diyAudio (
-   Multi-Way (
-   -   Understanding John Dunlavy's Crossover Designs Crown Prince (

Zero Cool 23rd June 2012 11:08 PM

Understanding John Dunlavy's Crossover Designs Crown Prince
2 Attachment(s)
I am working on rebuilding the crossovers in my Duntech Crown Prince speakers. After speaking with Duntech AU they suggested replacing all the capacitors and resistors as Technology in the last 20 years has greatly improved these parts. So I have torn one crossover apart and mapped out the schematic and verified all the parts values with the exception of the Inductors which i will have measured next week.

There are many interesting things and some puzzling things as well about this crossover design. I will start with the top working down on the schematic.

Zero Cool 23rd June 2012 11:25 PM

The first thing I noticed is the upside down Zobel network forming more of a Boucherot cell in the tweeter network. As expected a 6db per octave first order network as i have always read Dunlavy used.

The midrange gets quite interesting. I have read that each speaker was "voiced" to fit within a chosen frequency response window. I was a little concerned that the left and right speakers would be different. But both crossovers check out to have all the same values...BUT the resistors while the same value, differ in construction! there are 3 distinct types (and colors) of resistors used and they are not the same parts in the same places. again they are the same values however!

Interesting in the midrange section is the zobel across the high pass caps. I checked this many times to make sure that this is in fact exactly how it is wired and it is!

The group of caps totaling 45uf at the input to the midrange section and Inductor L4 is expected to form a 6db first order network. But L2 in the middle of the voltage divider was not. I'm curious about that. L4 is followed by two Tank circuits where they must be tuning out some peaks in the response at a guess??

Where things get interesting is C4. this is a 50uf single very large cap. Everywhere else in the circuit he used multiple paralleled caps to form a larger value. why in this location only did he choose to use one single large cap?

Moving down to the woofer network see L10 (typo, was supposed to be L1) which is a very large inductor. followed by a Zobel network with multiple paralleled caps made of 10, 10, 10, 20, 20uf caps which seems odd to me that they didn't use 20, 20, 20, 10 instead? and again why did he use one large 50uf cap in the midrange tank circuit and not kept form by using 20, 20, 10uf as elsewhere? maybe i am making more of this then there really is but it is interesting.

The following resistor switch network is a low frequency room compensation network that was done away with in later versions. as suggested I will be removing this network as well.

auplater 23rd June 2012 11:31 PM

Doesn't look too exciting... Dunlavey was big on linear response...

Looks like some response shape filtering as well as shelving and notching going on...

Hard to tell anything else w/o the driver compliment

Zero Cool 23rd June 2012 11:33 PM

The tweeter is a Dynaudio D28AF. Not sure what the midranges are yet but the woofers are Dynaudio 24W75's in a WMTMW config.

bear 24th June 2012 04:23 AM

You might want to simulate that midrange bit in LTSpice or similar and see what it really does..

Do your speakers have good bass?

Dunlavy claimed they could do a good square wave midband.
Problem with the speakers is/was higher IM on the tweeter...

_-_-bearwith the spea

Zero Cool 24th June 2012 04:28 AM

The bass is very good! surprisingly good in fact! I was quite shocked when I opened them up and realized how SMALL the woofers are for the amount of bass they produce!

David Gatti 24th June 2012 09:07 AM

From memory the mids are from Vifa but I could be wrong.
Very interesting project.

Zero Cool 24th June 2012 12:18 PM

I would really like to know what the mids are but I am just not going to tear the cabinets apart to find out. the woofers are being re-foamed right now so I had to take them out and I can see the back of the tweeter from the inside of the crossover cavity.

To me the speakers sound just slightly recessed in the midrange. like vocals are slightly further back in the hall then I am used to hearing from my Hales Revelation 3 speakers I upgraded from. Also the Duntechs just don;t have the same transient response the Hales have! which to me is my number 1 pet peeve about loudspeakers. I mean if you stand next to a guy banging on a snare drum, the sound is very very sharp and it makes you jump/flinch. same with a triangle. the sound while soft has very large transients that when played back on 99% of the speakers I have ever hear, just round everything off into mush. The Hales were very very good at that. the best i ever heard in fact. where as the Duntechs are very soft sounding...

I am hoping that rebuilding the crossovers will open the soundstage up a bit and maybe bring that midrange forward. according to Duntech AU new caps and resistors really bring these speakers we will see.

Oh yeah BTW, the recessed Midrange MIGHT have been from a resistor that wasn't soldered!!! One one crossover board in the midrange section at the input, there are two paralleled resistors. one wasn't soldered to the other on either end. it was just wrapped around loosely! and the glaze they dipped the board in was holding it in place!! ooops! a factory mistake!!!

bear 24th June 2012 02:16 PM

Frequency response and peaks therein can often have the effect of making a given speaker sound "fast"... in theory the Duntechs have nearly ideal transient response, which is why they can do a 1kHz. square wave more or less correctly... or at least that is what some of his speakers did, not sure about this model.

Also, just a thought, depending on the vintage, later on John Dunlavy was having some troubles with the business, so he/they may have put speakers together with whatever they had on hand to do the job, as in the differences in the xovers - or perhaps you have two different production runs for the xovers and/or the cabinets? Only speculation.

I kinda doubt the resistors will have much effect unless perhaps if you switch to thick film types. The caps are more likely to have an effect, imo...


a.wayne 24th June 2012 03:00 PM

Changing out those sand cast resistors to wire wound will make a difference, we fond goop degrades the sound (dead) this maybe the biggest issue in my books when redoing that xover.

Also your Hales have better bass than the Duntechs , this goes along way in dynamic reproduction and pretty much a weak spot in all of Dunleavy's speakers IMO, aside he was pretty crafty and creative ...

All times are GMT. The time now is 03:27 PM.

Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 18.75%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Copyright ©1999-2018 diyAudio