SpeakerTalk Forum Index SpeakerTalk
This forum has been set up to facilitate discussion of 1970s KEF speakers and drive units. The owner of the Forum has no connection with KEF Audio.
 
 FAQFAQ   SearchSearch   MemberlistMemberlist   UsergroupsUsergroups   RegisterRegister 
 ProfileProfile   Log in to check your private messagesLog in to check your private messages   Log inLog in 

Refirbushing Kef 105.1, NEED CROSSOVER SCHEMATIC, etc..
Goto page Previous  1, 2, 3, 4  Next
 
Post new topic   Reply to topic    SpeakerTalk Forum Index -> KEF Speakers from the 1970s
View previous topic :: View next topic  
Author Message
ColinR
‹ber Contributor 1000+


Joined: 31 Jul 2004
Posts: 1175
Location: Staffordshire

PostPosted: Tue Feb 24, 2009 5:20 pm    Post subject: Reply with quote

Surely the idea behind any active loudspeaker design is to reduce distortion, improve headroom and increase the spl by reducing distortion and non-linearityís cause by passive components.

For a design without corrective circuits you do require each drive unitís useable frequency response, the flatter the better.

Now with respect to both the B110 A&B it appears their output is good down to below 160Hz according to KEFís frequency plots.

This has been confirmed by ear when integrating B139, B200 and B300 bass units at 160Hz with my 24dB/octave Behringer CX3400 with KEF Ref. 101s, Codas, JR 149s, Dalines, Stags, Klang Ton TLs and that substandard BBC design, clones of which cause the destruction of so many fine loudspeakers.

So an active crossover for a Ref. 105-2 centred at ~150Hz isnít a surprise to me.
_________________
This post or any other information supplied to this website or any other by myself is not available for any form of commercial purpose i.e. to hi-fi magazines or as sales and marketing material for sleezeBay or Audiodogging pimps and the like.
Back to top
View user's profile Send private message
audiolabtower
VIP Contributor 500+


Joined: 06 Jan 2009
Posts: 544

PostPosted: Tue Feb 24, 2009 7:24 pm    Post subject: Reply with quote

ColinR wrote:
So an active crossover for a Ref. 105-2 centred at ~150Hz isnít a surprise to me.


neither is it a surprise to me, it was the idea of the passive being so that is difficult to understand unless there was some major late revision which would have meant major crossover redesign, and greater expense.

Thinking about say the LS3/5a volume in comparison, the B110 enclosure even on the Mk1 "should" be fine for this active rolloff given that 100Hz is already ~12dB down from the shaped peak. The LS3/5a in comparison goes "flat" to 70Hz with the old B110, power handling issues notwithstanding, but this whole octave up in crossover for the newer B110B would hugely increase power handling and SPL for the active 105 in comparison, given the B300 has the real job of diaphragm excursion.
Back to top
View user's profile Send private message
audiolabtower
VIP Contributor 500+


Joined: 06 Jan 2009
Posts: 544

PostPosted: Sun Mar 01, 2009 7:24 pm    Post subject: Reply with quote

Has anyone considered building this active circuit? What op-amps would replace the 351's nowadays that don't cost a relative fortune, after all 18 are needed Smile ?
Back to top
View user's profile Send private message
BobL
Intermediate Contributor 25+


Joined: 27 Jan 2004
Posts: 38
Location: West Midlands, UK

PostPosted: Mon Mar 02, 2009 1:09 pm    Post subject: Reply with quote

audiolabtower wrote:
Thinking about say the LS3/5a volume in comparison, the B110 enclosure even on the Mk1 "should" be fine for this active rolloff given that 100Hz is already ~12dB down from the shaped peak.


It seems the R105 mk1 has a 7 litre mid enclosure and the mk2 has an 8.5 litre enclosre. As dropping from 400 Hz down to 150 is more than an octive I can see that the 20% larger enclosure of the mk2 wouldn't help that much.
Back to top
View user's profile Send private message
BobL
Intermediate Contributor 25+


Joined: 27 Jan 2004
Posts: 38
Location: West Midlands, UK

PostPosted: Mon Mar 02, 2009 1:16 pm    Post subject: Reply with quote

audiolabtower wrote:
Has anyone considered building this active circuit? What op-amps would replace the 351's nowadays that don't cost a relative fortune, after all 18 are needed :) ?


I built it on vero board over 20 yeras ago. It measured very well just like the graph shows it should. It had an obvious and substantial improvement in the bass but I couldn't get the speaker as a whole to sound quite so convincing in active mode. This may have been because I couldn't at the time find 6 identical power amps. Some where bipolar and others fets so not sure if that was the cause.

I also ran just the bass part active with the turnover changed 400Hz. that was interesting.

Bob.
Back to top
View user's profile Send private message
BobL
Intermediate Contributor 25+


Joined: 27 Jan 2004
Posts: 38
Location: West Midlands, UK

PostPosted: Mon Mar 02, 2009 1:54 pm    Post subject: Reply with quote

audiolabtower wrote:
The crossover is 4th order (Linkwitz Riley ?).


I know that is what KEF said at the time but I would love understand why. Take the mid range xover. To my eyes it has 3 components rolling off the bass and 3 components rolling of the treble. Isn't that 3rd order?

OK soy so you add the cabinet/driver response to make up the bass roll off to 4th order, I can't believe the driver would roll off at the top end just right to make up the other 6db per octive we're short of 4th order there.

Same with the B300, to me that's a 3rd order xover ie. 2 inductors and a capacitor. Especially as the active version definately is 4th order.

If someone could explain I would be very interested.
Back to top
View user's profile Send private message
audiolabtower
VIP Contributor 500+


Joined: 06 Jan 2009
Posts: 544

PostPosted: Tue Mar 03, 2009 12:12 am    Post subject: Reply with quote

Hi, yes the design was computer optimised for the drive unit equivalent L, R, and mechanical impedances and their rolloffs and these then make the total measured design 24dB/octave rolloffs through and down the crossover regions. This Linkwitz-Riley total filter gives the excellent phase response with no sharp discontinuities. This was all described in great detail in Keftopics Volume 3 No 1, the design of the 105. It is also mentioned in the official spec
http://www2.kef.com/gb/about/museum
click on 1970s then 105.

I suspect the active bass is 4th order because the crossover has changed from 400Hz to 150Hz and at this lower frequency the B300 is not rolling off to the extent it would at the higher frequency, thus the extra section? The mid and treble active sections are 3rd order and designed for the rolloff just as the passive crossover was with the addition of the required response shaping sections. The B110 seems fine with the extra bandwidth without another section one assumes because of the "hump" shaping? As mentioned before if Andrew Jones was available he could give the definative answer.

I have gone through my caps and luckily have all the 2.5% polystyrene available so will have a go at building this active circuit. I've done the vero layout which comes out pretty neat and have some smallish toroids which will do the power with 78 and 79 regulators and plenty of smoothing. I have 6 channels of Audiolab power amp available, they are all gain matched etc, so drive is not a problem.

The only problem is the op-amps. I built a Quad 405 when they first came out and it has been upgraded with all the latest (Ludwig/Snook) mods over the years (but not used in the main system anymore) and used OPA134 there. However 18 of those would be close to £100 so I'm thinking more of TLE2071 which is the lower noise/distortion/higher bandwidth next generation TLO71 which would only be around £1 each. I'd use 5534 but worry about unity gain stability without a scope to check things out anymore.

Certainly rebuilding the crossover with poly caps matched to better than 0.25% of the original values gave good benefit, but separating the network sections, tri-wiring throughout and placing the crossovers at the amp terminals and running speaker cables from crossover output to drive units just magnified all the original attributes of the 105, superb pin sharp imaging and depth etc. Experience says the active should be better if the chips are good enough not to degrade the rest of the electronic chain so plenty of local decoupling on each op-amp etc and the choice of chip is then important.
Back to top
View user's profile Send private message
audiolabtower
VIP Contributor 500+


Joined: 06 Jan 2009
Posts: 544

PostPosted: Tue Mar 03, 2009 1:31 am    Post subject: Reply with quote

update, Farnell do the OPA134 at £1.20, I thought they were much more expensive than that - they were the last time I bought some - so OPA134 it is.
Back to top
View user's profile Send private message
speakerguru
VIP Contributor 750+


Joined: 18 Nov 2005
Posts: 958
Location: Green Hut, Tovil

PostPosted: Tue Mar 03, 2009 11:57 am    Post subject: Reply with quote

BobL wrote:
To my eyes it has 3 components rolling off the bass and 3 components rolling of the treble. Isn't that 3rd order?

In short, it is the total unit/eq/crossover/box response which is 4th order, NOT just the electrical response of the passive filter.

In long, the important thing about Linkwitz-Riley or Butterworth Squared crossovers is that the high pass and low pass ACOUSTIC responses be in-phase for about a couple of octaves either side of the x-over frequency. Otherwise the main summation axis will not be pointing in the design intent direction. Because of high frequency roll-off, directional characteristics and voice coil inductance, the order of the electrical filter often can be less than the final resulting acoustic filter shape over the required band.
Back to top
View user's profile Send private message
BobL
Intermediate Contributor 25+


Joined: 27 Jan 2004
Posts: 38
Location: West Midlands, UK

PostPosted: Tue Mar 03, 2009 1:47 pm    Post subject: Reply with quote

audiolabtower wrote:
update, Farnell do the OPA134 at £1.20, I thought they were much more expensive than that - they were the last time I bought some - so OPA134 it is.


I will be very interested to hear how you get on.

Bob.
Back to top
View user's profile Send private message
BobL
Intermediate Contributor 25+


Joined: 27 Jan 2004
Posts: 38
Location: West Midlands, UK

PostPosted: Tue Mar 03, 2009 1:56 pm    Post subject: Reply with quote

speakerguru wrote:
BobL wrote:
To my eyes it has 3 components rolling off the bass and 3 components rolling of the treble. Isn't that 3rd order?

In short, it is the total unit/eq/crossover/box response which is 4th order, NOT just the electrical response of the passive filter.


Thanks for that.

Yes, I can see it in some cases like the roll off of the B110 at the bass end dues to the driver cabinet combination but its much harder for me to see the none electrical roll off in other places. Especially as they occur just at the right frequency and have the right 6db slope which added to the 18 electrical gives 24.

Would it be true to say that if the active xover was used at 400Hz and not 150Hz that the bass electronic xover would need to be reduced to 3rd order because of natural roll off at 400Hz? Would the phase aligment between B110 and B300 work better at 400Hz/3rd order than 400Hz/4th order?

Bob.
Back to top
View user's profile Send private message
speakerguru
VIP Contributor 750+


Joined: 18 Nov 2005
Posts: 958
Location: Green Hut, Tovil

PostPosted: Tue Mar 03, 2009 2:47 pm    Post subject: Reply with quote

BobL wrote:
but its much harder for me to see the none electrical roll off in other places. Especially as they occur just at the right frequency and have the right 6db slope which added to the 18 electrical gives 24.

Yes it's hard without the complete set of amplitude and phase measurments of each component and combination in front of you to look at. In fact, it's impossible, so I would advise that you don't attempt to see it all intuitively. What most people outside the speaker design engineering community don't realise is just how far down into the passband the phase response is affected by voice coil inductance and stopband roll-off and this has to be taken into account in the design of the crossover and any integral passive eq.

BobL wrote:
Would it be true to say that if the active xover was used at 400Hz and not 150Hz that the bass electronic xover would need to be reduced to 3rd order because of natural roll off at 400Hz? Would the phase aligment between B110 and B300 work better at 400Hz/3rd order than 400Hz/4th order?


I don't have any data to answer those questions directly but I can say that if you can find any two complementary high and low pass filter shapes (of any order including assymetric) which result in an in-phase acoustic addition on the axis you are designing for, then the x-over will work. If the resultant amplitude is not flat then you can apply the same eq to both sections (this will avoid disturbing the relative phase) and it will then still add up correctly in space.
Back to top
View user's profile Send private message
audiolabtower
VIP Contributor 500+


Joined: 06 Jan 2009
Posts: 544

PostPosted: Tue Mar 03, 2009 3:22 pm    Post subject: Reply with quote

Thinking about the lower crossover I had intended to built the 400Hz version but the 4th order B300 active is obviously not following the design intent of the original here. I think I shall layout both sets of components with a few solder links to swap between the 2 frequencies.

The B110 cutoff is complicated by the shaping network R11, R12 and C12, maybe this is why 3rd order is still ok here?

It all may be theoretical whether you will hear minor differences at 400Hz compared to swamping effects of room modes etc?

I am beginning to bend towards the theory that the optimum low crossover is 150Hz, but it was just not possible to achieve economically or practically with passive components.

Or alternatively the active is an invention of AJ as a thought experiment or one off experimental based on his wide experience of this design when he was planning Kubes etc for the next generation of Reference series?

We seem agreed that the B110 volume should not be a hindrance in running down to 150Hz, I can see nothing but advantages in room by giving the B100 more low mid to output and pushing the B300 crossover down further providing the total power handling and max spl stays the same. The original 400Hz specs were 35Vrms sine for the B300 and 28Vrms sine for the B110. I don't think music peak signal will have much of a difference in that we are listening at home rather than at studio monitoring deafening levels.
Back to top
View user's profile Send private message
audiolabtower
VIP Contributor 500+


Joined: 06 Jan 2009
Posts: 544

PostPosted: Tue Mar 03, 2009 3:42 pm    Post subject: Reply with quote

personally I've never understood how the large series B300 capacitor worked. It must have done something for the total bass loading or else Kef would not have spent money on making up such a relatively expensive component. The active seems to have no equivalent. Confused
Back to top
View user's profile Send private message
speakerguru
VIP Contributor 750+


Joined: 18 Nov 2005
Posts: 958
Location: Green Hut, Tovil

PostPosted: Tue Mar 03, 2009 4:21 pm    Post subject: Reply with quote

I haven't spoken to Andrew Jones recently and I only have a dim recollection that he'd done an active 105. Was it before he joined KEF? Mike Gough did both passive 105 and 105.2. I did the 105.4. I will try to answer any specific technical questions as they come up, but, in case you're wondering, I'm sorry I don't have the time for general speculative chit-chat and I don't do "musicality" or directional wires.
Back to top
View user's profile Send private message
Display posts from previous:   
Post new topic   Reply to topic    SpeakerTalk Forum Index -> KEF Speakers from the 1970s All times are GMT + 1 Hour
Goto page Previous  1, 2, 3, 4  Next
Page 3 of 4

 
Jump to:  
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum


Powered by phpBB © 2001, 2005 phpBB Group