The LEDs on the optocouplers of the Warpspeed are operated at more acceptable voltage/current levels anywhere you desire the volume control. At or near extremes on the Lightspeed will send it past the specified maximum 50mW...try it there for a length of time and you will lose the optocouplers. Keep it at pot mid-point please...
By comparison the Warpspeed actually has typical resistances on its listening range at very much lower levels than where the the Lightspeed operates. This is why the Lightspeed prefers higher input impedance amps. By this, the Lightspeed is more similar to typical high R pots used for passive volume control.
It's a solution and it is imperfect (which one is perfect anyway?) but imho closer to perfection than others. It will remain a riddle how it is done though, for now...
By comparison the Warpspeed actually has typical resistances on its listening range at very much lower levels than where the the Lightspeed operates. This is why the Lightspeed prefers higher input impedance amps. By this, the Lightspeed is more similar to typical high R pots used for passive volume control.
It's a solution and it is imperfect (which one is perfect anyway?) but imho closer to perfection than others. It will remain a riddle how it is done though, for now...
Several points on this.
A series and shunt resistors forms a voltage divide. Increasing the series and reducing the shunt at the same time will bring the division to a fraction of the input voltage. This is how the "Dead" silence is achieved. Not by just increasing the series resistance. If you have a series and shunt LDR and remove the power supply, this would still be voltage divider and you would theoretically get 50% of the voltage out - with not load. Depending on the input impedance of the load, the voltage division will change.
Second point. Turning "off" the LED will leave the LDR at the highest resistance. This is a natural state, not pumped up. In this state the LDR (depending on which you are using) will have about 1 Megaohm of resistance.
Johnson noise is proportional to thermal values in resistors, that is power dissipation, and voltage. The greater the resistance, the lower the power dissipation and the lower the noise. The voltage across the LDR will change according to its role - shunt, series, and the load. This would be a constant regardless of design - if the design uses a 2 LDR's in a voltage divider.
The ability to fine tune the current, hence the LED and the Light will be the ultimate decision on how to design the circuit. Going even further, the noise on the voltage of the LED will have more effect the the Johnson noise.
As a final note though, we have to pick our poison as all solutions will have some trade offs, some of them subjective.
Although the Lightspeed is a voltage divider, the range of operation is limited by the power delivery design to the LEDs -it still operates where amps still has enough voltage at the input to amplify and be heard.
I don't think Johnson Noise should even be considered, unless someone hears some kind of white noise coming from the volume control without the music playing.
I don't think Johnson Noise should even be considered, unless someone hears some kind of white noise coming from the volume control without the music playing.
I suppose Johnson Noise is not going to be audible but certainly it is changed (increased/decreased) by resistance.
Vn=SquareRoot(4KTRB) where k is Bozeman's constant, B is bandwitdth, T is absolute temp, R is the resistance. Its a pretty low amount of noise until we take the resistance up into and above megohms and then when added to any other noise in the circuit it really can have an effect.
This is my understanding from Self's Small signal book Page 10 Table 1.2 and he is talking about plain resistors as well as pots.
At anyrate, I feel like I took the bait to start talking about another's product on the Lightspeed thread.. again.
1 Megohm when turned off? Thats not a NSL32SR2 or an NSL32SR3. Sounds like a NSL32 or NSL32SR5, if its a Silonex Audiohm product. Ooh, or a combo of one of those for series and SR2 for shunt..
Cant wait to see pics of that massive optocoupler from HP. From a printer? That would make sense. Sounds fun
Uriah
Vn=SquareRoot(4KTRB) where k is Bozeman's constant, B is bandwitdth, T is absolute temp, R is the resistance. Its a pretty low amount of noise until we take the resistance up into and above megohms and then when added to any other noise in the circuit it really can have an effect.
This is my understanding from Self's Small signal book Page 10 Table 1.2 and he is talking about plain resistors as well as pots.
At anyrate, I feel like I took the bait to start talking about another's product on the Lightspeed thread.. again.
1 Megohm when turned off? Thats not a NSL32SR2 or an NSL32SR3. Sounds like a NSL32 or NSL32SR5, if its a Silonex Audiohm product. Ooh, or a combo of one of those for series and SR2 for shunt..
Cant wait to see pics of that massive optocoupler from HP. From a printer? That would make sense. Sounds fun
Uriah
Last edited:
I'll repeat, the Lightspeed's listening range has the series/shunt combination at higher R values, 4k to 5k on my version with the volume pot at midpoint. So if J Noise is a factor, which I doubt even with the Lightspeed, it is more affected.
At any rate, I don't need bait to get you hooked
...in fact you're the one who tossed a fishing line. You can choose not to answer if you feel you are being baited...
At any rate, I don't need bait to get you hooked
...in fact you're the one who tossed a fishing line. You can choose not to answer if you feel you are being baited...
In my case I just replied to what I think is wrong analysis about the design and optocouplers...and yes there is already a Thread and data
.
Diameter of each plate is 22mm.
If you can get a hp 3420b cheap somewhere...
Agilent HP 3420B
Price : Call
Equipment Location : USA
This dealer's latest update: 13 days ago
Seller : Test Equipment Value, Inc.
Phone : Please Use RFQ form
Fax : 818-340-2090
Used-Line Dealer since 2005
Description : HP 3420B For Sale in U.S.A Only
Agilent HP 3420B - For Sale at Used-Line
Last time it was running about USD$500.
I don't agree, but am interested in this thread you talk about. If you have subscribed to it as well, I'd appreciate the info since I keep track of only a few threads.In my case I just replied to what I think is wrong analysis about the design and optocouplers...and yes there is already a Thread and data
Thanks alot for the information.
I am using 68r resistors for the series ldrs (according to the manual) and 1k for shunt.
My preamp is Audio Note, poweramps are Quad II's (low gain) . Speakers are diy 15" coax with approx 96dB effiency.
In fact when volume pot is at its lowest the sound is "normal listening level".
An alps 100k log pot is in order. I have a noble pot 100k also I think so I may test it tonight..
Also changing the series resistors.
One more question. Would it be possible to change the 100k pot with for instance a pga2311 kit?
I am not sure.
http://cgi.ebay.com/PGA2311-Volume-..._HomeAudioHiFi_Amplifiers&hash=item3a631ddf75
Thinking about a remote option. Or should I go for motorized alps solution?
Sincerely
Last edited:
Hi,
The PGA2311 wont work like you want it to. It could be a preamp all by itself but it does not work the same as a pot. Its actually an opamp with programmable gain.
The 100k pot is going to help you a lot. Right now your series LDRs are very very low resistance at their max resistance compared to what they would be with a 100k pot. If you have normal volume at min with a 50k then I think you will be fine with the 100k.
Another Lightspeed builder has had success with those motoroized pots but make sure that you check the pinout of the motorized pot as its not typical. Also be careful with those motorized pots. They might work for a while and then die. They can only handle about .05W, not .5W. That said, I know one is successfully in use and probably more. They might be underrated.
Make sure your 100k is DUAL LOG. Try that noble if its a dual log.
Dont worry about changing any resistors until you try a 100k pot. Seriously. Those are just bandages to fix it when your entire system has to much gain for the 100k pot version. Try 100k, then decide if you need to replace any resistors.
Uriah
The PGA2311 wont work like you want it to. It could be a preamp all by itself but it does not work the same as a pot. Its actually an opamp with programmable gain.
The 100k pot is going to help you a lot. Right now your series LDRs are very very low resistance at their max resistance compared to what they would be with a 100k pot. If you have normal volume at min with a 50k then I think you will be fine with the 100k.
Another Lightspeed builder has had success with those motoroized pots but make sure that you check the pinout of the motorized pot as its not typical. Also be careful with those motorized pots. They might work for a while and then die. They can only handle about .05W, not .5W. That said, I know one is successfully in use and probably more. They might be underrated.
Make sure your 100k is DUAL LOG. Try that noble if its a dual log.
Dont worry about changing any resistors until you try a 100k pot. Seriously. Those are just bandages to fix it when your entire system has to much gain for the 100k pot version. Try 100k, then decide if you need to replace any resistors.
Uriah
The 100k dual log did not make any difference.
I will try increasing the series resistors.
I built the module after the manual
https://docs.google.com/fileview?id...DUtZTkyOS00YjU2LTk1ZDEtZjZiYjg1OTNjMzdm&hl=en
from buildanamp.com
It says
68R for R2 and R4 (in series with the shunt ldr)
1K for R1 and R3 (in series with the series ldr)
This is what I am using. Now with a 100k dual log pot.
So increasing the series resistance will be tested..
I made a typo in an earlier post. I said 68r for series. 1k for series is correct.
Sincerely
I will try increasing the series resistors.
I built the module after the manual
https://docs.google.com/fileview?id...DUtZTkyOS00YjU2LTk1ZDEtZjZiYjg1OTNjMzdm&hl=en
from buildanamp.com
It says
68R for R2 and R4 (in series with the shunt ldr)
1K for R1 and R3 (in series with the series ldr)
This is what I am using. Now with a 100k dual log pot.
So increasing the series resistance will be tested..
I made a typo in an earlier post. I said 68r for series. 1k for series is correct.
Sincerely
Last edited:
I know this sounds nuts, but try a bigger knob. I wonder if its just to hard to dial it slowly while the shunts rise quickly in resistance. A larger diameter knob will make it much easier to dial small amounts than a small diameter knob. Now, if you tell me you are using some 2" diameter knob already I wont know what to do. Well, I will have some suggestions such as limiting the shunts ability to raise quickly by paralleling a resistor with the shunt side of your 100k pot. Very odd that a 68,000R resistor and then another 9,000R didnt help much.
Uriah
Uriah
It's not nuts at all. I fitted a larger knob on my Lightspeed for the same reason and it worked a treat.
Slightly off topic but other tricks are (1) to squirt the likes of Deoxit onto the tracks of the pot and work it in a couple of times on successive days and (2) to cover the back and front (i.e. not near the solder joints) with as much BlueTak as you can comfortably fit.
No, neither tweak makes any difference to the sound but they do make that $2 pot feel smoooooth and serrfistykated.
(Squidging Bluetak over a stepped attenuator does make it mechanically quieter.)