Planar Speaker Asylum

Welcome! Need support, you got it. Or share you ideas and experiences.

Return to Planar Speaker Asylum


Message Sort: Post Order or Asylum Reverse Threaded

PLLXO question related to LDR

70.160.141.247

Posted on February 26, 2012 at 06:52:39
johnvb
Audiophile

Posts: 324
Joined: October 26, 2011
Been wanting to try one of the new light dependent resistor attenuators as a passive pre in my system, to replace the old Channel Island jobber I use now. The recommended system requirement specs for this design is a source of around 200 ohms output impedance and a power amp of 47k ohms or more input impedance.

I'm good with the output impedance spec, will be using a Dodd tube buffer after my DAC, utilizing a 6H30DR which will give me about a 100 ohm output. Power amp wise my Aragon(s) are low, 22k ohms, although folks have used amps with this low of input impedance and reported success. I was wondering, can a passive line level crossover be designed with a high input impedance, say 100K ohms? That way I can take care of two issues at once.

 

Hide full thread outline!
    ...
RE: PLLXO question related to LDR, posted on February 26, 2012 at 09:07:12
neolith
Audiophile

Posts: 3094
Location: Virginia
Joined: February 21, 2002
Contributor
  Since:
December 2, 2004
I would think a 100k input load would give you too much insertion loss to be practical. However I think this would work for your MMG's.


R-HP = 6190 ohm
C-HP = 1.7 nf
R-LP = 5600 ohm
C-LP = 120 nf (0.12 mfd)
I calculate the insertion loss to be -2db and the minimum input impedance seen by your tube stage is 4740 ohm.
Here's the Bode diagram comparing to stock:

Here's my PLLXO spreadsheet in Open Office.
 


 


 

RE: PLLXO question related to LDR, posted on February 26, 2012 at 09:28:17
johnvb
Audiophile

Posts: 324
Joined: October 26, 2011
Thanks Neolith, I sent you a PM

 

RE: PLLXO question related to LDR, posted on February 26, 2012 at 09:33:03
LDR-based attenuators exhibit a variety of non-linearities and high-ish distortion levels. I think the CI attenuator you have now is a superior device.

Anyways, Neo is certainly correct. Creating a low-pass filter with high input impedance (47k) into a power amp input resistance of 22k nominal guarantees an insertion loss of at least 10db. (Probably unacceptable.)

Cheers,

Dave.

 

RE: PLLXO question related to LDR, posted on February 26, 2012 at 09:51:24
johnvb
Audiophile

Posts: 324
Joined: October 26, 2011
Hey Dave. I figured after that discussion we had a few years ago regarding the Sony TAP preamp's non-linearities, that you probably had reservations with the LDRs. But the fans of these things just lov em. :)

The CI preamp is pretty long in the tooth, bought it used, years ago off the Audiogon. My fall back idea was to replace it with a Goldpoint Mini-V. That actually would be less expensive, and easier to build. I want to build two matching chassis, one for the tube buffer, and another for whatever passive I go with. And of course still interested in a PLLXO.

 

RE: PLLXO question related to LDR, posted on February 26, 2012 at 10:47:28
JBen
Audiophile

Posts: 2430
Joined: May 18, 2008
Contributor
  Since:
December 0, 0000
John, I would have to agree that LDRs have improved greatly in stability. After you mentioned your plan in another Asylum forum, I dusted my old research and then looked at what you saw. This was weeks ago and I am still keeping track of it for the future.

By and large, it looks promising. Too much, perhaps. I asked and the kits are not being made available anymore, or at least for now. They seem to be selling all the selected LDR paired sets that they can get as finished product.

As Davie and Neo suggest, insertion losses could be a problem. The thing I liked about this LDR design is the low input impedance. It appears that my existing PLLXO setup would hardly be affected by being right after the attenuator. If you consider that my LP is currently being set as 2nd order, I already have the bulk of losses accounted for. I can still go to 1st order on it and remove impedance.

BTW, this may work just as long as I don't want to add a properly timed subwoofer. Things get trickier for this and Davey may have a good perspective on this also.

 

RE: PLLXO question related to LDRse, posted on February 26, 2012 at 11:10:30
neolith
Audiophile

Posts: 3094
Location: Virginia
Joined: February 21, 2002
Contributor
  Since:
December 2, 2004
The problem with having a passive pre (regardless if it is an LDR or not) is that the output impedance varies if you change the volume. This means the the fc of the filters and the xo point of the PLLXO will also vary everytime you change the setting on the passive. You could put an opamp buffer between the passive and the PLLXO but then you might as well go to an active crossover.
 


 


 

RE: PLLXO question related to LDR, posted on February 26, 2012 at 11:21:25
johnvb
Audiophile

Posts: 324
Joined: October 26, 2011
I've read the Warpspeed is no longer offered as a kit form, but Uriah Dailey is still selling a kit (actually you can get the parts, or the boards pre-assembled). AFAIK, he made an agreement with the original Lightspeed designer, George Stantschleff, not to sell completed units. See the link below.





 

RE: PLLXO question related to LDRse, posted on February 26, 2012 at 12:56:15
JBen
Audiophile

Posts: 2430
Joined: May 18, 2008
Contributor
  Since:
December 0, 0000
Neo, I may have failed to realize this. It could mean that my UBS DAC and the pre-amp have the same output impedance, which I don't know for a fact. I can plug into each separately and measure no change in xover points.

I never saw a spec call for it while designing the PLLXO filters and simply assumed that input impedances was all I had to worry about.

Well, time to start lining up the active stage ducks, then. It was in that "future projects limbo" for when a proper subwoofer integration became a need anyway.

 

RE: PLLXO question related to LDRse, posted on February 26, 2012 at 13:03:15
You have to consider/include 'everything' when designing PLL's. :) Even interconnect capacitance might be an issue.

You can measure output resistance of a source fairly easily. Take a voltage measure unloaded and also with a 1k resistor attached.

Rout = 1000 times (Ropen-Rload)/Rload

You'd be surprised how high the source resistance is on many components.

Cheers,

Dave.

 

RE: PLLXO question related to LDRse, posted on February 26, 2012 at 13:11:31
JBen
Audiophile

Posts: 2430
Joined: May 18, 2008
Contributor
  Since:
December 0, 0000
Thanks Davey! Where do I find the impact relative to changes in Rout in a given configuration?

 

RE: PLLXO question related to LDRse, posted on February 26, 2012 at 13:12:44
johnvb
Audiophile

Posts: 324
Joined: October 26, 2011
So is it's OK to place the pot after the PLLXO? I'm thinking of a dual deck (4 channel) stepped unit.

 

RE: PLLXO question related to LDRse, posted on February 26, 2012 at 14:01:37
neolith
Audiophile

Posts: 3094
Location: Virginia
Joined: February 21, 2002
Contributor
  Since:
December 2, 2004
Shouldn't that be Z = 1000*(Vload-Vopen)/Vopen?
 


 


 

RE: PLLXO question related to LDR, posted on February 26, 2012 at 16:41:23
Satie
Audiophile

Posts: 2508
Joined: July 6, 2002
Do you have any other source other than your DAC coming into the tube buffer?

The way to get a PLLXO and an LDR to work in tandem with minimal problems is to put the LDR in front of the buffer as a load to the DAC, and use the buffer's likely high input impedance as load for the LDR, also works well for TVCs (the ones based on Stevens and Billington's TX102 are excellent). The buffer's output impedance is ok for any load the PLLXO creates so long as it remains over 10Kohms, beyond that, the higher the better.

I would imagine the HP will end up being just one cap and a potentiometer (to control treble/bass relative volume) if you are doing 1st order, and the LP will be 10k or more resistor and cap - unless you are using an active for that purpose. In the latter case you would not need the pot on the HP and you will probably be ok on the combined input impedances as most actives have something around 20k, so combined you will have something like 10k ohms for the load the buffer sees.
If you end up having a problem, then just add a second parallel buffer for the bass after the LDR/volume control/pre, you can use a simple Y connector for it.

BTW, in Neo's calc, I would replace the fixed R-HP with a 20k pot (carbon or wirewound, linear - not log).

 

RE: PLLXO question related to LDR, posted on February 26, 2012 at 18:09:06
johnvb
Audiophile

Posts: 324
Joined: October 26, 2011
All my sources are upstream, digitally. I find it easier to hide some stuff in a closet on the front wall, and run a long 2 meter digital cable to my amp's location. There I can keep the analog cables short.

The buffer I'm building is the simple single I/O version, so yes I can place it before or after the volume control.

I know there is a diy Nelson Pass B-1 solid state buffer kit, that folks have used with the LDRs. So I could always add something like that if I need to buffer both ends. Fun!


 

RE: PLLXO question related to LDR, posted on February 26, 2012 at 19:47:34
Satie
Audiophile

Posts: 2508
Joined: July 6, 2002
Ah. The B-1 is great for these purposes. It is nearly fully transparent given top notch parts, a little more so is the DC coupled version. Davy (colin) has used a TX102 based TVC in conjunction with both the B-1 versions beating out the Lightspeed LDR and the highly regarded Arcadia pre and the Quad 99.

 

Page processed in 0.013 seconds.