Simple Attenuators - Design And Testing

[BlueX]

I think you are ok with all that. I've been assuming 18 gage is 1mm diameter.

Thanks! I thought it could be of general interest to have the "SI" dimensions also.

 

[mike_lawyer]

Thanks! I thought it could be of general interest to have the "SI" dimensions also.

I have two cliff jacks, but they have 4 solder lugs instead of 6. Does it make any difference for this build?

 

[JohnH]

I have two cliff jacks, but they have 4 solder lugs instead of 6. Does it make any difference for this build?

Sounds like a mono jack. That'll work for everything except if you are building an 8 Ihm M2, and you want to include Out 3 which is for a 16 Ohm speaker.

Also, the stereo jacks just feel better, they grip firmer due to the extra spring contact, even if it's not connected.

 

[NoNaMe]

The simplest attenuator for Effects loop. Marshall MA100C. Works perfect!

 

[Barnsley Boy]

The simplest attenuator for Effects loop. Marshall MA100C. Works perfect!

View attachment 102109

Ummmm!! ..... not really! That if I am very much not mistaken is a volume control that sits between the pre and power amp stage. I have a Donner Emo amp controller (effectively the same thing) that I use with my Peavey Classic VTX. It's the only way I can get it anywhere near domestic playing levels.

The VTX is my next candidate for one of @JohnH 's super duper attenuators. @NoNaMe please do yourself and your lovely Marshall a favour and get building!!!

 

[JohnH]

The simplest attenuator for Effects loop. Marshall MA100C. Works perfect!

View attachment 102109

That's a neat version of how to put a volume control in the FX loop. I wonder if the spacing between jacks is consistent enough for it to fit across a range of models?

That sort of simple device will let you reduce the volume if you don't have a master volume control, or as an alternative to it. Your poweramp stage will stay clean and you find your tones in the preamp or with pedals. But as BB notes, what it cant give you is the distinctive tone, and dynamic feel, of a power amp running hard at high power. For that the power amp actually has to be working hard. Its that which our attenuator is designed to provide, at any volume that you want.

Your MA is likely to to respond well to that.

 

[mike_lawyer]

In terms of switches, what amp and voltage ratings are fine for the DPDT switches? I have some rather large switches on hand but they are harder to throw, I'd like to get some mini toggles instead.

 

[Barnsley Boy]

Afternoon all,

The soldering iron has been unsheathed again, with an additional -28db stage added to my 8 ohm M2, and replacement of mini-toggle switches with some serious chunk-mongers. Pics will be posted on the completed build thread in due course, but at the moment I'm struggling getting the -28db to work.

For me to get it clear in my head, I have to draw it out so that it makes sense.

@JohnH 's M2 design schematic, with the stages re-ordered in ascending order, and additional stage tagged onto the end:





To make sure I understood the switching for that last stage, I extracted it and "exploded" the DPDT switch:




Finally, I incorporated into my colour coded pictorial layout, loosely based around @Gene Ballzz 's really helpful layout diagram, but adjusted to suit the layout in my enclosure:




Note - stage 2, 3 and 4 have been drawn as SPST switches to make it a bit easier to trace the signal path, and to reflect my latest modifications.
When I throw the switch to engage stage 5, it cuts off the sound. Disengaging, and the other stages work perfectly (as they did previously before I started tinkering).

@JohnH and @Gene Ballzz can you see any mistake that I may have made?

 

[JohnH]

hi @Barnsley Boy , the schematic bits look ok. On the wiring diagram, the -28 stage is first after stage 1 and that's ok too. I think RA and RC are swapped. Coming from the amp side, the signal should go through RA (6.8) first. But that swap shouldn't cut out all sound. I may be missing something, or there could be something in the build not on the diagram.

 

[Barnsley Boy]

Hi @JohnH .... Ah, you may have a point!!



I think maybe it should look like the image above. I'll fiddle around with it tomorrow and see if it makes a difference. The cutting out bit may just be a bit of dodgy soldering on my part.

Am I correct in assuming that connecting a multimeter across one of the output jacks and switching through each of the combinations from -7db up to -59.5 (GULP) db will give me a progressively higher resistance reading? I've worked out that it will give 16 combinations in -3.5db steps? Should the difference in resistance be the same for each step? I'm thinking that it would be a good measure of checking that everything is working correctly.

Thanks for coming back to me on this, your time and effort is very much appreciated

 

[Lancer X]

Hi @JohnH , @Gene Ballzz and attenuation crew. Thanks for sharing your amazing work with with the community!

My questions will require some setup. Like @len_zwerf , who posted a few pages back, my most recent project was a micro amp. In my case, Rob Robinette's Deluxe Micro Mod design, which is a 1.0-2.0W single-ended class A version of a 5E3 Tweed Deluxe. I built it in a Vibro Champ combo cab (open-backed ) with a 10" Weber 10A100T alnico speaker (8 ohm). Rob's design does include a switchable negative feedback loop option, but of course the default Tweed setting does not.

The point of this amp build was to simplify my practice rig. Previously, I had built a full-sized 5E3 and ended up getting a Weber MiniMASS since a Tweed Deluxe is not family-friendly once you get it singing. Instead, this little Deluxe Micro is only 1W and offers a Master volume. Still, I would like to keep the power section grooving without ticking everyone off, but even this little amp can still make about 95 dB. Your M2 seems like just the ticket.


Question #1 - if my goal is limiting to ~60-70 dB with the master dimed, which M2 stages do you recommend that I include? I assume that your -7 or -14 dB references are relative to a 50W amp, right??

Question #2 - could I order say 10W and 5W power rating resistors, since this amp is only 2W max?

Question #3 - would anyone be interested in assisting me to tweak the M2 design to model the reactive response of an alnico speaker in an open-backed cab? (Or perhaps that exercise isn't necessary??) If so, I'd propose a 12" Jensen P12R 8 ohm as our model - Jensen publishes the impedance response curve on the product page. Should be broadly applicable for vintage "American" amps, and useful info for the "rest of us" who aren't rocking Marshalls. (Expecting that '65+ Fender-style designs probably track pretty well with your existing ceramic Greenback model.) Maybe this is a deep endeavor - I don't want to presume? But willing to lend my effort if doable and of interest.

 

[JohnH]

Hi @Lancer X , welcome to our world!

Taking on your numbers at face value, if the dimed amp can do 95db, and you might like to hear that as low as 60db, then that's a reduction of -35dB. The full M2 is -31.5db, but I'm sure that that will be enough. But you might not want to leave any stages out, just in case you need them sometimes. I reckon you'll mostly use less attenuation though.

Yes you can scale down the power rating of the resistors. For a nominally 50W amp, R1 is spec'ed at 100W rating, ie 2x the amp power. It takes about 30% of the power of the amp, which gives is a factor of x3 for a case-mounted resistor. You might chose to try the ceramic block resistors, for which Id give a bigger factor. say 5. 2W x 30% x 5 = 3W. So you could go 5W or 10W.

Id still suggest to use the same 18 or 19 gage coil. Obviously it can handle a lot more power, but we need it also to be low resistance too.

On the Jensen speaker, actually, no need to change anything if you are building the M2 design, and will be using that speaker with the attenuator. The coil in conjunction with the circuitry around it is modelling the rise in impedance at higher frequency, That Jensen is within about 10% of the Celestion G12M measurements in terms of high frequency impedance. At the low end, the resonance and the influence of the cab type is not modelled, because your real speaker will do that itself. Instead, the attenuator is trying to feed a signal to the speaker like that of an amp. so the speaker can do its own thing. Feed it into a closed back cab instead and that is what it will sound like.

We do however, also have design M3, which does indeed model the bass resonance. this is mainly for when it is used as a loadbox with no cab. In that case, a resonance could be matched to the Jensen. But with a cab, this makes a very slight difference, only if you play the low bass notes, and generally we have found its not really needed of you are using a guitar cab.

 

[JohnH]

Hi @JohnH .... Ah, you may have a point!!

View attachment 102194

I think maybe it should look like the image above. I'll fiddle around with it tomorrow and see if it makes a difference. The cutting out bit may just be a bit of dodgy soldering on my part.

Am I correct in assuming that connecting a multimeter across one of the output jacks and switching through each of the combinations from -7db up to -59.5 (GULP) db will give me a progressively higher resistance reading? I've worked out that it will give 16 combinations in -3.5db steps? Should the difference in resistance be the same for each step? I'm thinking that it would be a good measure of checking that everything is working correctly.

Thanks for coming back to me on this, your time and effort is very much appreciated

Diagram looks better....

The resistance readings shouldn't continue to increase at higher settings. The idea is that they stay much the same.

With no amp connected, but speaker plugged in, the reading across a cord into the input (ie where the amp should go) should be 7 to 10 ohms in all settings, after subtracting for the meter leads. With the new -28db stage, id should be even closer, like 8 or 9 ohms.

 

[Barnsley Boy]

Nice one @JohnH . Fingers crossed, I should be up and running tonight!

 

[Gene Ballzz]

Hi @JohnH , @Gene Ballzz and attenuation crew. Thanks for sharing your amazing work with with the community!

My questions will require some setup. Like @len_zwerf , who posted a few pages back, my most recent project was a micro amp. In my case, Rob Robinette's Deluxe Micro Mod design, which is a 1.0-2.0W single-ended class A version of a 5E3 Tweed Deluxe. I built it in a Vibro Champ combo cab (open-backed ) with a 10" Weber 10A100T alnico speaker (8 ohm). Rob's design does include a switchable negative feedback loop option, but of course the default Tweed setting does not.

The point of this amp build was to simplify my practice rig. Previously, I had built a full-sized 5E3 and ended up getting a Weber MiniMASS since a Tweed Deluxe is not family-friendly once you get it singing. Instead, this little Deluxe Micro is only 1W and offers a Master volume. Still, I would like to keep the power section grooving without ticking everyone off, but even this little amp can still make about 95 dB. Your M2 seems like just the ticket.


Question #1 - if my goal is limiting to ~60-70 dB with the master dimed, which M2 stages do you recommend that I include? I assume that your -7 or -14 dB references are relative to a 50W amp, right??

Question #2 - could I order say 10W and 5W power rating resistors, since this amp is only 2W max?

Question #3 - would anyone be interested in assisting me to tweak the M2 design to model the reactive response of an alnico speaker in an open-backed cab? (Or perhaps that exercise isn't necessary??) If so, I'd propose a 12" Jensen P12R 8 ohm as our model - Jensen publishes the impedance response curve on the product page. Should be broadly applicable for vintage "American" amps, and useful info for the "rest of us" who aren't rocking Marshalls. (Expecting that '65+ Fender-style designs probably track pretty well with your existing ceramic Greenback model.) Maybe this is a deep endeavor - I don't want to presume? But willing to lend my effort if doable and of interest.

@Lancer X ,
First> to the forum!

Next> I strongly recommend that you build the full blown, full wattage version of the M2. Once you plug your "full sized" 5E3 into the M2, your Weber MiniMass will collect dust! You may find that your Deluxe Micro also collects dust, as two triodes in parallel for the power section simply do not have the dynamics and overall sound of two 6V6s in push/pull.

While the Weber is a nice, convenient unit for some things (4/8/16 ohm settings) and just shaving "a little off the top" volume wise, it won't even come close to holding a candle to the performance of the @JohnH unit! I too own a MiniMass and will never get rid of it, but.......... Trust me when I state that you will not miss (or need) that Treble compensation switch of the MiniMass. Any initial, perceived cost, complexity and/or size savings of building a "trimmed down" version will end up, IME, being a waste of the extra effort! The M2 will quickly become the most used piece of gear in your arsenal and you will want to use it with any other amp you happen to acquire! The only part you might want to consider "trimming down" could be the extra, compensated 16 ohm output jack, if building an 8 ohm unit.

Also> Can we assume that both of the mentioned amps have an 8 ohm output tap and an 8 ohm speaker?

Happy Attenuatin'
Gene

 

[Gene Ballzz]

@JohnH ,
Has it ever been determined whether or not the 8 ohm M2 is as friendly to a 4 ohm load as the 16 ohm version is to an 8 ohm load?
Just Curious?
Gene

 

[Lancer X]

Hi @Lancer X , welcome to our world!

Taking on your numbers at face value, if the dimed amp can do 95db, and you might like to hear that as low as 60db, then that's a reduction of -35dB. The full M2 is -31.5db, but I'm sure that that will be enough. But you might not want to leave any stages out, just in case you need them sometimes. I reckon you'll mostly use less attenuation though.

Thank you so much for your advice, response, and generosity @JohnH and @Gene Ballzz !!

RE my 95 dB number, that was a quick back-of-envelope estimate using an online SPL calculator and the sensitivity of a Jensen P10R (Weber doesn't publish specs, which is super annoying). Calculator says 92-94 dB from clean to gritty (1.0-1.5W output). Please do correct me if my SPL assumptions are faulty!

I hadn't understood that the dB attenuation numbers were universal regardless of amp wattage - thanks. I will look up the theory/math on that topic.

One last question if you'd be so kind: from what I'm reading here, sounds like that at this low wattage a simple aluminum Hammond chassis should suffice as a heat sink?? How about if I build an M2 for my full-size Tweed Deluxe (15W, No NFB)?

Thanks again - so excited to build this critter!

 

[Gene Ballzz]

A Hammond 1590D will work well, but @JohnH prefers the slightly larger 1550G. These are the fairly thick die cast boxes, not the thin aluminum sheet metal units! I implore you to build it as a well ventilated tank and you will never have a problem!
Just Sayin'
Gene

 

[Lancer X]

@Lancer X ,
First> to the forum!

Thank you, sir! I arrived here specifically Googling for reactive attenuators. Lucky me!!

Next> I strongly recommend that you build the full blown, full wattage version of the M2. Once you plug your "full sized" 5E3 into the M2, your Weber MiniMass will collect dust! You may find that your Deluxe Micro also collects dust, as two triodes in parallel for the power section simply do not have the dynamics and overall sound of two 6V6s in push/pull.

I may build one dedicated to the big 5E3, which I am selling, as an added feature. Likewise for the Deluxe Micro, integrating them into the back of the combo cabs. The 5E3 was my first build, so still jonzing for more soldering.

As such, I plan to build Rob Robinette's 2W Tweed 5F6A Bassman Micro with the proceeds of the 5E3 sale. Specifically, he also designed a variant with two EF80 pentodes in push-pull. 99% of my playing for the foreseeable future will be at home, and this little gal will be absolutely perfect for my needs. So, I think I'm right with you re my direction.

That said, you'd be surprised how lovely this Deluxe Micro sounds with just that little 12AU7. While it likely differs a bit from a 5E3 (also lacks the cathodyne phase inverter of course), there's something special about single-ended Class A tone.

Also> Can we assume that both of the mentioned amps have an 8 ohm output tap and an 8 ohm speaker?

Yep, precisely!

 

[Lancer X]

A Hammond 1590D will work well, but @JohnH prefers the slightly larger 1550G. These are the fairly thick die cast boxes, not the thin aluminum sheet metal units! I implore you to build it as a well ventilated tank and you will never have a problem!
Just Sayin'
Gene

At the moment, I am chasing these Micro amps, so my application is a 2W bedroom amp. @JohnH confirms that I can get away with 5-10W resistors, and I'd like to go in that direction. (I'm not keeping the 5E3 regardless.) I want this attenuator to fit into, and be at least semi-permanently integrated in the amp chassis (edit: oops - I meant inside the amp cabinet, not in the chassis). I know it's OCD of me, but the Weber MiniMASS sitting on top of my Deluxe always bugged me.

Given the 2W amp output and a M2 with 5-10W resistors, do you think that a ventilated sheet metal style chassis would offer sufficient cooling?