6k with a 40R shunt gives about 43db of attenuation. Higher Rtot gives greater attenuation range but this is where I like mine set, 6k.
Voltage divider - unloaded or open-circuit
Voltage divider - unloaded or open-circuit
Last edited:
Uriah Hi
I'm pleased to see you endorse getting the LDR's close to the power amp.
I intend putting mine smack bang in between two LM 3886 output stages
that are on separate large heatsinks.
The thinking was getting rid of the set of interconnects completely between pre and power ( well, apart form the very short silver wires that will connect them together internally ) thus shortening the signal path considerably.
There's little or NO excess heat inside my case plus the traffo's and caps are underneath and in another vented case ( both are large 2 U alloy cases )
Maybe I should put additional screening between the LDR's and power amps with some sheets of copper or aluminium to be sure.
Just how sensitive are these things ?
Do I need to go overkill on the heat issue ?
What about the warm sunny days when the room temp rises - we get sunshine here sometimes !!!
Andrew
I'm pleased to see you endorse getting the LDR's close to the power amp.
I intend putting mine smack bang in between two LM 3886 output stages
that are on separate large heatsinks.
The thinking was getting rid of the set of interconnects completely between pre and power ( well, apart form the very short silver wires that will connect them together internally ) thus shortening the signal path considerably.
There's little or NO excess heat inside my case plus the traffo's and caps are underneath and in another vented case ( both are large 2 U alloy cases )
Maybe I should put additional screening between the LDR's and power amps with some sheets of copper or aluminium to be sure.
Just how sensitive are these things ?
Do I need to go overkill on the heat issue ?
What about the warm sunny days when the room temp rises - we get sunshine here sometimes !!!
Andrew
I have to admit to 'only' using my ears to confirm this in my system but I feel the same way. Matched LDRs should change the same amount and there is no chance for them to change resistance in a direction opposite of each other. They will all rise or all fall with temperature changes and it makes sense to me that this change will not be audible. I see no reason not to pot them if you have proved your system works and you want to keep this set of LDRs in there then pot away! Changes in temp will them be very gradual and very uniform and there is no way you would ever notice it.
I think getting them in the amp and removing the extra set of RCAs is a good economical mod and a good sonic mod.
Uriah
All you have to do Stuart is to sent a 500mV 1khz sine wave through a Lightspeed Attenuator that has yet to be potted and watch it on a dual trace scope both left and right channels at equal levels.
Blow on the NSL's (ldr's) and see how much they can go out of calibration, even the heat from my halogen work light can effect them unevenly if too close while doing the calibration stage.
I do not advocate putting them in this environment, there has been some bad posts about "diy" lightspeed's drifting, and this sort of thing should not be encouraged as I will give the Lightspeed Attenuator bad reputation eventually.
Cheers George
So the enclosure of choice should be of large a thermal mass as practical, and be devoid of ventilation? I hope so, that's what I did.
My stupid, cheap Radio Shack Alps pot failed in just a few weeks. I'm going to put in a blue Alps, still <$10.
Stuart
How about that - this device is old enough to be new again.
I designed a tremolo circuit for a guitar amplifier some 40 years ago, when we still used vacuum tubes. This simple circuit used a resistive photocell and a 6 volt lamp. With the lamp hooked up to a simple oscillator, the sound output would vary slightly at the oscillator frequency. Drive the lamp harder and you get more variation. Because of the filiment heating and cool down, you got a smooth transition between states.
I still have a couple of these in a parts cabinet but they're oficially antiques!
Just a couple of points about LED's
The LED is a current device, so varying the current produces a more repetable brightness. LED's will get dimmer over time even at their rated current, about 1/2 the output in about 10 years. Driving LEDs below their rated current will increase their half-life. LED's are intended to be indicators, not light sources. As such, the light output for the same current and voltage will vary slightly between LED's in the same batch, and will vary much much more between batches of LED's.
If you use a second LED / photocell to control the current through the LED's you can self compensate for the loss in brightness over time.
Have fun!
I designed a tremolo circuit for a guitar amplifier some 40 years ago, when we still used vacuum tubes. This simple circuit used a resistive photocell and a 6 volt lamp. With the lamp hooked up to a simple oscillator, the sound output would vary slightly at the oscillator frequency. Drive the lamp harder and you get more variation. Because of the filiment heating and cool down, you got a smooth transition between states.
I still have a couple of these in a parts cabinet but they're oficially antiques!
Just a couple of points about LED's
The LED is a current device, so varying the current produces a more repetable brightness. LED's will get dimmer over time even at their rated current, about 1/2 the output in about 10 years. Driving LEDs below their rated current will increase their half-life. LED's are intended to be indicators, not light sources. As such, the light output for the same current and voltage will vary slightly between LED's in the same batch, and will vary much much more between batches of LED's.
If you use a second LED / photocell to control the current through the LED's you can self compensate for the loss in brightness over time.
Have fun!
Ski shops?
Melted wax is no where near as hot as a soldering iron.
I insist that calibration be done before potting and also after potting when the wax has cooled to room temperature.
You will see how the wax itself when hot effects both channels differently, even though they (ldr's) are all getting heated by the same amount, and once back to room temperature hopefully you will see that they all come back to pre wax calibration. This can take up to 1hr.
Cheers George
Melted wax is no where near as hot as a soldering iron.
I insist that calibration be done before potting and also after potting when the wax has cooled to room temperature.
You will see how the wax itself when hot effects both channels differently, even though they (ldr's) are all getting heated by the same amount, and once back to room temperature hopefully you will see that they all come back to pre wax calibration. This can take up to 1hr.
Cheers George
George Hi
That wasn't a joke
I mean sports shops that sell Ski 's and maybe........erm...wax?
I don't know what I'm .....g talking about here - Candle wax, Ski Wax...
This is almost going off thread
I'll put the whole thing in a small cast aluminium case ( with a lid ) in between the power amp output stages and when it's calibrated fill it to the top with cold two pack polyester resin - if it doesn't work it'll be a posh paper weight.
That wasn't a joke
I mean sports shops that sell Ski 's and maybe........erm...wax?
I don't know what I'm .....g talking about here - Candle wax, Ski Wax...
This is almost going off thread
I'll put the whole thing in a small cast aluminium case ( with a lid ) in between the power amp output stages and when it's calibrated fill it to the top with cold two pack polyester resin - if it doesn't work it'll be a posh paper weight.
I measured the output from my LDR attenuator (LDRs unpotted, using Uriah's board) and found that the balance between channels varied very little over time. A couple of hours after switch-on the balance had changed only by just over 1%. I used a 1kHz test signal from a CD and measured the voltage at the speaker terminals.
The attenuator is sited in the same enclosure as four Hypex UcD180 power amp modules. It gets mildly warm in there as time goes on.
So it seems that, at least in my case, thermal drift is not an issue.
The attenuator is sited in the same enclosure as four Hypex UcD180 power amp modules. It gets mildly warm in there as time goes on.
So it seems that, at least in my case, thermal drift is not an issue.
How about this, Andrew:
ski wax bars items - Get great deals on Sporting Goods items on eBay UK!
Jon
ski wax bars items - Get great deals on Sporting Goods items on eBay UK!
Jon
I measured the output from my LDR attenuator (LDRs unpotted, using Uriah's board) and found that the balance between channels varied very little over time. A couple of hours after switch-on the balance had changed only by just over 1%. I used a 1kHz test signal from a CD and measured the voltage at the speaker terminals.
The attenuator is sited in the same enclosure as four Hypex UcD180 power amp modules. It gets mildly warm in there as time goes on.
So it seems that, at least in my case, thermal drift is not an issue.
I also found this to be the case. The box of the first one I made is open on the bottom with 2 Class A amps running below on the floor. At the left back of the equipment is a heater vent...yeah over here heater set at 65F still comes on even with 2 stereo class A amps dissipating 100W/ch.
With that setup I'd say you'll have definate image drift going on, maybe your just not seeing/hearing it?
They say some gifted ears can hear 1db difference, but most need 3 db to make a judgement call. The ocilloscope does'nt lie.
Cheers George
Nothing of that sort experienced. I have a spot on the top center of my fireplace and Norah Jones is always spot on. The soundstage of another unit I made (no longer a LS)does not collapse even at low levels -it is always centered. LDR match at my typical listening level is less than 1dB. Surely there will be some temp related drift but in my system it is mimimal enough not to shift soundstage, sound level, or any of that sort.
The drift you might be measuring is during the first few minutes after the control pot is varied but usually stabilizes after a while unless you have some noise/ripple/bad regulation from the psu or a crappy pot.
The drift you might be measuring is during the first few minutes after the control pot is varied but usually stabilizes after a while unless you have some noise/ripple/bad regulation from the psu or a crappy pot.
George Hi
No one would doubt your knowledge on this for a second, particularly those
members with no means of testing channel balance and drift properly.
I don't - so as a precaution I'll still give it some respect and do a thorough encapsulation in a small alloy enclosure.
I want the best out of it so it's worth doing properly.
It is interesting though that some don't notice any drift in extremely testing conditions - maybe this is further testimony to the sound quality of your discovery.
Like many, I've done a lot of trawling and ' Googling ' while I wait for mine to arrive from Uriah and there is not one negative comment anywhere on the internet whether DIY vero board versions or serious production units.
How astonishing is that ?
Andrew
No one would doubt your knowledge on this for a second, particularly those
members with no means of testing channel balance and drift properly.
I don't - so as a precaution I'll still give it some respect and do a thorough encapsulation in a small alloy enclosure.
I want the best out of it so it's worth doing properly.
It is interesting though that some don't notice any drift in extremely testing conditions - maybe this is further testimony to the sound quality of your discovery.
Like many, I've done a lot of trawling and ' Googling ' while I wait for mine to arrive from Uriah and there is not one negative comment anywhere on the internet whether DIY vero board versions or serious production units.
How astonishing is that ?
Andrew