I recently saw a 'Lightspeed' passive LDR attenuator which looked pretty tasty as a pre-amp. The blurp says that it uses a optocoupler device that is a sealed unit which consists of a high performance LED that shines on a light dependent resistor (LDR). This means that it is a passive preamplifier with 'no contact points in the form of a "wiper" as in the case of a potentiometer or "switches" as may be with a discrete stepped attenuator'.

All sounds pretty good, but I guess that it usually works with poweramps (otherwise in intregrated amps the volume control mechanism there would undo all the passive LDR attentuator good work, right?). Also it works best apparently with amps with an input impedance of more than 50K. So....

1) What do people think would be the best way to convert the ZKIT1 into a poweramp? I was thinking that I could remove the pot and put a resistor along each signal path where the pot was with a value a fair bit greater than 1k. Is that correct? Also given the way I have soldered my pot and made my case getting at the pot and removing it might be quite difficult without desoldering a lot of connections (transformers, RCAs, speaker jacks) to remove the board. Is there an feasible alternative that doesn't involve removing the pot?
2) Does anyone know what the input impedance of the ZKIT1 would be? How do you measure input impedance?
3) Also if anyone has heard the 'Lightspeed' or another passive LDR attenuator, any thoughts?
Cheers,
Matt

I've recently modded my TS GG pre to incorporate optocouplers as the attenuating device.
I can't praise this mod enough. removing the attenuator from the circuit has astonishing results.
Don

Dan, that's the same conclusion I came to.  Here's a DIY link with a full description:

http://diyaudioprojects.com/Solid/DIY-Lightspeed-Passive-Attenuator/

Chris

I have been looking at info re: digital attenuation (via computer software) vs LDR attenuation and theoretically digital attenuation should work better. Given that I only use computer as source, it makes sense that I should first try bypassing the pot in my zkit1 and use the computer's attenuation to adjust volume. Am I right in thinking that the resistance of the pot varies between 100R and infinite, so I will to replace the pot with a resistor greater than 100R? What value would people suggest? I don't want the amp to clip.

Thanks

Matt

Thanks Dan.
Sorry about my rudimentary electronics knowledge. I didn't really know how pots worked so from what you are saying it acts like a variable voltage divider. Because of the logistical nature of sound if I find a volume I like then I will have to measure Vin and Vout and calculate the resistances: Vout=(R2/R1+R2)*Vin. Am I right?

The reason I wanted to remove the pot was as per information given about the LDR attentuator which suggested that it is a pot's wiper that adds distortion to the signal. The LDR reportedly removes the distortion by removing an contacts in the attenuation process. That was my undertsanding anyway. Given that 1) I have heard only great things about how the LDR attentuators improve the sound and 2) that digital attentuation is reportedly the most accurate, I figure that I might be able to get a simliar effect to having a LDR attentuator by removing the pot from the signal path in the zkit1 and using digitial attenuation to change volume. This mod would be much easier to do than try to get hold of a matched pair of optocouplers and build the LDR attenuator (particularly as I couldn't just add a module within the amp enclosure as the heat would likely decalibrate the attenuator, so would have to build a separate enclosure). I figure that if I leave the pot in the signal path this distortion will still exist even if is turned up to full volume (and at full volume the hum does get a bit annoying).
What do you think? Is my rationale sound?

As a side issue, not all digital attenuation is accurate. If you do try out XXHighEnd like I suggested in the DAC-1 thread, hover your mouse cursor of the volume controls to pulll up and information box on how to use them because they aren't obviously .

Thanks again for your help Dan,

Matt

I think for the voltages produced by a battery operated VOM the impedance of the grid would look infinite and not really affect the accuracy of this measurment for practical purposes.  Alternatively the tube could be removed and then make the measurement.

Matt,
My comment was in regard to my opinion you do not need to remove  the pot in order to measure the the equivalent R1 & R2 resistances for your purposes.

But I like Dan's suggestion to turn the pot at maximum and leave it there.  Why not leave in the pot for now?  That way you could actually observe how the lightspeed controls the volume by itself and whether you need to insert a fixed resistor divider in place of the pot at all.  

After some listening and playing around with volume settings and sources you will either find the optimum pot setting that you'd want to replicate with a fixed set of resistor or you may find you don't need to replicate the pot at all.  

Chris

Thanks Chris and Dan.

Glad to hear that I can measure the pot resistance without having to remove it. So to make it absolutely clear to me to work out the fix resistor values I turn the amp off and measure 1) input to wiper, and 2) wiper to ground of the pot and replace with resistors right?

In terms of turning the pot to maximum, I assume that the rationale is that the wiper is no longer in contact with resistance band of the pot so won't influence the sound. Is that right?

My concern with that is 1) that the hum returns quite noticeably when the volume is turned up to the final 1/8 of a turn. So if I compared  pot attentuation to 'lightspeed' attenuation the pot would have the advantage because it could reduce amplier hum, whereas the 'lightspeed' could not moderate  this directly - so any gains of the 'lightspeed' might be difficult to detect (I hate that hum). 2) My other issue is that I haven't got a 'lightspeed' attenuator yet so will have to get parts and build. Other than the cost there is also quite a delay in getting matching optocouplers and if the fixed resistors work as well, for me the attentuator will become redundant.

On the converse, its going to be difficult to do A-B-A comparisons with fixed resistors and I am not game for using alligator clips in a high voltage circuit!

Overall, I think that I am going to replace the pot with fixed resistor because as long as the resistors are good quality, I can't see that it will make the sound worse. I will try to listen out for any differences and let you guys know. Any recommendations re: type of resistors to use?

Thanks for your help,

Matt

Matt

If you are going to replace the pot with fixed resistors you really must remove the pot from the circuit.  Whether you physically remove it, or remove it by removing J4, J5, and the two 1k resistors connected to the input tube control grid's is up to you.

I just set mine to "8" on a 0-10 scale and measured the pot.  I got wiper to ground = 34k and wiper to RCA input = 16k (remember I'm using 50k pots, not the 100k you are using).  If we assume yours are double this, it would be close to 68k for R2 and 32k for R1.  I would use 68k and 33k.  I would remove J4, J5, and the two 1k resistors, solder one end of both 68k and 33k resistors in the 1k resistor hole that connects to the tube grid.  So now you have two sets of 2 resistors sticking up vertically out of the pcb.  Connect the two 33k's to the J4/J5 hole that connects to the RCA connectors with 2 short wires, and run another short wire from the both 68k's to ground.

Dan

Matt

If it were mine, that's what I'd do.   R1=16K, R2=34K.  But I'd round up/down to standard resistor values so I'd use  R1=15K, R2=33K.  Then, if you want to adjust slightly, you can change just one of the resistors - say R1 = 10k to make it slightly louder, R1 = 20k to make it slightly quieter, etc  


Dan