DIY atenuator?

[pissdrunx]

http://blog.makezine.com/archive/2008/07/diy_attenuator_box_for_yo.html

 

Is there an L-pad that would take 50w at 16 Ohm? If there is, is this safe to use with JCM 800 50w? I really don`t wanna fry it!

[el_tonto]

You can make a fixed 3dB attenuator to take it down to ~25W prior to the L-pad to be on the safe side.

[WRGKMC]

Marshalls are Very sensitive to impediance changes. I would never recomend a straight resistance. You take a big chance at blowing the output transformer and power tubes.

Marshall has an impediance plug/switch on them for proper impediance matching. A missmatch can be very bad.

 

Using resistors for dropping the output down only affects one frequency. You're dealing with impediance not resistance on the output of an amp. The impediance varies with frequency from a couple of ohms up to several hundred. Putting a straight resistor will only affect one frequency. What you need is an active impediance that reacts like a speaker does. A Hot Plate acts like a transformer to step the voltage and current down. I would look for a diagram for building a hot plate. If it was a cheesy littel solid state amp with a lower output its one thing, a marshall is another. If there was a cheap safe method they would build it into an amp. Some amps have ways of turning off a few of the power tubes and using a different tap on the output transformer. They dont just throw an Lpad on there. Other than a Hot Plate, You can buy Yellow Jackets. Theyre tube adaptors that let you run low wattage tubes for lower volumes which are prooven to be very safe to use and require no internal mods to be done.

[Doctorx]

I've built several attenuators like this with the 100 watt version from Parts Express, and they work fine for small amps (Up to about 25 watts rms) but no way would I use a 50 watt Marshall or any other amp with that much power with one. Even with smaller amps I use a fan to keep it cool.

[aortizjr]

I would check out the ones from tedweber.com.

 

I think you can DIY a MASS and save some money that way. But basically it uses a speaker motor as the load. So it essentially is a speaker.

[isaac42]

 

Marshalls are Very sensitive to impediance changes. I would never recomend a straight resistance. You take a big chance at blowing the output transformer and power tubes. Marshall has an impediance plug/switch on them for proper impediance matching. A missmatch can be very bad. Using resistors for dropping the output down only affects one frequency. You're dealing with impediance not resistance on the output of an amp. The impediance varies with frequency from a couple of ohms up to several hundred. Putting a straight resistor will only affect one frequency. What you need is an active impediance that reacts like a speaker does. A Hot Plate acts like a transformer to step the voltage and current down. I would look for a diagram for building a hot plate. If it was a cheesy littel solid state amp with a lower output its one thing, a marshall is another. If there was a cheap safe method they would build it into an amp. Some amps have ways of turning off a few of the power tubes and using a different tap on the output transformer. They dont just throw an Lpad on there. Other than a Hot Plate, You can buy Yellow Jackets. Theyre tube adaptors that let you run low wattage tubes for lower volumes which are prooven to be very safe to use and require no internal mods to be done.

 

 

Only one frequency? That strikes me as a strange thing to say. Which frequency are you saying it affects, to the exclusion of all others?

[WRGKMC]

Issac24, maybe not the best wording there. I should say the resistance doesnt change with frequency the way a speaker impediance does. The resistor will only match the transformer impediance at one frequency might be a better way of putting it.

 

Pulled the first google search here and it had a graph to explain a littel better. http://www.bcae1.com/resvsimp.htm

 

You can see the speaker and resistor matching at one frequency where the two cross over at 50hz in this example. Its probibly the resonant frequency of the speaker.

 

This isnt the best explanation I've seen because of the lack of reasoning as to why this occurs. The consequences EMF in a coil causing phase shifting of current and voltage and all the jazz about "eli the ice man" and impediance isnt there but it suffices to know that a resistor definately isnt a sunstitute for a moving coil and should never be used on hi powerd amps.

 

One exception to the rule is in 70V audio transmissions. Lpads are used in intercom type systems. High voltage is used for longer diatance audio transmission. The voltage is high but the current is low in those applications.

I dont remember pot values but they match that kind of impediance range if its tapped in before the transformer at the speaker that changes the high voltage to higher current.

[djcgtr]

 

Issac24, maybe not the best wording there. I should say the resistance doesnt change with frequency the way a speaker impediance does. The resistor will only match the transformer impediance at one frequency might be a better way of putting it. Pulled the first google search here and it had a graph to explain a littel better. http://www.bcae1.com/resvsimp.htm You can see the speaker and resistor matching at one frequency where the two cross over at 50hz in this example. Its probibly the resonant frequency of the speaker. This isnt the best explanation I've seen because of the lack of reasoning as to why this occurs. The consequences EMF in a coil causing phase shifting of current and voltage and all the jazz about "eli the ice man" and impediance isnt there but it suffices to know that a resistor definately isnt a sunstitute for a moving coil and should never be used on hi powerd amps. One exception to the rule is in 70V audio transmissions. Lpads are used in intercom type systems. High voltage is used for longer diatance audio transmission. The voltage is high but the current is low in those applications. I dont remember pot values but they match that kind of impediance range if its tapped in before the transformer at the speaker that changes the high voltage to higher current.

 

 

To help clarify the use of an L-Pad in a 70 volt or other distributed audio system, the 8 or 16 ohm L-Pad is always after the transformer at any given speaker location.

 

A 70 volt speaker transformer may have a fixed wattage primary to a given rated speaker load or may have a multi tapped primary that will provide a certain chosen wattage to a speaker load.

 

The wattage delivered is relative to the AC audio voltage in the distributed system at any given time. The fixed or tapped transformer primary rating would be max watts delivered if 70 volts of audio is actually appearing at the primary. Most of the time it is less.

 

The L-Pad maintains a nearly constant load on the secondary winding of the 70 volt transformers, by using two variable resistors on one shaft.

That when the L-Pad is connected to the appropriate matching speaker impedance, the changes in volume is happening by these two variable resistances going in opposite directions of resistance value.

 

I fully agree that a tube amplifier has a certain response in its voicing by the interaction of the output circuit and the speaker or speakers loading, along with the speaker enclosure.

 

A fixed resistance that is not a speaker will essentially have no affect on any one frequency more than another frequency. A wirewound resistor will have a certain amount of inductance that will begin to create the phase shift between voltage and current - and will be more as the frequency goes higher.

 

It is very true on the changes in actual impedance of a speaker at any given applied frequency.

 

On a tube output amplifier, as long as it does have a reasonably matched load, there should not be any damage done. The no load situation at a sufficient output level can create problems with arcing either in the output transformer windings or inside the output tubes and certainly to be avoided.

 

Using an L-Pad on a tube amp should not damage anything as long as the rating of the L-Pad is sufficient to not overheat and subsequently burn out which would lead to the ''no load'' problem.

While I do not see any damage done with a L-Pad, the characteristics of the amp voicing would certainly change

[WRGKMC]

djctgr, I think I know about the application you're talking about there. I was thinking more along the lines of an in wall attenuator that would control speakers in a room or in a ceiling. They attenuate the 70 volt line and not the speaker output but are more likely a stepped transformer or combination of resistors and a transformer. Lpads on speaker outputs become scratchy an burn out unless they are a high quality/wattage rehostat. Either way, most home intercoms suffer from the scratchy volume syndrome and its probibly why 70 volt attenuators are favored over them.

As far as using an Lpad on a tube amp I'll stand firmly against that one.

Transistor amp at lower wattage fine, but a guitar amp pushing alot of current you're asking for troubble. As far as a wire wound resistor developing any kind of decent EMF its not going to happen. Theres no iron core to magnify the magnetic field like a choke or transformer would produce.

The best advice for safety and fidelity is a hot plate, power break, power soak type design because it matches impediance and is the only advice I would give someone in this case.

I wouldnt advise anyone otherwise otherwise. The risks of damage to a classic amp head outweighs any kind of cost savings as well.

 

Another option that hasnt been mentioned here is to is to install a post inverter master volume control. depending on the design you might be able to add a 100K dual ganged pot in place of two fixed resistors. May need to research it though. I've read good and bad results same as any kind of speaker attenuation. My buddy had one with a master volume on his and it worked well for hime but it was a plexi like mine not a jmc.

 

I personally say screw any attenuation at all. I just use different speaker boxes and speakers with different SPL levels for lower volumes. I've built or bought 1X10", 2X10" 1X12" 2X12", 4X10", and a full sized 4X12" marshall cab that will not only provide different volumes but different tones as well. What I would do is go to parts express, pick up a dayton 120w rms 10" speaker that are on sale for $23 and build a box for it. You can crank the head up as much as you want, and still get tube saturation at lower volumes. Do a nice job and its great for smaller gigs or recording. Be sure to make the box so it will fit the head and finish it nice.

[djcgtr]

 

djctgr, I think I know about the application you're talking about there. I was thinking more along the lines of an in wall attenuator that would control speakers in a room or in a ceiling. They attenuate the 70 volt line and not the speaker output but are more likely a stepped transformer or combination of resistors and a transformer. Lpads on speaker outputs become scratchy an burn out unless they are a high quality/wattage rehostat. Either way, most home intercoms suffer from the scratchy volume syndrome and its probibly why 70 volt attenuators are favored over them. As far as using an Lpad on a tube amp I'll stand firmly against that one. Transistor amp at lower wattage fine, but a guitar amp pushing alot of current you're asking for troubble. As far as a wire wound resistor developing any kind of decent EMF its not going to happen. Theres no iron core to magnify the magnetic field like a choke or transformer would produce. The best advice for safety and fidelity is a hot plate, power break, power soak type design because it matches impediance and is the only advice I would give someone in this case. I wouldnt advise anyone otherwise otherwise. The risks of damage to a classic amp head outweighs any kind of cost savings as well. Another option that hasnt been mentioned here is to is to install a post inverter master volume control. depending on the design you might be able to add a 100K dual ganged pot in place of two fixed resistors. May need to research it though. I've read good and bad results same as any kind of speaker attenuation. My buddy had one with a master volume on his and it worked well for hime but it was a plexi like mine not a jmc. I personally say screw any attenuation at all. I just use different speaker boxes and speakers with different SPL levels for lower volumes. I've built or bought 1X10", 2X10" 1X12" 2X12", 4X10", and a full sized 4X12" marshall cab that will not only provide different volumes but different tones as well. What I would do is go to parts express, pick up a dayton 120w rms 10" speaker that are on sale for $23 and build a box for it. You can crank the head up as much as you want, and still get tube saturation at lower volumes. Do a nice job and its great for smaller gigs or recording. Be sure to make the box so it will fit the head and finish it nice.

 

 

Yes, the stepped wall attenuators are a multi tapped auto-transformer that maintains a match for the distributed audio line, and reduces the feed to the local speaker transformer.

And, yes, a much preferred approach for distributed audio volume control.

 

On the L-Pad thing....

I did not advocate using one.

I said " the characteristics of the amp voicing would certainly change ".

And as long as there is a load on any tube amp - it will not damage the amp, with an appropriate resistive or speaker load.

 

Whether the amp is solid state or tube output, it still is the same current for whatever the rated wattage is of the amp. 50 watts is 50 watts.

 

I mentioned the inductance of a wirewound resistor because there is some there. Most of that minor inductance ( reactance ) is insignificant at audio frequencies. But I still use non-inductive wirewound resistors as a load for amp testing either solid state or tube work.

Wirewound L-Pads have a minor amount of inductance similar to a fixed wirewound resistor.

 

I agree that the Hot Plate or Power Soak type units would be a better way to reduce levels to a speaker in a tube guitar amp......

If the performer wishes to over drive their output tubes for that "sound".

 

I am perhaps in the minority, but I have a very good overdriven sound from my tube amp and do not run wide open.

I have recommended to my customers that my preference would be to not use PowerSoak type units and seek preamp overdrive, or external processing before the amp or on a in/out loop from the amp.

Being fully aware of the "magic" of overdriven output tubes, it is "To each their own".

I prefer more life out of my output tubes by not driving the car with the gas pedal all the way to the floor and holding the brake pedal part way down, to maintain a certain speed.......

 

I also have to wonder how many in the audience observing and listening to a band are aware that the guitar sound would be that compromised by using a amp that doesn't have a PowerSoak in the output, and OD being done a different way....

 

That, and the speakers have their contribution to a sound by the amount of power applied to them.

 

My Marshall is a master volume style. Even though I feed the amp with my processed signal and run the amp "clean"....

I have done requested modifications to non-master volume Marshalls and other tube amps by installing a dual master volume pot in the post phase inverter location and they were happy.

 

But - as I said - To each their own.

Everyone is entitled to do what suits them.

Options for guitarists are many fold.

[isaac42]

 

Issac24, maybe not the best wording there. I should say the resistance doesnt change with frequency the way a speaker impediance does. The resistor will only match the transformer impediance at one frequency might be a better way of putting it. Pulled the first google search here and it had a graph to explain a littel better. http://www.bcae1.com/resvsimp.htm You can see the speaker and resistor matching at one frequency where the two cross over at 50hz in this example. Its probibly the resonant frequency of the speaker. This isnt the best explanation I've seen because of the lack of reasoning as to why this occurs. The consequences EMF in a coil causing phase shifting of current and voltage and all the jazz about "eli the ice man" and impediance isnt there but it suffices to know that a resistor definately isnt a sunstitute for a moving coil and should never be used on hi powerd amps. One exception to the rule is in 70V audio transmissions. Lpads are used in intercom type systems. High voltage is used for longer diatance audio transmission. The voltage is high but the current is low in those applications. I dont remember pot values but they match that kind of impediance range if its tapped in before the transformer at the speaker that changes the high voltage to higher current.

 

 

It seems to me that it's speakers that match the transformer at only a few frequencies. Resistors selected to match the nominal impedance of the output transformer will match at all frequencies. But the interplay between the output transformer, with its relatively high output impedance compared to solid state amps without output transformers, and the complex impedance of speakers does affect the frequency response and therefore the sound of the amp. Bob Carver demonstrated this when he modified a solid state amp to sound like a high-end tube amp by simply adding a small resistance in series with the load. The golden-eared audiophiles couldn't hear the difference.

 

So I don't think there's any chance of damaging the amp by having a more resistive load on the output, but I would expect it to sound a little different, not just quieter. FWIW, I've used a Scholtz Power Soak with my Sunn Sentura II lots of times, and never noticed any difference other than volume, but the Sunns are a pretty hi-fi design, with their ultralinear output stage.

 

And that's not a very good representation of a speaker's impedance for our purposes. The speaker resonance is either 20 Hz or below, not at all what we see in guitar speakers, or even most bass speakers.

[WRGKMC]

Good discussion guys. All opinions noted and no problems there dcjtr. Back in the late 70s when I did pro audio repairs for a living we used a straight resistive load to drop the output down on out test speakers. Also had caps across the speakers to preven the speakers from blowing if the amps went DC. There was no way we could test 500w power amps at maximum power without some resistance there, but we were testing repairs. the added load stressed the output and if something was going to blow breakup it would be with that additional load running a frequency generator through it.

Still think you're both missunderstanding my one point about frequencies and impediance matching and what will occur with a steady resistance without an inductive load on the transformer. My point is the transformer will only see the proper 8 ohm load at one frequency. Its is not what the resistor sees.The resistor will see all the frequencies. What kind of load it prduces will remain the same. A speaker load varies with frequency. Its not a steady 8 ohms. Without this varience in impediance load, youre going to take chances with blowing the transformer because that EMF the transformers producing isnt going to be absorbed by the speaker and will result in heating up the transformer.

 

I found this creative site that has a good illustration that plots the results in a more 3d way. Its a hard concept to understand without in depth study but viewing it here is a snap. The second page of the article shows the results I'm talking about which is pretty neat. http://www.hometheaterhifi.com/features/technical-topics/a-new-way-to-plot-speaker-impedance-the-smith-chart.html

 

You can clearly see there is a higher load above 5K which is a marshalls trebble voicing region.

 

Also Issac.

I have an older 70s Concert Lead Sunn head in my collection as well. While replacing caps and studying the circuit I discoverd they used a multitap choke transformer and added coils between the power transistors to add inductance to the output. Its probibly why Its one of the best sounding tube like transistor heads I've heard.

[Doctorx]

You guys smart and well informed. As I said in my earlier post, I have built several guitar attenuators using the Parts Express 100 watt L-Pads.

 

I've never pushed one of these to the point of burning up, but experiments with amps like a Bassman head or a Bandmaster show they get really dangerously hot for comfort.

 

There's no way either of these amps even dimed are putting out more than 100 watts, so how do they get that 100 watt rating?

 

 

The best I can come up with is that they expect these things to be used after a crossover controlling only high frequencies? They don't state that though, it's just a 100 watt rating.

[WRGKMC]

You're exactly right on them being used after the crossover for mids and tweeters. All the low frequencies that would have a large motor action on woofers is removed by crossover. The caps passing the mid and trebble to those can be safely attenuated by with a rehostat and have minimal loading on the overall impediance as well. Still you can cook them in a Hi Fi application with too much wattage, and is the speakers dont have fuses.

[djcgtr]

 

You guys smart and well informed. As I said in my earlier post, I have built several guitar attenuators using the Parts Express 100 watt L-Pads . I've never pushed one of these to the point of burning up, but experiments with amps like a Bassman head or a Bandmaster show they get really dangerously hot for comfort. There's no way either of these amps even dimed are putting out more than 100 watts, so how do they get that 100 watt rating? The best I can come up with is that they expect these things to be used after a crossover controlling only high frequencies? They don't state that though, it's just a 100 watt rating.

 

 

As far as a rating on your 100 watt L-Pad, that is an overall rating of the ability to handle, and dissipate into heat, up to a 100 watt amount.

And yes they will get hot. Consider that since the load is staying fairly steady and the driver has more or less delivered to it for power, depending on the L-Pad adjustment, that extra power has to go somewhere - therefore heat.

 

You seen the explanation on the use as an attenuator for highs and occasionally mids in a speaker enclosure.

 

The 100 watt versions ( and most others ) are wirewound dual potentiometers ( inside ) and built into a ceramic housing with metal because of the heat.

Some older L-Pads I have seen are literally two pots on a common shaft and wired to achieve a resistance going lower to the driver and higher on the shunt resistance as you turn it ''up''. It electronically works the same as what you have.

 

The unit you have is internally connected so that when using the outside terminals ( of 3 ) going to the source of the audio - it maintains fairly closely, the desired resistance for a ''load'' regardless of the position of the pot shaft.

 

That trait, in conjunction with the center and ''lowside'' common end terminal going to the proper driver rated impedance, allows an amplifier or on some instances like a hi-pass out of a passive crossover to ''see'' a relatively constant load.

 

As far as trying to use a non L-Pad style single pot in series ( a rheostat ) with a driver - that is when the overall load in ohms ( combination of a series resistance and driver ) increases. Yes - that way does reduce volume output of the driver but does not maintain a relatively constant load.

 

This is why an L-Pad is what it is.

 

Passive crossovers rely on a specific driver load in ohms to actually ''cross-over'' at the designed point in frequency.

 

Yes, it has been discussed that drivers vary in their actual impedance at any given frequency. It is interesting to look up graphs of drivers that plot frequency vs impedance. And if you had an impedance meter, you could independently verify your chosen driver in a specific enclosure by sweeping the meter output freq on through the bandwidth of your enclosure and even plot it yourself.