Building an attenuator?

[iamtherealmungo]

Hi

Before buying a Marshall power break or a thd hot plate I was wondering how difficult it expensive it is to build one yourself. I've seen a lot of DIY resistor based volume controls that people are calling calling attenuator but I've read that proper attenuator are different and the resistor based ones can damage an amp and generally sound bad..

What's the difference between the expensive ones and how would I go about making one?

I need it to handle a 50w head and 8ohm speaker load but I would like to make it switchable for various speaker setups.

Cheers

[DougH]

1. Google "amp garage".
2. Go to the site and register.
3. Download plans for the Air Brake.

It's pretty simple. I built one with surplus parts for about $30 or so, and it took an hour or so of my time. There's nothing wrong with using resistive attenuators. This one, designed by Ken Fischer, will work with up to 100W amps and any impedance. Don't get caught up in some of the hand-wringing over impedance- Ken knew what he was doing. Since it is resistive instead of inductive, the impedance is higher than you would expect, as it mimics the average impedance of an inductive load (which is much higher than the specified impedance- look at an impedance curve for a speaker to see what I mean). It is well designed, I've used mine a lot for a few years with no problems at all.

99% of the problems you hear about with attenuators are due to people diming their amps since now they can, with an attenuator. This wears out power tubes more quickly, and with vintage pieces can stress output transformers and so forth. If you push your amp 100% full throttle for long periods of time you will find weak spots in the amp.  I use a combination of master volume and attenuation to achieve the best sound. Attenuating a 100W amp down to bedroom volume almost never sounds good, for a large variety of reasons, many that have nothing to do with the attenuator itself. But this attenuator sounds really good, pretty transparent, and is cheap and simple to build.

[iamtherealmungo]

Thanks so much for the info!

Apologies for the lack of technical knowledge aha. But how does that build you have linked me to compare with a high end attenuator on the market?

What are the differences in the design and components?

I'm going to be running a 50w head with two 30 watt speakers so I'm hoping to get a much more useable distortion!

[DougH]

Give me an example of what you're referring to as a "high end" attenuator and I'll see if I can answer.

[iamtherealmungo]

Sorry that should have been made clearer. I'm referring to the expensive attenuators such as the THD Hotplate and the Marshall power brake. As well as the Weber mini mass attenuateor.

These very in price up to £200 so I am curious if I can make something just as good and safe for my amp head but much cheaper?

[slacker]

Those use some sort of inductive load or other trickery to try and make the amp sound the same when using a lot of attenuation as it does when played loud, the Webber thing uses what's basically a speaker without a cone. There's info here http://www.tedweber.com/atten.htm and here http://www.tedweber.com/understandingattenuators.html or google Fletcher–Munson for more theory.

These are no safer for your amp than what Doug posted, not that any of them are dangerous. No idea if they work any better.

[DougH]

Sorry that should have been made clearer. I'm referring to the expensive attenuators such as the THD Hotplate and the Marshall power brake. As well as the Weber mini mass attenuateor.

These very in price up to £200 so I am curious if I can make something just as good and safe for my amp head but much cheaper?

As Ian implied, these have some custom parts, for one. I believe the Marshall unit has a custom wound transformer. Not sure about the hotplate. The weber unit uses a "speaker motor" to provide an inductive load, which is a custom part they manufacture. Marshall and Weber provide an inductive load to "more accurately" model the impedance load a speaker provides to an amplifier. With an inductor, the impedance will vary according to frequency. At some frequencies, the actual impedance will be several times higher than what the speaker is rated.

If you compare say, an 8 ohm inductive load to an 8 ohm power resistor used as a load, you will hear a difference. The resistive load generally sounds more midrangey with less highs (from my experience). This is where Ken was smart with the Air Brake. The resistive load is actually 30 ohms or more (IIRC, been a while since I looked at the design). This is taken as sort of an average of the impedance load over the frequency range of the speaker. He looked at the impedance curves for a speaker (which BTW are different for different speakers) and based it on that. Compare this load to a 4/8/16  ohm inductive load and you probably won't hear much difference if at all. And this is much easier to build with easy to find parts. And it is safe and reliable (from my experience). And even with an inductive load, too much attenuation will start to sound bad.

Re. price- the cost I quoted was parts cost alone for surplus parts. With new parts (yes, there is a mouser parts list available for this) it will run around $100. Factor in labor, taxes, profit, etc, and those commercial units you mentioned are not really overpriced.

[azrael]

I've been looking at L-Pad system on eBay. Very cheap.

[iamtherealmungo]

I've read about the L Pad attenuators but it seems there are conflicting reviews on whether an l pad can safely attenuate a valve amplifier above 5-15 watts.

Does anyone have some insight on whether it's possible to create an l pad attenuator that could safely handle a 50 watt head?

[Mark Hammer]

ALL attenuators will attempt to let you run your amp louder but provide less of that output power to your speakers, while still letting the amp "see" the sort of output load it is designed to work with.

There are two kinds, basically.  One simply strives to turn that output into waste heat via a network of suitable power resistors that enable your speakers to mimic a much bigger speaker array, in conjunction with those power resistors.  So, if I have an amp with a single 8-ohm speaker, I might use a quartet of 16-ohm power resistors in parallel, run in series with the speaker and another duet of 16R power resistors in parallel with that.  The total nominal impedance is  8 ohms, but the entirety will behave like a sort of 7-speaker array with each power resistor "using up" some of the output power as waste heat.

The other kind recognizes that a speaker does not behave identically to a fixed resistor, and attempts to mimic the more nuanced electronic properties of a larger speaker array.  Those circuits are the pricier and more complex ones.  They also tend to be more complex and costly because they are often used for providing a line-level output for the mixing board, where replicating the behaviour of speakers is going to be more important.

How big a sonic difference between the two?  I have absolutely no idea, since I have used neither.  In any given circumstance, a difference MAY be audible, or it may not be.

[gruemungus]

My amp tech "Guy Claret" came up with a circuit he called  "control for electronic amplifiers" back in 79. It's a  simple circuit with a low parts count, and it works great.
http://www.google.com/patents?id=Jh84AAAAEBAJ&printsec=abstract&zoom=4#v=onepage&q&f=false

[Mustachio]

I was wanting to build an attenuator a while back and just went ahead and bought a 100watt 8 ohm L-Pad off ebay. Then I read more and more about how its probably not safe and all of my amps seem not right for the set up.

I have a 1973 peavey vintage two 12 model It has 4 6L6GC output tubes and I believe its around or over 100watts , I thought I read it was 110watts. Great amp Like a bassman from what I understand, but it has a solid state preamp. I have a late 80s early 90s Fender deluxe 85 and a newer Harke GT100 I recently got in trade for doing some work on a friends equipment. I also have a 1973 Ampeg I believe its a GV22-15 Which I have to rebuild. the ampeg was my first tube amp I gave away to a friend that plays harmonica. He had given it back to me about 8 or 9 years ago but broken and I haven't fixed it yet. It was a great amp two 12's and I believe 4 output tubes as well really loud!

Ok so those are my amps and I'm pretty sure the L-Pad I bought isnt a good idea to use with any of them. I thought I read something about L-pads not being good to use with a solid state amp. But some of my friends have smaller combos like fender hot rod deluxe . My question is what would be a good match with a 100watt 8ohm L-pad. I'd hate to give this to a friend and have it mess up their amp.

I've been thinking about building an amp from ax84 something like the high octane maybe. Would it be a good idea to save the L-pad to build into an amp like that. Figured Id ask in this thread since the convo was new.

And I dug up an old thread I bookmarked that had some good info

http://www.diystompboxes.com/smfforum/index.php?topic=88931.msg753017#msg753017

And here's another link I bookmarked a while back. This is what I was gonna follow to make mine.

http://guitarprojects.weebly.com/diy-attenuator.html

[azrael]

I've read about the L Pad attenuators but it seems there are conflicting reviews on whether an l pad can safely attenuate a valve amplifier above 5-15 watts.

Does anyone have some insight on whether it's possible to create an l pad attenuator that could safely handle a 50 watt head?

The ones I found on eBay are already built, cost less then 20 USD, and depending on which one you are looking at, are rate for higher wattage amps. You want something at least 1.5 times your amp's wattage, because when your amp's power section distorts, it will exceed the rated clean wattage.

[iamtherealmungo]

Hey,

So are L-Pads safe to use for higher wattage amps or is it a bad idea to try running a 50w head through a 100w L-Pad based attenuator?

I've seen the ready made L-pad attenuators on ebay but I know I could make it cheaper.

[Mike Burgundy]

I wouldn't risk your amps output section on that.
I might be wrong, but as far as I know Lpads are designed for attenuating tweeters in a 2-way speaker setup. I think they might suffer from the same mis-infomation tweeters do - the rated wattage is *not* the actual power the thing can handle, but an indication of *total system power*, while the tweeter only sees the higher spectrum bit of that. Also one of the reasons tweeters specify a minimum crossover freq.
Since there is a LOT less energy in higher frequencies, a "150W" tweeter crossed over at 2kHz might only see 20% of the full-range system power, so 30W.
I really fear Lpads are somehow rated in a similar manner.
That said - using them for attenuating tweeters should work just fine (I've got a 150W one on a "150W" SWR horn tweeter *after the crossover filter* without problems).
Using that full range on a 50W amp scares the heebiejeebies out of me.

[DougH]

I don't know much about L-pads intended for hi-fi use. But in general, be careful about applying equipment designed for hi-fi to musical instrument equipment- especially something intended for a solid-state device that you want to use with a tube amp. The "in-wall volume controls" are a good example. I played with one a few years ago. They are essentially an auto-transformer that attenuates the signal by creating a massive impedance mismatch of the speaker load to the amp. This is okay for solid state, but can cause problems with tube output circuits.

They may be making assumptions wrt power rating and etc that work just fine for a solid-state hi-fi circuit with a crossover, but fall apart when you plug in a 100W tube amp, like others mentioned.

[markeebee]

This page is a build report for a POSSIBLY suitable L-Pad, and has links to a nice pad calculator:

http://www.regiscoyne.com/ampoline/

I used something quite similar (but not built into a petrol can!) for a few years with a silverface Pro Reverb that was rated at 40W, and it didn't seem to harm the amp in any way.  That's just my experience though, so if your amp catches fire it's not my fault. The theory is pretty sound though.  The pad gets pretty hot - I initially built it into a closed plastic case and it got so hot that the case melted and stuck to the inside of the amp cabinet.  Also, I used it to reduce the volume of the nearly-fully-cranked amp from "nuclear meltdown" to merely "threshold of pain";  attenuating down to bedroom level will obviously dissipate even more heat, and that might get hairy.

I did notice that the pad attenuated the high frequencies more than the rest of the spectrum, so much so that I used to leave a wah switched on to boost the treble.  And that, children, is how I arrived at my signature sound, subsequently coveted by absolutely no-one.

Oh yeah, and, I have a newish Vox AC4 with built in L-Pad and I *think* I can notice a drop off in high frequencies when the L-Pad is maxed, but I can't trust my ears too much. 

[artifus]

I did notice that the pad attenuated the high frequencies more than the rest of the spectrum, so much so that I used to leave a wah switched on to boost the treble.

this would be due to the difference between impedance and resistance and the confusion (including my own!) caused by both being referenced in ohms, no? as alluded to earlier in the thread and the resistors vs speaker emulators, etc, discussion. i've been reading a little about current drive vs voltage drive recently and am still a little confused.

is there more to impedance vs resistance than just frequency response in audio signal circuits? or is that just a handy shorthand way of thinking about such things? educate me!

*edit* i googled:

The short answer is -- impedance includes reactance, and reactance includes effects which vary with frequency due to inductance and capacitance.

i already knew this, just a little fuzzy today - need another coffee. i think what i was getting at was that could an attenuator be made with an l-pad and a (current drive?) treble boost circuit to compensate for hf loss, as markeebee suggests, or are there other factors to be considered too?

[markeebee]

Hmmmm,  I think the perceived loss of high frequencies is maybe partly due to reactive attenuation,  and partly due the rolloff in normal hearing. Slacker hit the spot,  I think, when he mentioned Fletcher Munson.

Because I'm lazy and defeatist,  I wouldn't dream of calculating the amount of boost(s) that should be applied to the upper frequencies. There are at least two sets of varables,  and only one life.  And less than one brain.  I'd be inclined to monkey with a graphic eq on the input to compensate for any damage to the output.

That's the view from the shallow end.

[artifus]

true, but f&m curves relate to human perception with regard to loudness and are often compensated for with a smile eq curve - that's what the loudness button on your hifi does - dips the mids - not what signal is sent to a loudspeaker and how it interacts with it.

as for treble boost calculations - that's why current drive popped in my head but i'm still a little fuzzy on that. but, yeah, life is short and eq works.

*gah!* i'm going for a walk while the coffee kicks in to spare you all any more of my incoherent rambling...