Simple Attenuators - Design And Testing

[stickyfinger]

Should a switch fail could this cause harm to the amplifier? Looks like we would just have a series of resistors after the 1st reactive.

I ask as I can't find full sized rocker switches I would like. These are my best option https://www.mouser.com/ProductDetail/611-CM101J12S205QA but I'm unsure if the quick connect terminals will have enough space to solder.

Also, they are high current but the least life expectance at 10,000 cycles at full load than most other switches.

 

[AtomicRob]

Should a switch fail could this cause harm to the amplifier? Looks like we would just have a series of resistors after the 1st reactive.

I ask as I can't find full sized rocker switches I would like. These are my best option https://www.mouser.com/ProductDetail/611-CM101J12S205QA but I'm unsure if the quick connect terminals will have enough space to solder.

Also, they are high current but the least life expectance at 10,000 cycles at full load than most other switches.

Electrically no problem, you would need at most a 5A rating and even Carling 5A mini toggles are fine for that - I used mini toggles on my 100w 16ohm build. I don't have the calculation sheet in front of me but IIRC worst case those stage switches see around 2A in my build.

The main issue with a rocker is cutting the darn hole. I guess you'd need a nibbler or a dremel, something like that, anyway a lot more effort than just drilling a hole.

 

[JohnH]

hi @stickyfinger

Those switches look like theyd be ok, 16A rated at 125Vac. id think that the number of cycles would be for when actually switching the full power. With the attenuators, although you can switch the stage switches (but not a bypass switch) on the fly, you generally wouldn't be switching at full volume, You'd be switching in a brief pause with little power going through at the moment of change . Plus, we are dealing with far lower voltages than the switch is rated for, so less tendency to arc as contact change.

Other than if you include a bypass switch, the attenuators are designed to always have some constant load through the Stage 1 fixed components and through the series chain of the other stage resistors. So a switch failure, although a cause for disappointment, wouldn't be expected to result in the nasty damaging transients of a full disconnect.

Drilling rocker switches and other rectangular things takes a bit more work compared to a round hole. But it can still be neat since there is a lip around the outside of the switch to cover up workshop sins. Start with some small round holes, then drill them bigger so they meet and then expand into the corners using a file?. Aluminum is quite workable.

 

[stickyfinger]

Yep drilling for rockers are hard but a router and a file will do it.

I'm planning 2 50w builds for 2 15 watt combos. I am hoping to keep the attenuators permanently sitting in the combo hence why I want Rockers as they are a bit more low profile. Hopefully the M2s will stay cool enough in there.

 

[Gingerale08]

But I have another thing I'm doing, which will be relevant, and I've been picking away at it for several weeks. Its a full 100W M4 wiring diagram, based entirely on 50W resistors. Some are in pairs to get the value. But using all the same-size parts makes layout very consistent, but there are quite a few of them! Ill finish it in a few days, its almost done.

Great, looking forward to that! I've been planning to pick up that project at some point so that wiring diagram would be great to have.

 

[Dave_11]

I recently got an SV20 and today noticed a high pitched squeal/chirp when I touch my M2 attenuator with a Tube Screamer pedal turned on. It only does this with the TS9 and no other pedals so far. It only happens when the TS9 is on. Unplugging the guitar makes the pitch lower when I touch the attenuator, and almost sounds like a bird chirping if I tap-tap-tap on it. Since this never happened with my Ori20 I'm guessing it's an idiosyncrasy of the TS9 interacting withe the SV20 and not something I wired up wrong in the M2, but I thought I'd post here and get some input from the experts.

 

[JohnH]

Hi @Dave_11

The only think I can thing of is this:

The way we show the M2 wiring, the case is not connected to anything and the jacks are insulated plastic types, eg Cliff or similar. So like that, there's nothing that's being contacted when you touch the M2 case. But if its built with metal jacks, then one of the speaker wires effectively goes to the case. This might usually be a ground wire and not cause any issues, or it might!

Worth checking into that but its only an unlikely guess. These kind of issues can be tricky to trace, since its probably about very small signals being picked up through the player and the system and fed back through the guitar and gain stages, and it may not even need contact with anything

 

[Dave_11]

Yes, this one has plastic Cliff jacks with plastic nuts. It does have metal switches/screw rings for the stages that make contact with the case, and I first noticed the noise when I went to flip a switch. I just had a look inside through the large bottom vent holes (they do not have the mesh installed) to see if I could see any wires touching the case, and one of the coil wires was possibly touching the case. But aren't those enameled? Anyway, I moved it away and now it no longer makes the noise when I touch the case. I tried moving the coil wire back to touch the side just to see if that was the cause, but I couldn't make it touch so maybe it never was. Oh well, I guess the problem is solved one way or the other, for now.

 

[Dave_11]

Update - it's doing it again. Also, I can take a guitar cable, plug one end directly into the SV20 amp, then bring the unplugged end into close proximity to the attenuator case and it makes the same noise. The closer the jack tip gets to the case, the louder the noise gets.

 

[diego_cl]

Update - it's doing it again. Also, I can take a guitar cable, plug one end directly into the SV20 amp, then bring the unplugged end into close proximity to the attenuator case and it makes the same noise. The closer the jack tip gets to the case, the louder the noise gets.

Did you check your amp ground connection?

The reverb channel of my deluxe reverb had a noise that went away while touching the amp panel, plug body or some pedal cases.
After spending a lot of time rolling preamp tubes and tapping them with a pencil, I realized that my step-up converter ground ring was misassembled.

 

[JohnH]

Hi @Dave_11

It could be a feedback loop from the sensitive high-impedance amp input, through ghe gain of the amp and pedal and then picking up from the field around the attenuator coil? If so , there maybe nothing wrong, but you could explore the effect so you know how to place things to avoid it.

 

[JohnH]

M4 - 100W Construction layout

What if you wanted 'one with everything'? at 100W power? What might that look like?

Maybe like this. It's exploring a few ideas for how to put it together, just to see what it might look like:



Its a pretty-much to-scale drawing, starting with one of the larger Hammond die-cast cases, a 1550N which is 250 x 250 x 97 mm.

I'm showing big switches, 15A rating to handle high power, and Cliff jacks.

I'm trying a layout with all consistently size resistors, at 50W each. So a number of them are doubled in parallel to combine power ratings. The idea of this is to see if it makes a neat consistent layout, plus smaller parts are more efficient at dissipating heat than larger ones. By laying them out on a skew as shown above, then the actual drilling layout is all parallel to the case on account of the diagonally arranged tabs on the resistors. The component sizes are based on Arcol resistors.

Its got most of the M4 feature set, including:

3.5 / 7.5 dB switch at Stage 1
Bypass switch - on the back
Line-out on the front, along with stage switches
Output jacks for 4, 8 and 16 Ohms - basing this on an 8 Ohm build
Fan cooling.
Bass resonance circuit, sized to deal with a 100W build

See page 158 for more on the M4 design. I didn't include the footswitch here for this high-power version.

Having drawn this out, it's hopefully not too hard to strip it back to say a 100W M2, or a 50W M4, or something in-between.

I've put an applicable version of the schematic in the corner, showing which resistors are doubled. eg R1A on the diagram, is in two parts R1A (1) and R1A (2) on the layout. The combined component values should match the M4 design on page 158, but where doubled, as near as possible to 2x the value is used for each half so the net value is close. (Typically, two of the parts for a 16 Ohm version in parallel, will be right for an 8 ohm unit built this way.)

I drew the wires with an emphasis on trying to be clear about what they connect to and what they don't connect to, without covering labels etc. So a real build could be neater so long as the same things are connected.

I coloured the resistors to match with the shading in each attenuation stage of the schemstic diagram. All wires that are connected together are coloured the same.

The case is upside down, with the lid forming the base, and all resistors on top. The reactive components are on the base. See previous comments on fixing parts, spacers etc. There are only a few wires going from base to top, so its easy to open up the case to work on it.

I show the switches rotated flat so you can see them, with the intent that for the Stage switches and the bypass, the 'louder' position has the switch lever upwards in use. This corresponds with most toggle-switches where lever 'up' connects common lugs 'down', though not all switch models do this.

Drilling for vents would be in the base, for the fan, and along the upper side edges. It also needs feet to raise the base.

I hope that's of interest. It's a bit of a beast! and obviously not built yet though all the circuit design is tested.

 

[jesone]

I guess you’d call that one the Zen version

 

[JohnH]

"make me One with Everything", as the Dalia Lama said to the waiter in the burger restaurant.

 

[diego_cl]

I got some inexpensive components to build the active cooling circuit for a M2 100W:

12V PSU


FAN Controller


Thermometer


Arctic P12 (12cm silent fan)


Bimetal thermostat



To ensure that the amplifier is always working with the cooling circuit energized, I will supply power from the attenuator to the amplifier with an IEC C13-C14 cable.

I will use a normally open bimetal thermostat set at 40°C, in physical contact with R1, so that the thermometer and the fan controller are energized only when the attenuator is working. I have some experience with Arctic P12 fans, and I really like them for their low noise levels and reasonable prices

I will use another normally closed bimetal thermostat set at 150°C, also next to R1, to cut off power to the amplifier in case the attenuator gets too hot.

All of this is highly experimental, and I have no experience with bimetal thermostats or the inexpensive fan controller I purchased. However, I hope it will work as intended. I'm unsure whether 150°C is an appropriate value for an emergency power cut, considering that the thermostat will be in contact with R1.





I hope to start building it in the next few days. Cheers

 

[What?]

What if you wanted 'one with everything'? at 100W power? What might that look like?

Maybe like this. It's exploring a few ideas for how to put it together, just to see what it might look like:

Looks like a monster. I bet it woiuld take a minute to put together. How would it do with a 2 ohm output and load?

 

[chocol8]

Looks like a monster. I bet it woiuld take a minute to put together. How would it do with a 2 ohm output and load?

The physical size of an L2 inductor for a 2 ohm version might be a problem. You might need to go to a bigger box or leave off the bass resonance circuit if you can't find the right value, with sufficient current capacity in a small enough size.

What 100 watt 2 ohm amp are you thinking of using it with?

 

[What?]

The physical size of an L2 inductor for a 2 ohm version might be a problem. You might need to go to a bigger box or leave off the bass resonance circuit if you can't find the right value, with sufficient current capacity in a small enough size.

What 100 watt 2 ohm amp are you thinking of using it with?

Not a 100 watt 2 ohm amp, but versatility for a 100 watt 4/8/16 ohm amp down to a 40-watt 2 ohm amp (super reverb).

 

[JohnH]

Hi @What?

That package sounds like it'd be best served by two builds, being a dedicated M2 at 2 Ohms for the Fender, built for up to 50W, and a 100W of M4 or M2 spec, with 8 Ohm input and 4, 8 and 16 Output.

There's a few things taken into account in that:

A 2 Ohm amp is most likely an open-backed Fender. The resonance on an open backed cab (if its 12") is most likely around 80hZ or low E, the last note on the standard-tuned scale (unless you're doing drop D etc). So might as well omit the expensive resonance circuit.

The values for the bass resonance are targeting around 110 to 120 hZ, for closed-back cabs. To make it at 80hZ is even bigger values. Caps for a resonace circuit at 2 Ohms are huge!

But all the designs, M, M2, M3, M4 give the full bass response into a guitar speaker. The bass resonance circuit if provided, is fine-tuning and it can also help the line-out signal. Otherwise It may add a few bass harmonics, but only if you play those low notes. The tone you hear from a guitar cab is generated by the natural response of the cab, whether it's open or closed etc, and that's one of the key features of these designs.

The M2 and M4 designs are dedicated to one input Ohms value, and then the output can get tweaked for x2 or x1/2 Ohms. So an 8 Ohm build can work well with 4, 8 or 16 Ohm cabs, but 2 Ohms is too much of a stretch to keep tone and load within the fairly tight bounds that enable these designs to provide consistent results at all settings. There is an input circuit earlier in the thread for converting input Ohms and it works fine, but it soaks up power and if you convert Ohms both at the input and the output, you can lose a lot of power which could be limiting if just a few dB's reduction is needed.

Most people won't need the bass circuit IMO, and for a line out, a bit of low EQ down-stream can compensate, and the guitar cab out of the attenuator sounds fine anyway.

I think most packages of different amps and cabs may be best addressed by considering two builds, so each is simpler.

 

[stickyfinger]

M4 - 100W Construction layout

What if you wanted 'one with everything'? at 100W power? What might that look like?

Maybe like this. It's exploring a few ideas for how to put it together, just to see what it might look like:

View attachment 140786

Its a pretty-much to-scale drawing, starting with one of the larger Hammond die-cast cases, a 1550N which is 250 x 250 x 97 mm.

I'm showing big switches, 15A rating to handle high power, and Cliff jacks.

I'm trying a layout with all consistently size resistors, at 50W each. So a number of them are doubled in parallel to combine power ratings. The idea of this is to see if it makes a neat consistent layout, plus smaller parts are more efficient at dissipating heat than larger ones. By laying them out on a skew as shown above, then the actual drilling layout is all parallel to the case on account of the diagonally arranged tabs on the resistors. The component sizes are based on Arcol resistors.

Its got most of the M4 feature set, including:

3.5 / 7.5 dB switch at Stage 1
Bypass switch - on the back
Line-out on the front, along with stage switches
Output jacks for 4, 8 and 16 Ohms - basing this on an 8 Ohm build
Fan cooling.
Bass resonance circuit, sized to deal with a 100W build

See page 158 for more on the M4 design. I didn't include the footswitch here for this high-power version.

Having drawn this out, it's hopefully not too hard to strip it back to say a 100W M2, or a 50W M4, or something in-between.

I've put an applicable version of the schematic in the corner, showing which resistors are doubled. eg R1A on the diagram, is in two parts R1A (1) and R1A (2) on the layout. The combined component values should match the M4 design on page 158, but where doubled, as near as possible to 2x the value is used for each half so the net value is close. (Typically, two of the parts for a 16 Ohm version in parallel, will be right for an 8 ohm unit built this way.)

I drew the wires with an emphasis on trying to be clear about what they connect to and what they don't connect to, without covering labels etc. So a real build could be neater so long as the same things are connected.

I coloured the resistors to match with the shading in each attenuation stage of the schemstic diagram. All wires that are connected together are coloured the same.

The case is upside down, with the lid forming the base, and all resistors on top. The reactive components are on the base. See previous comments on fixing parts, spacers etc. There are only a few wires going from base to top, so its easy to open up the case to work on it.

I show the switches rotated flat so you can see them, with the intent that for the Stage switches and the bypass, the 'louder' position has the switch lever upwards in use. This corresponds with most toggle-switches where lever 'up' connects common lugs 'down', though not all switch models do this.

Drilling for vents would be in the base, for the fan, and along the upper side edges. It also needs feet to raise the base.

I hope that's of interest. It's a bit of a beast! and obviously not built yet though all the circuit design is tested.

Looks awesome my only negative I can think of would be a Marshall is 8.25" wide. My rockcrusher is slightly bigger than my Marshalls and the feet barley sit on the head. Cosmetically and structurally if it were around 8 x 12 I say it be perfect.