue now.

I've swapped out the MK I bass filter only, and dropped the MK II bass filter back in, nothing else has been changed. It sounded a bit odd, a bit dirty which I put down to perhaps the LF and MF now reproducing the same section of the audio range. The interesting thing is, no bass boom at all.

So, I opened the MF section and dropped back into the circuit the 2 x 50uF caps, one of them with the 3.3ohm resistor across it. Now there is loads of bass boom.

The 15 ohm resistor is present in MK I and MK II crossovers, in the MKI it is directly in parallel with the solen cap, in the MKII, as seen in the diagram earlier, it follows the 2 x 50uF caps. I'm going to move that resistor to a a position BEFORE the 2 extra caps, putting it back directly in parallel with the solen cap and see what happens.

I now suspect it is possibly incorrect in the circuit diagram for the MKII.

Brian

Brian · #4 25th July 2006, 08:04 PM

Only took a couple of minutes to make the change. The 15ohm resistor is now before the 2 x 50uF caps, seen after the caps in the MKII style crossover above. There is more bass power than the MKI but there is no bass boom. I'll have to leave it now for a few weeks to see what I think on a lot of music, but I'm making some progress.

I really do think now that the diagram in supplement number 32 is wrong, either that it's just my ears/room or something else.

Brian

Greg. · #5 25th July 2006, 11:08 PM

Hi Brian,

Your observations certainly confuse me. As you know I've built both versions and neither had bass bloom and both Mark 1 and 2 sound good. Having said that, I'll follow your thread with interest as it may end up being relevent to my own intentions on KLS3 tweaking. Please keep the info comming.

Best Wishes,

Greg

Brian · #6 26th July 2006, 08:20 AM

For ease of checking in one post, and for clarity because I know I don't explain these things very well, here are links to the 2 crossovers.

As explained earlier, in the MK_I crossover the 15ohm resistor is in parallel with the 8uF cap.

In the MK_II crossover diagram the value of the cap is now 6uF but the resistor is the same value and it is still in parallel, but it is now in a position after all of the additional components. Therefore, if looking at the MK_II crossover, I've moved the 15ohm resistor to an equivalent location as in the MK_I circuit. The 15ohm resistor is still in parallel with the 6uF cap, but the way I've constructed the crossover now places the new components after this section in the circuit.

It sounds very good now, tbh.

KLS3 MK_I_Crossover

KLS3_MK_II_Crossover

Brian

petercom · #7 26th July 2006, 05:20 PM

Quote:

Originally Posted by Brian

Peter

Thanks for the reply.

Right, there is some intrigue now.

I've swapped out the MK I bass filter only, and dropped the MK II bass filter back in, nothing else has been changed. It sounded a bit odd, a bit dirty which I put down to perhaps the LF and MF now reproducing the same section of the audio range. The interesting thing is, no bass boom at all.

So, I opened the MF section and dropped back into the circuit the 2 x 50uF caps, one of them with the 3.3ohm resistor across it. Now there is loads of bass boom.

The 15 ohm resistor is present in MK I and MK II crossovers, in the MKI it is directly in parallel with the solen cap, in the MKII, as seen in the diagram earlier, it follows the 2 x 50uF caps. I'm going to move that resistor to a a position BEFORE the 2 extra caps, putting it back directly in parallel with the solen cap and see what happens.

I now suspect it is possibly incorrect in the circuit diagram for the MKII.

Brian

The 15 Ohm resistor is nothing to do with the caps, it is meant to be in parallel with the drive unit.

What I think might be happening when you are listening is that, without the two 50uF caps, the midrange output is higher due to the considerable overlap between the bass and midrange. The resulting lack of bass boom is possibly due to the greater midrange power diverting your attention away from the bass.

However the two 50uF will also sharpen the roll off of the midrange unit and may produce a hump in the upper bass through the crossover region. Difficult to know without measurement as to what exactly is going on. As I said before it is instrumental to disconnect the bass unit so you can hear exactly what the midrange unit is doing in the bass end whilst you are fiddling around with the crossover.

Brian · #8 26th July 2006, 08:36 PM

Peter

I appreciate your help with this.

I've restored the crossover back to exactly how it appears in the MK_II diagram. There is loads of bass boom again.

I disconnected the bass driver. No evidence of boom from the mid unit.
I reconnected the bass and disconnected the mid. The potential is there for bass boom from the bass driver, it's less obvious though without the mid in the circuit and I need to turn up the volume to get it.

A couple of points, because I'm obviously pretty much stuck now.

1. The fact remains that by moving that 15ohm resistor to put it in parallel across the 6uF cap the bass boom is mostly gone.

2. With the proper crossover design for the mid, I turned my attention to the bass filter and slotted another 15ohm resistor in parallel across the 50uF cap, a significant level of the bass boom has gone.

I've read you say before that it's a case of sitting there with a bucket of components, I know what you mean.

Brian

petercom · #9 27th July 2006, 10:23 AM

All you are doing with that 15 Ohm resistor is moving the attenuation around - a parallel resistor will shunt energy away from the drive unit and present a reduced impedance to the crossover. It is trial and error. You just need to be careful that you don't lose any midrange/bass detail when inserting/moving components.

You could also try a combination of 50uF and 15 Ohms (in series) in parallel with the bass unit and the existing 50uF. You can then play around with the resistor, down to around 10 Ohms, to hear what works best. This combo will leave the lower bass untouched but should reduce upper bass boom as well as correcting the impedance better than just the plain resistor does.

Brian · #10 4th August 2006, 02:03 PM

My KLS3's are now back in MK I guise. I can't shift the excess bass in MK II. As Peter has said, moving the resistor simply disguises the problem.

One thing I noticed when I changed it back to MK I is that the wire used in the inductor I bought for the MK I design is a lot thinner than the wire used for the MK II. Other component quality is similar throughout.

Bottom line is I need a speaker that offers something between MK I and MK II.....

Brian