Simple Attenuators - Design And Testing

Joined
Jan 9, 2021
Messages
16
Reaction score
7
okay, did the test with pleasing results.

I used an Arduino due as function gen, not ideal but hey. Added a little circuit: dc filter cap, bleed resistor, rc filter and then a voltage divider. The rc filter is to make the sinus a bit cleaner. I found a little Arduino script with a function gen and controlled it with a serial terminal on my mac.
cruel function gen.jpg

I dragged everything to my little workshop, so nobody in the house has to listen to my loud beeps.

test setup.jpg

I used the scope to see there is no distortion. And 2 voltmeters to measure input and output. I measured at 3 frequencies, 500Hz, 1kHz and 4kHz. I had the readjust the output voltage (AC) from the amp at each frequency and stage / switch change on the attenuator. (In order to compare -7dB and -31.5dB). It is really attenuating a lot. The scope is on 20V/div on both channels:
both 20Vperdiv.jpg

Now "zoomed in" on 1V/div on second (speaker) channel:
scope 20V and 1 VperDiv.jpg
I also used this:
earprotection.jpg

As always, measurements, in practice things turn out a but different, but close enough to verify it functions as designed:
Naamloos.jpg

1/17 is about 0,059. Note that the output voltage of the amp is not as constant with different settings and the meter was jumping around a bit. I very very very tiny turn (tapped it) on the volume knob changed a lot. So, not very precise, but enough to tell verify it is in the ballpark
 

JohnH

Well-Known Member
Joined
May 31, 2008
Messages
5,050
Reaction score
3,268
Location
Wilton NSW
hi @Gert-Jan van der Heiden

That's perfect! Even the figure of 1/17 that I wrote, is just a simplification and the actual value is 1/16.78 which is 0.060. So that's exactly your value at 4000hz and the others are within a decibel which is still a great result over the range.

Some small variations like that are explained by differences in the amps and the different output impedances they have at various frequencies, whereas my numbers are based on a constant value.

All good!
 

RedHouse59

New Member
Joined
May 12, 2022
Messages
6
Reaction score
0
Hey JohnH, very sorry to be "one of those guys" (who skipped over stuff) but I made it to page 77 of this thread and decided it might just be better to ask a question.

I'm building your M2 version (16-Ohm input type with single inductor) as that would suit most of my gear.

I ordered enough parts to build 2 attenuators and for the second one I want to use it with a unique speaker cab I have that is 4-Ohms, it's an "Iso-Cab" (with internal mic) used as DI box.

My question is what values of resistors would I need to use/change in your basic M2/16-Ohm Input schem to operate with 4-Ohm speaker out as the norm?
 
Last edited:

JohnH

Well-Known Member
Joined
May 31, 2008
Messages
5,050
Reaction score
3,268
Location
Wilton NSW
Hey JohnH, very sorry to be "one of those guys" (who skipped over stuff) but I made it to page 77 of this thread and decided it might just be better to ask a question.

I'm building your M2 version (16-Ohm input type with single inductor) as that would suit most of my gear.

I ordered enough parts to build 2 attenuators and for the second one I want to use it with a unique speaker cab I have that is 4-Ohms, it's an "Iso-Cab" (with internal mic) used as DI box.

My question is what values of resistors would I need to use/change in your basic M2/16-Ohm Input schem to operate with 4-Ohm speaker out as the norm?

Hi Redhouse59, thanks for joining our thread.

That's a good question to ask, which has not come up before.

There's a few ways to use the attenuator to convert between ohms values between amp and cab. It all can work simply enough if you have the first -7db reactive Stage 1 always on, without adding a bypass switch (ie as on the base diagram). In this case, the amp mostly sees Stage 1.

Your amps are 16 Ohm, so that's what the attenuator needs to show to the amp, within a small margin

If you just use the basic 16 Ohm M2, into a 16 Ohm amp and a 4 Ohm cab, its actually safe, The amp sees somewhere between 14 and 18 Ohms in all settings. What diverts a bit is the tone, which will become more prominent in the high presence treble and bass resonance. This may be OK, or desirable, or not so. The difference is very much the same as changing from an amp with negative feedback, to one with no NFB. It comes from a 4 Ohm speaker expecting to be fed from an output with 1/4 of the impedance of that of a similar 16 ohm speaker.

Generally we have discussed this difference across a 1 step change, such as 16 to 8 or 8 to 4. At this change, the tone doesn't change much and is usually acceptable. But it can be adjusted for. From 16 to 4, there will be a bigger difference.

Id suggest to keep everything the same as the basic 16 Ohm version, but add a switch with a resistor across the speaker output. So, one end of the resistor goes to the 'cold' output the other to the switch, and the switch to the 'hot' output. Id suggest a value of 15 Ohm. For a 50W amp, it can be like the other small resistors, ie 25W rating, or even 10W. This will lower the apparent output impedance of the attenuator, correcting the speaker tone, but the amp will hardly know.

You could hard-wire it instead, but if you have it switched, you can try the different tones and see which sounds best..

One thing to note. When connected this way with the 16 to 4 conversion , all the attenuation levels are increased, by about -5db. So the Stage 1 value changes from -7 db to -12 db. Id expect that you'll still have plenty of volume available to record an Iso cab though.
 
Last edited:

RedHouse59

New Member
Joined
May 12, 2022
Messages
6
Reaction score
0
Hi Redhouse59, thanks for joining our thread.

That's a good question to ask, which has not come up before.

There's a few ways to use the attenuator to convert between ohms values between amp and cab. It all can work simply enough if you have the first -7db reactive Stage 1 always on, without adding a bypass switch (ie as on the base diagram). In this case, the amp mostly sees Stage 1.

Your amps are 16 Ohm, so that's what the attenuator needs to show to the amp, within a small margin

If you just use the basic 16 Ohm M2, into a 16 Ohm amp and a 4 Ohm cab, its actually safe, The amp sees somewhere between 14 and 18 Ohms in all settings. What diverts a bit is the tone, which will become more prominent in the high presence treble and bass resonance. This may be OK, or desirable, or not so. The difference is very much the same as changing from an amp with negative feedback, to one with no NFB. It comes from a 4 Ohm speaker expecting to be fed from an output with 1/4 of the impedance of that of a similar 16 ohm speaker.

Generally we have discussed this difference across a 1 step change, such as 16 to 8 or 8 to 4. At this change, the tone doesn't change much and is usually acceptable. But it can be adjusted for. From 16 to 4, there will be a bigger difference.

Id suggest to keep everything the same as the basic 16 Ohm version, but add a switch with a resistor across the speaker output. So, one end of the resistor goes to the 'cold' output the other to the switch, and the switch to the 'hot' output. Id suggest a value of 15 Ohm. For a 50W amp, it can be like the other small resistors, ie 25W rating, or even 10W. This will lower the apparent output impedance of the attenuator, correcting the speaker tone, but the amp will hardly know.

You could hard-wire it instead, but if you have it switched, you can try the different tones and see which sounds best..

One thing to note. When connected this way with the 16 to 4 conversion , all the attenuation levels are increased, by about -5db. So the Stage 1 value changes from -7 db to -12 db. Id expect that you'll still have plenty of volume available to record an Iso cab though.

Perfect, ...thank you John.
 

RedHouse59

New Member
Joined
May 12, 2022
Messages
6
Reaction score
0
Just to confirm, ...like this?:
(the added resistor and switch in GRAY color)

M2_16-Ohm_4-Ohm.jpg
 
Last edited:

Gene Ballzz

Well-Known Member
Joined
Feb 19, 2013
Messages
3,684
Reaction score
4,297
Location
Las Vegas, NV
@JohnH & All,

I'm looking for some opinions on some standardization for packaging/layout, as I'm getting ready to build a few of these attenuators. I'm using the Hammond 1590DE boxes at a little less than 8" X a little less than 5". I'm curious as to what most folks see as most convenient for the switches and jacks. Are they better along the short sides or the long sides, as per the layout drawings I provided in post #2,207 on page #111? Obviously the switches and jacks will be opposite each other. They will fit well either way. I'm kinda leaning in one direction, but I'll wait to share that, until after I get a few opinions.

TIA,
Gene
 
Joined
Jan 9, 2021
Messages
16
Reaction score
7
@Gene Ballzz The box is nice and spacious. I think the switches look best on the long side. I also prefer to align the switches to the right, looks better somehow. But that is taste of course. The advantage of this orientation is that the attenuator may fit better on your amp (or not). Just put the box on your amps first.

I don’t like the little solder logs on the switches on the post. I used 1,5mm2 wire, maybe a little bit to much… But anyway, you may want to use thick wire. I=sqrt(P/R). For a 50w / 16ohm version that’s about 1.8A. After stage 1 the amperage through the wires has dropped.

I didn’t use those isolated shrinking thingies. In my country there called shrinking socks (if I translate literally). The reason is that it is easier to service. But at the switches I used these connectors, to prevent shorts, as they are close to the resistors. Also easier to take apart.

The coil is flat. I placed it up right, because of space, but also interference with the amp. But with the metal box that idea might be a little over the top. (Faraday).

I also drawn the schematics on a paper with the values and resistor refs. I numbered the “nets” for easy reference. I then made a wiring diagram and added those net list names to the diagram. Made it easier for me.

I’ve drawn a wiring diagram for the one you pointed out wiring.jpg
I drawn it to get an idea of the wring. The pinkish coloured wires go to the switches. Looks pretty tidy this layout of conponents.

Hope this is to some use for you.
 

JohnH

Well-Known Member
Joined
May 31, 2008
Messages
5,050
Reaction score
3,268
Location
Wilton NSW
I also favour putting switches and jacks on the long faces. A more spacious panel layout. Also, the unit is then narrower front to back, so they sit on my combos better on the rear side of the top handles without resting on the handle.
 

junk notes

Well-Known Member
Joined
Dec 10, 2018
Messages
4,040
Reaction score
5,563
Thanks to your one-man-marketing-machine @Gene Ballzz I am intrigued to build your attenuator design.
As everyone else, I do like having the option of two channels as on some reamping units.
I recently made the decision to cut off the straight ends off one end of the speaker cable and solder on 90s. I was temporarily using 90 adapters for a trial run.
Less gravity pulling down the back of my attenuators.
 

Browneyesound

Active Member
Joined
Mar 22, 2022
Messages
121
Reaction score
207
I was thinking of making one that has switches and switched output jacks at each attenuation stage so a cable and box with a remote foot switch can be plugged into any of the attenuation stages to get a remote volume attenuation at the desired level. Is this doable?

The other question I had was there was a post a while back talking about the need to have the input (?) jack grounded to prevent oscillation from occurring when touching the unit. Is this correct? And all the other jacks should be not connected to the chassis?
 

JohnH

Well-Known Member
Joined
May 31, 2008
Messages
5,050
Reaction score
3,268
Location
Wilton NSW
hi @Browneyesound
IMO the best way to add a footswitch is via one TRS jack. You make a three cored cable out of 5A main flex.


That one has a single value of attenuation in the foot unit. A couple of guys here found that -3.5 db is a good value for that. But another switch or two could be added at the foot unit to change its value.

On the jack grounding, for basic attenuator use, I think all jacks should be Cliff type fully insulated. So the case is not connected to anything. If there's a line-out, then its possible thst the case could then be grounded to the line out ground, not definite, it'd depend on the overall rig set up.
 

Gene Ballzz

Well-Known Member
Joined
Feb 19, 2013
Messages
3,684
Reaction score
4,297
Location
Las Vegas, NV
@Gene Ballzz The box is nice and spacious. I think the switches look best on the long side. I also prefer to align the switches to the right, looks better somehow. But that is taste of course. The advantage of this orientation is that the attenuator may fit better on your amp (or not). Just put the box on your amps first.

I don’t like the little solder logs on the switches on the post. I used 1,5mm2 wire, maybe a little bit to much… But anyway, you may want to use thick wire. I=sqrt(P/R). For a 50w / 16ohm version that’s about 1.8A. After stage 1 the amperage through the wires has dropped.

I didn’t use those isolated shrinking thingies. In my country there called shrinking socks (if I translate literally). The reason is that it is easier to service. But at the switches I used these connectors, to prevent shorts, as they are close to the resistors. Also easier to take apart.

The coil is flat. I placed it up right, because of space, but also interference with the amp. But with the metal box that idea might be a little over the top. (Faraday).

I also drawn the schematics on a paper with the values and resistor refs. I numbered the “nets” for easy reference. I then made a wiring diagram and added those net list names to the diagram. Made it easier for me.

I’ve drawn a wiring diagram for the one you pointed out View attachment 110780
I drawn it to get an idea of the wring. The pinkish coloured wires go to the switches. Looks pretty tidy this layout of conponents.

Hope this is to some use for you.

Be very careful and attentive when using "right angle" plugs, especially for speaker cables. The Switchcraft 226 series has the sleeve/barrel only "pressed" into what appears to be a die-cast plug body and can become loose, causing intermittent connections! Many other brands are constructed in a similar manner.

226.jpg

I tracked down this malady/anomaly on a few of my guitar/instrument cables after chasing my tail in suspicion of the jacks in my guitars! Annoying for an instrument cable, but VERY BAD for speaker connections!

Upon contacting Switchcraft, they admitted to being aware of the problem and sent me a pile of twenty "New & Improved" plugs that they claimed had corrected the issue! I was still able to "loosen" them, through not very many repeated wiggles and/or flexes!

I'm not saying that we shouldn't use this style of plug, but we should be aware of the possible issues and keep a close eye on them, as they age! I won't even get started on the pitfalls and shortcomings of Switchcraft style "open" jacks and their points of possible connection failure!

Thanks also for the comments on attenuator jack and switch placement.

Just A PSA,
Gene
 

Gene Ballzz

Well-Known Member
Joined
Feb 19, 2013
Messages
3,684
Reaction score
4,297
Location
Las Vegas, NV
hi @Browneyesound
IMO the best way to add a footswitch is via one TRS jack. You make a three cored cable out of 5A main flex.


That one has a single value of attenuation in the foot unit. A couple of guys here found that -3.5 db is a good value for that. But another switch or two could be added at the foot unit to change its value.

On the jack grounding, for basic attenuator use, I think all jacks should be Cliff type fully insulated. So the case is not connected to anything. If there's a line-out, then its possible thst the case could then be grounded to the line out ground, not definite, it'd depend on the overall rig set up.

And @JohnH ,
I too have noticed the anomaly of getting a squeal when passing my hand (or guitar) near my attenuators or even touching them with my hand. While earth grounding may "cure" that specific problem, I'm not savvy enough as an electrical engineer to know or understand any other issues that approach may introduce? I can only guess and/or surmise at what forces may be in play here! I am however, somewhat aware of the premise that a "Faraday Cage" only functions as such when it is properly grounded? It seems that when NOT grounded, it simply acts somewhat as an antenna? There certainly are fair amounts of voltage, current and magnetic fields contained within the boxes!

On the other hand, addressing this issue may be somewhat "picking nits" and a bit similar to telling a doctor that my head hurts after banging it repeatedly into a wall, and the doctor's response nd solution is to simply "Avoid banging my head into a wall!"

Thanks, As Always,
Gene
 
Last edited:

Gene Ballzz

Well-Known Member
Joined
Feb 19, 2013
Messages
3,684
Reaction score
4,297
Location
Las Vegas, NV

JohnH

Well-Known Member
Joined
May 31, 2008
Messages
5,050
Reaction score
3,268
Location
Wilton NSW
And @JohnH ,
I too have noticed the anomaly of getting a squeal when passing my hand (or guitar) near my attenuators or even touching them with my hand. While earth grounding may "cure" that specific problem, I'm not savvy enough as an electrical engineer to know or understand any other issues that approach may introduce? I can only guess and/or surmise at what forces may be in play here! I am however, somewhat aware of the premise that a "Faraday Cage" only functions as such when it is properly grounded? It seems that when NOT grounded, it simply acts somewhat as an antenna? There certainly are fair amounts of voltage, current and magnetic fields contained within the boxes!

On the other hand, addressing this issue may be somewhat "picking nits" and a bit similar to telling a doctor that my head hurts after banging it repeatedly into a wall, and the doctor's response nd solution is to simply "Avoid banging my head into a wall!"

Thanks, As Always,
Gene

Hi Gene, thats a curious one and I hadn't come across it before relative to the attenuators. I just did some tests and I couldn't reproduce it. I was running at max power and gain, touching the strings or not, and touching, or just being close to the attenuator. What sort of setup and settings do you run when it happens?

Im not across all the science of screening etc. But, Ive screened some guitars, and it helps cut out 'spiky' RF buzz from fluoro lights etc, if the screen is grounded, but not the 50 or 60hz hum. Ive read that to suppress audio frequencies by screening needs a 'battleship' thickness of grounded metal!. Also, Ive been working with the notion that the 'cold' lead that comes out of the amp for the speaker, may or may not be a ground, and I assume its probably not. What's your experience of that based on circuits that you've worked with?

The whole output of the amp, through unshielded speaker leads is unscreened anyway. But the part in the attenuator that may create the most field would be the coil., as would the speaker. On mine, its down at the base of the attenuator.
 

Gene Ballzz

Well-Known Member
Joined
Feb 19, 2013
Messages
3,684
Reaction score
4,297
Location
Las Vegas, NV
Hi Gene, thats a curious one and I hadn't come across it before relative to the attenuators. I just did some tests and I couldn't reproduce it. I was running at max power and gain, touching the strings or not, and touching, or just being close to the attenuator. What sort of setup and settings do you run when it happens?

Im not across all the science of screening etc. But, Ive screened some guitars, and it helps cut out 'spiky' RF buzz from fluoro lights etc, if the screen is grounded, but not the 50 or 60hz hum. Ive read that to suppress audio frequencies by screening needs a 'battleship' thickness of grounded metal!. Also, Ive been working with the notion that the 'cold' lead that comes out of the amp for the speaker, may or may not be a ground, and I assume its probably not. What's your experience of that based on circuits that you've worked with?

The whole output of the amp, through unshielded speaker leads is unscreened anyway. But the part in the attenuator that may create the most field would be the coil., as would the speaker. On mine, its down at the base of the attenuator.

All the ones I've built for myself so far, have been M-Lite models, with both coils mounted on the top. My current build is an 8Ω M2. I'll let you know if it exhibits the same anomaly.

As for screening, I know that shielding guitar cavities does absolutely nothing until it is grounded. Additionally, the shielding plate or screen covering the open part, component side of an amplifier chassis doesn't do it's job unless making proper contact with the actual amp chassis. While I don't know much of the technical details, I do know the above through experience and experimentation. I do know that some amps have both the +positive & -negative speaker output leads sort of "floating" and mostly isolated from the rest of the circuit while other amps connect the -negative lead directly to amp chassis ground. In either case, I'm not sure that I would want the speaker output -negative doing double duty for providing a shield ground! I'm fairly certain though, that providing a chassis ground lug on the amp and a similar lug on the attenuator housing (given that the attenuator circuit electrically "floats" inside the the housing) and connecting them with croc clips and wire would not cause any issue and may fix the mentioned issue. Or not! I'll do some tests and let you know!
Thanks Again,
Gene
 

RedHouse59

New Member
Joined
May 12, 2022
Messages
6
Reaction score
0
... Also, Ive been working with the notion that the 'cold' lead that comes out of the amp for the speaker, may or may not be a ground, and I assume its probably not....

Most of the "usual suspects" of old style tube amps (Fender, Marshall, etc) the OT's secondary side "common" lead/terminal is grounded, to both circuit and chassis ground.

Attaching a couple of schematics for refence.
 

Attachments

  • Champ_5F1.jpg
    Champ_5F1.jpg
    103.1 KB · Views: 6
  • Deluxe_5E3.jpg
    Deluxe_5E3.jpg
    197.5 KB · Views: 7
  • Marshall_1959SL_Page_1.jpg
    Marshall_1959SL_Page_1.jpg
    163.3 KB · Views: 7
  • Marshall_1987.jpg
    Marshall_1987.jpg
    579.3 KB · Views: 7
  • Marshall_JCM800.jpg
    Marshall_JCM800.jpg
    286.2 KB · Views: 6
  • Princeton_5F2A.jpg
    Princeton_5F2A.jpg
    111.8 KB · Views: 6
  • TwinReverb.jpg
    TwinReverb.jpg
    600.1 KB · Views: 6
Last edited:

Latest posts



Top