Valab 23 step attenuator musings

[JamieMcC]

Valab 23 step attenuator musings

BHC+Speedball+Valab

Instillation

Firstly although this should be a fairly straight forward modification if you are doing this after you have built your Crack + Speedball it requires a bit more work than if you had installed it at the start of your build. Because the valab is quiet tall and the connections are located on its top I had to disconnect and replace all of the connecting wires with longer ones.

This meant un-mounting the small speedball boards (but not disconnecting the wiring) to gain enough access to the octal pins at A7 and A2.  Space is restricted, de-soldering and reattaching longer wires is a fiddly and time consuming task. I was very pleased to have an illuminated magnifying glass to check clearances and solder joints with as the speedball boards make it a bit darker and restrict your angle of view.

Fitting of the valab 23 step attenuator is quiet straight forward the standard volume pots hole needs to be enlarged by about 1-2mm. The smaller locating pin hole requires elongating slightly a dremel made quick work of these mods.
I initially thought the Valab would not fit but it did there is only just enough clearance for the Valab and it only fitted because the smaller A/B board can slot between the  small gap between the Valabs upper and lower banks of resistors (see pic).

Initial impressions with a few hours listening.

Tubes, Mullard 12au7 and Tung-sol 5998

Most noticeable is an improvement in clarity particularly to the mids and top end.  Music seems to have more of a crispness and openness to it.

Placement within the soundstage seems a little more precise. Instruments have a little more attack and bite to the leading edges of notes.

I can hear some woodiness texture to certain wind instruments I had not noticed before.

Listening as I type to Fleetwood Mac, Lindsey Buckingham

 

[Chris65]

Nice report!   Does it still fit inside the standard wooden case with the Valab fitted?

 

[JamieMcC]

Yes it still fits.

 

[mcandmar]

I like the little PCB's on that version, mine didn't come with those just a bag of resistors which kept me busy for an evening assembling it.

Interesting to hear your before/after impressions as i installed mine during the initial build so cant speak of any sonic improvement, i just love that click click click feel to it and being able to set exactly the same volume level every time.  In fact i'm so used to it now i could replace it with a three position switch

 

[Chris65]

Yes it still fits.

OK, thanks. Just considering it for my Crack (I have the Valab). Cheers.

 

[NightFlight]

Based on your photos, I don't think the KHOZMO 48 ladder I want to try is going to fit without some serious modification. 

 

[JamieMcC]

Based on your photos, I don't think the KHOZMO 48 ladder I want to try is going to fit without some serious modification. 

Wow what a monster that its

 

[Natural Sound]

Based on your photos, I don't think the KHOZMO 48 ladder I want to try is going to fit without some serious modification. 

No, but I see no reason why the Goldpoint next to it wont work.

 

[Grainger49]

Based on your photos, I don't think the KHOZMO 48 ladder I want to try is going to fit without some serious modification. 

That is one big MF! ! !

 

[grausch]

I know this is an old thread, but I thought I would post here, rather than start a new thread, since this does relate to the Valab attentuator. I know this is an upgrade, and I should not expect any assistance, but I am hoping someone that has installed the attentuator may be able to shed some light on its workings. Also, excuse the ignorance, but the Crack was my first ever exposure to an electrical circuit.

I have recently ordered 2, one to be installed in the Crack and the other for my soon-to-arrive Stereomour. Unfortunately, the PCBs are mounted upside down as can be seen in the picture.

Thus, I needed to figure out how to connect this to the Crack. After some internet sleuthing and tracing the wire paths, I have determined that my inputs and outputs go to the top and bottom holes while I need to ensure that the left and right channels stay separated. Since resistors have no polarity I am assuming that I can swap inputs and outputs.

However, what throws me off, is the grounding of the attentuator. Based on what I saw on the internet, the grounds are the two center holes. However, there are no conductive tracks from the ground holes to anywhere. Should there not be conductive tracks from the ground hole to something? The stock pot has some conductive tracks on the ground lugs which I assume will be grounded to the chassis itself via the pot. How would grounding work in this case?

I am also attaching a sideways shot of the attentuator.

The white-and-yellow wire leads to a "solder-spot" marked GND. It only connects to the right side and the left has nothing connecting to it. If I connect my ground wires to the indicated holes in the PCB, they will never reach the solder spot marked GND since the are no conductive tracks. Also, it does not look like this wire will ever connect to the chassis, so what is the point of it? I mean, the top of the wire just ends in a solder joint and I don't think it grounds to anything.

In this case, should I even connect my ground wires to the attentuator itself? Can't I just connect the black wires (1 from the RCAs, 1 from the headphone jack and 1 from T3) directly? Would this not have the exact same effect?

I would be grateful for any input.

Cheers,

 

[mcandmar]

Those two Black wires are the Inputs, two White wires are the outputs, and the White/Yellow is one of the grounds.  You should be able to buzz them out with a meter and figure out which holes on the top pcb they are connecting to.

If you look at the edge of each PCB it creates a ring joining all the resistors together, this is the ground.  It is important that the grounds on each side of the attenuator are joined together, all the white/yellow is doing is connecting the ground from the bottom half to the top PCB, but make sure you also join it to the ground ring on the top PCB.  The grounds must also connect to the chassis ground in the Crack too (equivalent of the center pin on stock pot).

 

[grausch]

Thank you for the reply. Figuring out the input /output was easy, although my one white is an input, the other is an output and the same for the blacks.

What had me stumped was the ground. I completely missed that the ring was a conductive track. After doing some resistance checks, I now see that the left ground (as aligned in the picture), connects to the bottom PCB, and the right ground connects to the top.

With the stock pot, the 3 grounds are all connected together (2 on the bottom lug, 1 on the top and then the lugs are joined). Thus, I am assuming that the connections need to be similar.

If my understanding is correct, I need to:
a) connect the two ground holes together since this connects the top and the bottom grounds;
b) connect my 3 black wires to any of the ground holes (also those on the ring of the PCB). At this point, it does not matter where I connect them since the grounds have been joined already in (a).
c) after the above 2 steps, I will have all of my ground wires connected together as per the stock pot;
d) connect the grounds to the chassis ground. I assume the easiest way would be to run a new wire from one of the grounds and connect it to nut at A2/A3. Is my understanding of this correct?

Cheers,

 

[Paul Birkeland]

You absolutely must not swap inputs and outputs.

Put pliers on the switch and turn it all the way down.  Get your meter out and measure between pairs of pads on the switch.  Two pairs should show nearly 0 Ohms between them.  Mark each of those pads with a black dot.

Put pliers on the switch and turn it all the way up.  Now two different pairs of pads will show nearly 0 Ohms between them.  One of those pads will have a black dot on it from the previous step.  Put another dot next to that solder pad.

The pads with one dot are grounds.
The pads with two dots are your outputs.
The pads with no dots are your inputs.

L/R doesn't matter, just make sure the white wires connect to the same set of pads, and the red wires to the other.

-PB

 

[grausch]

Thanks PB. I have done the checks, and they lined up exactly as I expected.

With the pot closed, I read 100k between input and ground and input and output and 0 between ground and output.

With the pot all the way open, I read 99 on the left and 97 on the right between ground and both input and output, with 0 between input and output. Will now be checking all of the steps to ensure that the resistance between input & output decreases on each step.

I now see that there is a difference between the inputs and outputs on the various settings. Does this mean that if I swap input and output that I reverse the attentuator? In that case would turning the attentuator clockwise result in a reduction in volume? Is there something else that would go wrong if input & output were reversed.

Cheers,

Edit: Turns out I should check ground and output for an increase when turning the attentuator clockwise. Measuring input & output did not really work.

 

[Paul Birkeland]

I now see that there is a difference between the inputs and outputs on the various settings. Does this mean that if I swap input and output that I reverse the attentuator? In that case would turning the attentuator clockwise result in a reduction in volume? Is there something else that would go wrong if input & output were reversed.

If you just swap in/out, you would end up with an attenuator that didn't attenuate.

If you swapped input and ground, you could end up with a control that got louder as you turned it counterclockwise. 

-PB

 

[grausch]

Thanks for the explanation. Still not sure exactly how the attentuator works, and I will need to think about this some more. Have found the Goldpoint website with a schematic which helped.

Resistance readings between input and ground show approximately 100k in all positions. This one does occasionally go higher or lower, highest I measured was 102 and lowest 95. I guess this is still within spec.

Ground and output will start off at 0 and then increase to 100k as I increase the steps and input and output will decrease as I increase the steps. However, the sum of these two readings will always be very close to the input and ground reading at each step. Measuring ground and output provided the easiest way of measuring the increases. Trying to measure input and output led to erratic readings, but this was not present on ground and output.

 

[Paul Birkeland]

Resistance readings between input and ground show approximately 100k in all positions. This one does occasionally go higher or lower, highest I measured was 102 and lowest 95. I guess this is still within spec.

Yeah, that's normal.  You can't always get the perfect pair of resistors for each step to give you both an even step size and exactly 100K. Maintaining step size is generally a lot more important than keeping the impedance perfectly constant.

 

[Doc B]

As you turn the attenuator down the signal level gets shunted to ground thru the progressively lower resistance of the path to ground from the wiper. As you turn it up that resistance increases and thus the signal level the grid of the tube sees is bigger since the shunting effect to ground is reduced by the increasingly higher resistance from wiper to ground.

It's kinda like the attenuator is a T valve with an inlet for the signal, an outlet to the tube, and an outlet to the ground. And the signal is a liquid. When you turn the volume down, the signal spills out all over the ground rather than making it to the tube. So the level of signal that makes it to the tube through the other outlet of the valve is lower.

 

[mcandmar]

The blow diagram makes it easy to understand.  For example a five position 100k attenuator would look like,

Rin - Rg
100k - 0k
75k - 25k
50k - 50k
25k - 75k
0k - 100k

 

[grausch]

Thank you all for the explanations,

I was pretty close to grasping this, but the T-Valve explanation makes this a lot easier to grasp. Rather than thinking of only one way for the electricity to flow, I need to realise that it can also flow to ground. As per high school science, electricity chooses the path of least resistance.

Am I right in assuming that since V=IR, as we increase the volume, the tube gets more current and therefore the volume increases? Thus, could the "level of signal" in Doc's post be replaced with current? Also, with the attentuator on anything other than a zero setting, does some of the electricity actually flow to ground, or is it current and resistance merely changed (again V=IR as per the above) and all of it flows to the tube?

Damn, I am now realising the extent of my ignorance when it comes to electricity...