L-pad improvements?

[jimboyogi]

Hi all,

I bought a few NOS "Noble" 8 ohm 15W L-pads. My intention was to use them as a quasi 'Master Volume' control in a SE 5W 6V6 build, wired between the OT and the speaker.

But..

They don't work so well. Main issue is that at very high attenuation, they have a very glitchy response between low volume to silent, and do not go completely silent at full CCW rotation. The next issue is that they appear to have a linear taper, so not intuitive as a volume control. They work very well at lower attenuations, but it's at the low volumes that a master is most useful for home/bedroom playing!

I have attached a pic of the measured resistance of the two resistive tracks in the pot below. Ignore the dotted line/cap, these were suggested by online reading for frequency response but I didn't try them.

My question for anyone with experience of L-pads is - is there any extra resistances or other components I can add to these to get a smoother reduction in volume at max attenuation?

[Chuck H]

I had the same glitchy response with my own attenuator design years ago. My solution may not work for you but I'll share anyway. I added a 1 ohm resistor between leg 1 of your Lpad and ground. Obviously this does make it so you can't turn down to zero but it does alleviate the glitchy noise from the wiper on the wire wound element. And...

You may want to employ a bypass switch. Unless some attenuation at full volume isn't a concern.

Sorry, but I don't have a solution for the taper.

[jimboyogi]

Thanks for the suggestion Chuck, I will give it a try. And apologies for the slow response, I posted the question above, and the next day I couldn't get onto the forum until yesterday. In the back of my head was a little voice telling me I had somehow blown up the forum

I understand why the 1 ohm resistor between leg 1 and ground will prevent turning all the way down, but I don't understand why a bypass switch would be necessary? If the L-pad is all the way up, leg 1 resistance becomes infinite as the wiper is lifted from the resistive track by an end stop. So resistance to ground from leg 3 becomes infinite also, and therefore zero attenuation of signal. I think...

Another question - in case of L-pad failure, to make sure that there is always some continuity between OT and speaker (and therefore some continuity to ground to protect OT), should I put a resistor between L-pad legs 3 and 1? say 100 times the speaker resistance so 80-100 ohms?

I have an idea for the taper problem, but I so far haven't been able to find the right part available. If I can find an 8 ohm ~15W STEREO L-pad, then I could wire the two sections in series. Two linear tapers in series like this forms a log taper or close enough as far as I understand. Another benefit would be the sharing of power dissipation between the two sections. Critique of this idea welcomed!

[Chuck H]

I'm sorry if this is offensive... Why do you think there is infinite resistance with the L-pad adjusted full up? This is not represented in the schematic and I don't know of any rheostat that does this. The load to your amp at full up would be the speaker load paralleled by a 35 ohm resistance. Roughly 25% of your amps power is wasted in the L-pad without a bypass switch.

[jimboyogi]

No offence taken at all. And apologies for not making this clear on the schematic. For L-pad leg 1, the wire wound track resistance is 0 - 35 ohms, but at it's end stop, there is a small ramp that lifts the wiper from the track, breaking any continuity. It's pretty clever IMO.

[Chuck H]

No offence taken at all. And apologies for not making this clear on the schematic. For L-pad leg 3, the wire wound track resistance is 0 - 35 ohms, but at it's end stop, there is a small ramp that lifts the wiper from the track, breaking any continuity. It's pretty clever IMO.

Ah. So a sort of built in bypass. I see now that you did indicate an infinity symbol for track 1 at full up. Yes, that's clever. I didn't know they did this.

[g1]

But the resistance from lug 3 to lug 1 is always there.
Not sure if I understand why one of the L-pads measures 35 ohm, and the 'why' of the way it's wired up?

[Chuck H]

But the resistance from lug 3 to lug 1 is always there.
Not sure if I understand why one of the L-pads measures 35 ohm, and the 'why' of the way it's wired up?

The two resistance tracks allow a graduation. As series resistance increases the speaker is shorted across it's terminals. At full attenuation the load is at 35 ohms. This isn't as bad as it seems since actual speakers have a highly variable impedance relative to frequency and are only at their specified ohms in a narow range. Being higher otherwise. So the "average" ohmage relative to bandwidth is at a higher ohmage than the speaker specification. A lot higher. Like in the 20 to over a hundred ohms range at frequencies other than where it's idealized somewhere in the mid frequencies. In fact the TrainWreck Air Brake has a high attenuated load in the thirties and players seem to like those quite well.

My own attenuator graduates similar except that it's played off an artificial load one side and the actual speaker on the other. So the ohm differential is lower. IIRC across it's range the ohmage is between 6.75 ohms and about 10 ohms. But that's the gross measure at resonance and actual impedance at frequency is still variable. With the overall design being sort of half resistive. My point is...

It's almost certainly fine to run an amp rated for 8 ohms into a purely resistive 35 ohm load. I don't have the tech chops to interpret why this may, or may not be an issue though. But "averaging" the actual, across the spectrum load for attenuation purposes is pretty common in several attenuators.