Simple Attenuators - Design And Testing

[LesPaul_Marshall_Rocker78]

Hi again, Thanks for the Welcome!!
Today I HAD to find out something!!?
(As a starting point for further Reference...)
- - - - - -

Soo I pluged in the (NMB plexi) 100watt SuperLead 1959 clone(Bugera 1960 Infinum) **I COMPLETLY DISENGADED THE PPIMV MASTER** ( =Should be Bonkers LOOOUD....!!!!!)

My Cab of choice for these Attenuation/volume tests :
* old 16-ohm(15-ohm?)original 73' 4x12 "Marshall 1935" Greenback Pulsonic cone 25w 55hz cab (large checkerboard cloth)

* THEN I put the UAD OX-Attenuator/Load in between speaker/amp "to limit the crazy speaker volume"

*Amp settings*
I set the 100w SuperLead amp-controls to ***EVH CRANKED!!!***(All knobs on full)
Presence : 10
Treble : 10
Middle : 10
Bass : 10
Volume -1 (gain) : 10
Volume -2 (gain) : 10
- - - - - -

*UAD OX Load box settings*
For FINAL volume control, I set the OX Attenuator Like this:
First Knob to 1: (Highest Attenuation)
Then Knob to 2: (Second Highest Attenuation)
And last Knob to 3: (Medium Attenuation, "getting" quite loud...)
Then I varied OX knob and amp knobs...

.......*tweaking*......
Then last I tweaked (varied settings) on all knobs *trying to keep the OX* at certain volume levels "as detailed above"

----------------------

Testing Results ...
*With UAD OX-Attenuator/Load knob # 1 :
In my opinion it sounds "too low" when FULLY *CRANKED* (all knobs on 10 including the two Superlead (gain)volumes)

*With UAD OX-Attenuator/Load knob # 2 :
Getting better, Amp breaths a bit better now, and sounds less congested.. but still far away from ?Loud? and almost "bedrom friendly" IMO ?

*With UAD OX-Attenuator/Load knob # 3 :
Waay better, starts to sound like a proper TUBE(valve)-AMP now... Amp-EQ is also starting to be more useable/responsive!
The extreme exessive Bass at low volumes and loss of highend/precence is going away a little now..


**** OX Attenuator knob at #3 ****
FULLY **CRANKED Amp (all knobs at 10, including Amp Volume 1 & 2 )**
AT OX-knob #3(medium Attenuation) is becoming Loud TV/very Loud bedroom ??!
"I STOPPED" (I don't need gigging/consert levels LOL) at this "Louder setting" and tweaked madly on all possible settings back and forth between the UAD OX Attenuator knob settings 1-3


- - - - - - - - - - -
MY CONCLUSION :
Amp does not NEED position 1
At Max Attenuation it needs 2 perhaps??
I think I could get away with an Attenuator that attenuates down to a MAXIMUM 2 - 2.5 ??? On the OX Load.. when it is FULLY CRANKED....

PS ! PS ! *Rabbit hole caveat* your ears close to the 412(as with a "on axis" micked up amp) sounds wayyy better even at lower attenuations thats a bit annoying hmmmm....




-----------------------------------------------
Regards
"LesPaul Rocker"

 

[Blaoskaak]

The coil looks fine

The switch is hard for me to read about, but there's an English data sheet there which says it is a 3-position switch. It's described as on-off-on, which means that in the centre position, nothing is connected, which isn't what we want. We need a 2position switch on-on There will be a version that does that, since it's a simpler design

Yeah, now i see it. On the page from the switch it says on/on. In the documentation on/off/on. Can i still use it when i dont attach middle or use only upper and middle? Otherwise i have to order some new one. That isn't a big problem too.

I'm glad the coil is the right one!

 

[matttornado]

hi @matttornado , great to hear from you again

Yes you can run the 8 ohm cab out of the 16 ohm attenuator, and you'll still plug into the 16 ohm amp jack . Of course, if you have a full bypass switch, you'd best not engage it unless you reset the amp to match the cab

awesome thanks!

 

[JohnH]

Yeah, now i see it. On the page from the switch it says on/on. In the documentation on/off/on. Can i still use it when i dont attach middle or use only upper and middle? Otherwise i have to order some new one. That isn't a big problem too.

I'm glad the coil is the right one!

Definitely don't use on-off-on switches for this. The disconnected middle position can cause nasty transients.

 

[Blaoskaak]

Thanks alot!

 

[JohnH]

hi @LesPaul_Marshall_Rocker78 welcome to our thread.

If i was you with what you have i would do as i did and build what we do!

But, for a commercial unit in the form of a passive attenuator, we hear great things about the Ironman designs. but personally, Id love to buy one from Mike Lind. He's a very cool guy who posts on TGP, and while developing his design, he generously shared a lot of very helpful testing data, which was a great help to me. i dont know exactly how his works, not the same as ours here, but it sounds to be right. I think he builds them to order.

But if you are up for a build project, stick around here!

Further on this, I had a read-up on the Ironman II. While it is evidently a great product, I have some questions/concerns:

Here's a page from its manual:


If I've understood it right, you get 3 ranges, solo, hi, and lo. Each separated by 3 db. Within each range, you get three steps of -4db, then two at -10 db. -10db is quite a big step, and to find places in between, it seems like you have to jump back and forth between the three ranges. Seems like a fiddle?

It offers the solo setting as a footswitchable way to get a change of level for a solo boost, even from 0db fully bypassed going down. But this implies switching and unswitching reactive circuitry on the fly, and this can create transients, which can cause damage. So the manual states that the solo and other controls should not be switched unless you damp the strings to stop the signal. This does not seem like a reliable procedure to require when under pressure, launching from loud chugging power chords into a face-melting solo at a gig.

I checked out the impedance curve, measured by Mike Lind on its smaller brother, the Ironman II mini. It doesn't look much like that of a real speaker.


Just some watch-its...

 

[LesPaul_Marshall_Rocker78]

Thanks so much! JohnH I rely on my Marshall brothers for valuable info!

It's just a pity Mike Lind don't have "any" video/recorded test-clips of his attenuator for his showing what it sounds like at all no quick and dirty iphone clips or anything?? at lot of people have waited roughly 2-3years to hear just something...

--------------
I have once again started reading "this forum thread" from the humble-beginning to sort my head/mind a little bit, after my recent testing session with Bugera "Superlead" I became a little bit wiser as what I probably want !!??

If I build "the simplest version" with one stage -14 or -18 desibel "reduction"
say a Resistive 16ohm or 8ohm (prefer 16ohms)
What do you think would be the result compared to feel/sound of the OX ?

At this stage of the Tone-Journey I feel we are "on our own" and probably have to build something.....

I have 1 year back in 1994(we had 3 year education plan for "94") as Electric/electronics education(basic priciples) of 3 year-school and used "the two other years" to do my dream ocupation draughtsman today i am Cad-designer(no engineer though)

I am not an Idiot when it comes to safety, But I would never dare doing a tube-amp(thats why I purchased a built Trinity I considered Metroamp JTM45 kit before that..) I am not a great solderer but I think I can manage ?(But I dont know if I have created a Cold solder or not??) I have a hard time understanding and calculating circuits though .... (Mathmatics is quite my nemesis..)

Question:
Would you guys help me understand it?
I sort of remember the symbols and I even have "2 multimeters"(cheaper ones)..
I ALSO have a quite good "genuine" Japaneese soldering station Hako FX888 70watt (adjustable watts/heat)

Problem is though getting suitable parts for build the attenuator case/box ?

Here in Norway I can get Arcol wire-wound-Resisors(would prefer High Quality build..even if more expensive..(there is Arcols in the OX) quite easily it seems probably some other small parts also, but that case/enclosure though

 

[JohnH]

Hi @LesPaul_Marshall_Rocker78

I reckon you have what you need to do this if you want to. It's actually a simple circuit, with nice big robust parts. There's nothing that wasn't invented at least 100 years ago, using theory from the 19th Century!

Arcol are perfect for the resistors. If you build the pure resistive version, you can expect it to be tonally at least as good or better than other resistive attenuators.

But seriously, Id recommend doing the reactive one based on design M2 (or part of it). It's no more complicated really. You can order the coil that it needs from various places in EU, and there are links in recent pages. Just have to interpret the diagram. You could build one stage, though it can be a bit tricky to decide what level to go to, and in any case multiple settings are useful just for different uses and even different times of day.

Units like OX and other high-end ones pack a heap of added functionality, effects loops, IR loaders, reamping ect. Our design is just an attenuator. It just takes a loud amp and reduces the volume. But, after a couple of years with this design, I reckon it does that with virtually no difference in tone or feel or dynamics . That's based on testing, analysis and listening. I've never compared it to an OX or other similar but I compare it the original sound of the amp and speaker, which is the best comparison really.

For the case you can can buy one. I used a Hammond die-cast one. All it needs is drilling, which with careful marking out isn't too hard to do with care.

Happy to discuss further!

 

[CO_Hoya]

Hi John,

First, thank you very much for starting and maintaining this thread. I've been needing an attenuator for years and I think you've finally motivated me into action. You've been very generous with your time and knowledge.

Also, kudos to the many people here who have posted their projects resulting from this thread, letting the rest of us learn from you.


Beyond purchasing a pile of power resistors via eBay, I'm still in the planning stages, figuring on a 4/8/16 ohm version.

I've put together a proposed circuit for my own version, which I'm calling the M3V - this is the full main reactive load to preserve a bit more of the bass resonance, as well as the ability to switch between inputs via a 4P3T rotary switch (C&K A40305RNZQ).



The C&K rotary is 125VAC-rated for 5A continuous, 2.5A switched. Per Randall Aiken, this *should* be enough in most cases with the understanding that the user would never switch the position while the amp is powered. The unattenuated signal from the amp always passes thru 2 conductors within the switch - after one stage of attenuation, I assume the load would be reduced enough that it could then pass thru a single switch conductor.


If you don't mind, I'd like to ask a couple of questions:
1. Do you see any issues with the proposed circuit? I've also sketched this for a 3P3T slider switch but that circuit is still half-baked and a bit messy, while the rotary switch seems straight-forward.
2. My attenuator would be primarily (exclusively?) for open-backed cabs, 1x12 (e.g. Fender Deluxe Reverb). Is that frequency curve different enough that I should consider some changes to the component values that you spec?


As I'm new here, I've had to scrub out embedded links - hopefully you can find that rotary switch via Google.

 

[JohnH]

hi @CO_Hoya welcome to our forum and to our thread. Thanks for your drawing. It looks like Digikey Schem-it? i use that too.

Looks like you are on the right lines, and a few observations may help.

The bass resonance circuit (C1 and L3 above) is not so important in this design a it is in some others. Only one person i know of has built design M3 with that part and that's @dbishopbliss , who might like to comment further here on the difference it makes. His review on builds of M2 and M3 are on page 89.
If you dont have the bass circuit, you still get the bass peak because in this design, it is generated by the speaker itself. What you might not get is the added bass harmonics as the amp responds non-linearly to the resonance when driven hard. Instead, the bass may be a very slightly cleaner and more solid.

Ive not built M3 myself but I did some tests trying to provoke any difference between the bass tone of the full unattenuated speaker, and the attenuated. i couldn't hear any difference driving very hard with drop-D tuning trying to hit those frequencies on my VM2266c amp. Relevant to your set up, this is also a 2x12 open back combo, and it probably has a resonance at about 80hz instead of about 110hz on a closed-back.

The bass circuit may have more need if you use a line out into an IR system, but you can also probably just add a bit of EQ if needed, and a number of builders here have used M and M2 designs with line-outs.

But if you do have the bass circuit, and if the input ohms options are important, then note that the added front end with L2 does not add a share of the bass resonance. Still it will work fine as described above, but with half of whatever the bass circuit does. We could work out another L and C for it if wanted.

if you want to target open-backed combos, then the bass resonance should be lowered using larger values. i could work something out if you like. Personally, I don't think the bass circuit its really needed since it cant do anything much except to the every lowest notes on the 6th string!

The switch sounds fine and its good that it lists the 5A carry/non-switched rating. I guess most switches would have that added capacity when not operated, but its nice to see it declared. Also of interest is whether its shorting or non-shorting. Shorting would be preferred, if only for the accidental case of moving the switch in use.

On the output side, see the red parts on the M2 design. These correct the tone when 16 ohm cabs are used on the basically 8 ohm circuit. Safe to use without but with a bit less high treble.

Last thing I have for now: the idea of making the switched conversion by adding L2/R12/R13 inherently uses up 50% of the power ie another -3db reduction of max volume level. That may be significant if you need low attenuation settings since he least attenuation is then -10db. On page 87 posted a full double circuit that allows different ohm settings with no such extra loss. You could instead, just consider building two dedicated attenuators. Or, if you need an overall -3db reduction, use it as a load box in parallel with the speaker, and use a lower amp tap.

Good luck with the build!

 

[Jordan Prysmiki]

Hey guys, was there a -1, -7, -7, -14, -3.5 version?

It's the addition of the "-1" that differs from the main Model M2 that I'm after.

 

[JohnH]

Hey guys, was there a -1, -7, -7, -14, -3.5 version?

It's the addition of the "-1" that differs from the main Model M2 that I'm after.

Hi Jordan, we dont have anything drawn with a -1db stage. That's a very small step, so for most purposes it would generally not be needed. Even the -3.5db is quite small, and most players are finding they don't want to subdivide it further. But, if you can describe what you have in mind and why, and what you want to use it for, lets discuss, its probably possible. It's significant as to whether or not you need this stage to run on its own, or only in combination with other stages.

 

[Jordan Prysmiki]

Hey John. Thanks for responding.
OK so it's the -3.5db I'm after! I built mine with the -7,-7,-14 thinking I would not need the -3.5 originally. In wanting to build my second one for another amp and cab I wrongfully remembered a -1db option instead of the -3.5db. So I should be good to go now!

 

[JohnH]

All good! when you have that -3.5db stage, in any of the designs, you can then control the switches to get every even step in small, equal -3.5db increments from min to max.

 

[Blaoskaak]

Who knows where i can get a aircore in europe? I ordered one, but they are scammers.

I searched on ebay, but they don't have 0.9. Can i get 1.0 too?

 

[CO_Hoya]

Hi John,

Thanks for the reply, responses are in-line.

hi @CO_Hoya welcome to our forum and to our thread. Thanks for your drawing. It looks like Digikey Schem-it? i use that too.

I discovered Schem-it thanks to this thread, in spite of being a regular Digikey shopper. It's a bit quirky, but good enough for this sort of work.

Looks like you are on the right lines, and a few observations may help.

The bass resonance circuit (C1 and L3 above) is not so important in this design a it is in some others. Only one person i know of has built design M3 with that part and that's @dbishopbliss , who might like to comment further here on the difference it makes. His review on builds of M2 and M3 are on page 89.
If you dont have the bass circuit, you still get the bass peak because in this design, it is generated by the speaker itself. What you might not get is the added bass harmonics as the amp responds non-linearly to the resonance when driven hard. Instead, the bass may be a very slightly cleaner and more solid.

Ive not built M3 myself but I did some tests trying to provoke any difference between the bass tone of the full unattenuated speaker, and the attenuated. i couldn't hear any difference driving very hard with drop-D tuning trying to hit those frequencies on my VM2266c amp. Relevant to your set up, this is also a 2x12 open back combo, and it probably has a resonance at about 80hz instead of about 110hz on a closed-back.

The bass circuit may have more need if you use a line out into an IR system, but you can also probably just add a bit of EQ if needed, and a number of builders here have used M and M2 designs with line-outs.

Understand on all points above. I have room for one additional toggle in my current layout and so I've been toying with the idea of incorporating the bass circuit and then being able to bypass it (although I may go a different way with that toggle). I figure that I'd only ever build one (or two) of these units, so why not go all out with the "Cadillac" design.

More practically, I'd at least like to design a layout that can accept those additional components down the road if desired.

But if you do have the bass circuit, and if the input ohms options are important, then note that the added front end with L2 does not add a share of the bass resonance. Still it will work fine as described above, but with half of whatever the bass circuit does. We could work out another L and C for it if wanted.

I think having bass resonance on both reactive loads may be overkill, even for me. I've two 8-ohm amps (and one 4-ohm) and so I'd primarily be using the attenuator in the 8-ohm position with the bass circuit designed for that, which would be fine for me.

if you want to target open-backed combos, then the bass resonance should be lowered using larger values. i could work something out if you like. Personally, I don't think the bass circuit its really needed since it cant do anything much except to the every lowest notes on the 6th string!

I think I'd want to target something around 90 Hz. I can work out an LC pair that would give me that resonant frequency (e.g. 10 mH + 300 or 320 uF), but this would provide a different L/C ratio than yours and its not clear to me how important that ratio is.

I'm finding that there is a big jump up in price once you get above 10mH so I'm trying to design around that if possible.

The switch sounds fine and its good that it lists the 5A carry/non-switched rating. I guess most switches would have that added capacity when not operated, but its nice to see it declared. Also of interest is whether its shorting or non-shorting. Shorting would be preferred, if only for the accidental case of moving the switch in use.

Yeah, you got me on that one - that's a non-shorting switch, which isn't great.

I can find only one available version of that switch that's both shorting and has solder lugs (p/n A40303RSZG), and I'm only seeing that from a single vendor with a min qty of 2. Since that version is double the price, that becomes an expensive "upgrade" to the inputs. Alternatively, there is a thru-hole shorting version (p/n A40303RSMCGK) but that would take some delicate soldering skills to use.

On the output side, see the red parts on the M2 design. These correct the tone when 16 ohm cabs are used on the basically 8 ohm circuit. Safe to use without but with a bit less high treble.

Thanks, but I don't have a 16-ohm cab so don't see the need. I did consider building a 2/4/8 ohm version, and then I would use that circuit for the extra output.

Last thing I have for now: the idea of making the switched conversion by adding L2/R12/R13 inherently uses up 50% of the power ie another -3db reduction of max volume level. That may be significant if you need low attenuation settings since he least attenuation is then -10db. On page 87 posted a full double circuit that allows different ohm settings with no such extra loss. You could instead, just consider building two dedicated attenuators. Or, if you need an overall -3db reduction, use it as a load box in parallel with the speaker, and use a lower amp tap.

Honestly, I'm guessing that even -7dB would be too little for my application, which is to get bedroom volumes with 1x12 combos.

Good luck with the build!

Cheers!

 

[JohnH]

Who knows where i can get a aircore in europe? I ordered one, but they are scammers.

I searched on ebay, but they don't have 0.9. Can i get 1.0 too?

The value for the 0.9mH coil is fine in the range about 0.8mH to 1mH with only a very small difference in theory

These people sell 0.82mH and 1.0mH You should get 1mm wire which is also called 18 gage

VIS SP 5013: VISATON SP coil - 0.82 mH - 1.0 mm at reichelt elektronik

 

[JohnH]

Hi John,

Thanks for the reply, responses are in-line.


I discovered Schem-it thanks to this thread, in spite of being a regular Digikey shopper. It's a bit quirky, but good enough for this sort of work.


Understand on all points above. I have room for one additional toggle in my current layout and so I've been toying with the idea of incorporating the bass circuit and then being able to bypass it (although I may go a different way with that toggle). I figure that I'd only ever build one (or two) of these units, so why not go all out with the "Cadillac" design.

More practically, I'd at least like to design a layout that can accept those additional components down the road if desired.


I think having bass resonance on both reactive loads may be overkill, even for me. I've two 8-ohm amps (and one 4-ohm) and so I'd primarily be using the attenuator in the 8-ohm position with the bass circuit designed for that, which would be fine for me.


I think I'd want to target something around 90 Hz. I can work out an LC pair that would give me that resonant frequency (e.g. 10 mH + 300 or 320 uF), but this would provide a different L/C ratio than yours and its not clear to me how important that ratio is.

I'm finding that there is a big jump up in price once you get above 10mH so I'm trying to design around that if possible.


Yeah, you got me on that one - that's a non-shorting switch, which isn't great.

I can find only one available version of that switch that's both shorting and has solder lugs (p/n A40303RSZG), and I'm only seeing that from a single vendor with a min qty of 2. Since that version is double the price, that becomes an expensive "upgrade" to the inputs. Alternatively, there is a thru-hole shorting version (p/n A40303RSMCGK) but that would take some delicate soldering skills to use.


Thanks, but I don't have a 16-ohm cab so don't see the need. I did consider building a 2/4/8 ohm version, and then I would use that circuit for the extra output.


Honestly, I'm guessing that even -7dB would be too little for my application, which is to get bedroom volumes with 1x12 combos.


Cheers!

Ill have a look at values for a 90hz resonance. When I looked at it before, it was getting good bipolar capacitors with high current and low losses was a big cost factor.

 

[CO_Hoya]

Ill have a look at values for a 90hz resonance. When I looked at it before, it was getting good bipolar capacitors with high current and low losses was a big cost factor.

Thanks.

The caps I'm considering are the non-polar electrolytics offered by Parts-Express.com. Those are spec'ed with a max DF = 5%, which gives an ESR = 0.26 ohm at 300 uF / 1 kHz, if I did the math correctly.

Most are out of stock at the moment, but I'm a long way off from purchasing.

On the other hand, bumping up to 12-15mH may not be as expensive as I thought.

 

[aceofbones]

I’m just about ready to wire my M2 up and had a question about wire gauge. I know you’ve said 18awg, but would 20awg solid core be fine? As far as I can tell, 20awg 80*C 1000v solid core wire should be good up to 10amps. I’m asking because I have a ton of this wire, but I’m out of 18.