tube amp direct line out

[famousmockingbird]

So my drummer whom thinks he understands more than he does about sound engineering and electronics won't shut up about my amp not having a direct out. We practice "live" in our facility but he wants to mix each instrument separately for a demo CD and claims that there is too much bleed through from other instruments into the mic he uses on my speaker. I am not a sound engineer but I told him hanging an SM57 directly down in front of my speaker isn't the way to do it. He refuses to use my Sennheiser e609 which is what I have always used in other bands and feel it's a better tool for the job.

I figure I will not waste my breath arguing with the drummer and install some resistors to pad down the output in order to send it to the mix board. I personally think this will sound horrible at the mix board and I feel it will effect the tone through the speaker by loading the amp differently now. Won't the resistors flatten out the response characteristic of the speaker? If I use large enough resistors where it isn't affecting the loading then I don't know if it will drive the mixer. I figure the mixer input impedance is <10k. Most I have seen are 1k-3k. I honestly care less about the mixer sound but I still want the sound through the speaker to be the same.

My amp is a custom built 12 watt push pull 6v6 tube amp. I was thinking of a resistive divider of 100 and 10 ohms, that should bring the 10v signal down to 1v or less. I was looking through my resistor drawer and I don't have anything less than 100 ohms so I don't want to order a 10 ohm resistor and pay for shipping or drive a half hour away to RadioShack if this is all for nothing and a waste of time. So if anyone has any experience doing this please share your results, it will be greatly appreciated. I am sure other tube amp users will find this information useful.

Thanks,

-bird

 

[famousmockingbird]

Nobody has tried this?

F^c$ it I will parallel enough resistors to get my 10 ohms and report back.

 

[Enzo]

10 ohms?

If you want to make a line out on your speaker leads, it ought to take more than 10 ohms.

Instead of 100/10, how about 10k over 1k, or even higher. Certainly 11k parallel your speaker will not affect the speaker tone in any way.

By its nature, this sort of circuit is unbalanced, so it will be a high Z output. If you want to feed it down a snake, then add a transformer after this.

Many amps have made a line out by divider across the speaker.

A straight one sounds flat to me, flat as in lifeless, not flat as in EQ. Good ones add some speaker emulation.

Here is an example of one I think works fairly well:
http://bmamps.com/Schematics/Peavey/Peavey_Windsor_Studio_Schematic.pdf

For LOTS more information or examples, google "speaker line out" and I'd skip right to the images, where there are schematics of all sorts of variations, both with and without emulation.

 

[famousmockingbird]

Hi Enzo,

10 ohms?

If you want to make a line out on your speaker leads, it ought to take more than 10 ohms, but still keep the 100 ohm on the "top" resistor. Total resistance 110R.

Instead of 100/10, how about 10k over 1k, or even higher. Certainly 11k parallel your speaker will not affect the speaker tone in any way.

I was thinking of a resistive divider of 100 and 10 ohms, that should bring the 10v signal down to 1v or less.

I have the 100 ohm resistors but not the 10 ohm ones which is why I said I might have enough to parallel down to 10 ohms. The "top" resistor is still 100R, total resistance is 110R.

The issue I was pointing out with using a larger resistance to not load down the amp is a high source resistance to the mixer, too low a resistance and it loads the amp and changes it's characteristic tone.


In the end you answered my question in that it won't sound good. And thank you for the tip on the speaker emulator!! It seems that the people that do this and don't like the sound end up getting a speaker emulator. Or from watching a few videos online of my favorite guitarists, they have a box built around the speaker mic so it doesn't get stage noise bled into it.

Something like this: Post Audio New ARF 50 Drum Mic Isolation Protection Tool Cymbal Bleed Blocker | eBay

 

[Enzo]

Yeah, I understood your resistors, I was just suggesting going for a higher resistance than 110 ohms, by a factor of 100 or more.

Any number of companies make speaker direct boxes. PV does, Behringer does, and others. Example:
Behr:
Behringer Ultra-G GI100 DI | Musician's Friend

They include an emulator circuit (with in/out switch) if you want, and it plugs in series with the speaker, between it and amp chassis.

https://peavey.com/products/index.cfm/item/895/117104

 

[cbdb]

The speaker is a huge part of the amp sound so don't expect it to sound the same.when ever I used a di ( almost always on the bass not the guitar ( unless you want it without distortion) ). I would also record a track with a mic and mix the 2. I have a couple of suggestions. Use a di straight out of the guitar and then use an amp simulator when mixing or do the usual studio thing, over dubs. The only real way to get everything separate.

 

[ArcticBreaze]

hi there, the sm57 is a very good instrument mic (or at least it was 20 years ago) its also very directional if he points it at the floor (ie hanging over the amp) lots of bleed can be expected. put it on a mic stand pointed at the speaker half way from the center to the outside edge, and in nice and close so gain is real high. that way when you turn it down bleed will be minimal.
I play harmonica and the speaker is a BIG part of the sound colour, i have tried speaker lvl di's and they just didn't sound the same.

other option is throw a mic INSIDE the speaker cab

 

[famousmockingbird]

Hi Arctic, I agree with you. That mic needs to be facing the speaker and not pointed down. I have not tried putting the mic inside the cab, I have seen people do this. Thanks I wouldn't have thought to try that. I have practice Thursday so we'll see how it goes. I put the resistive divider in but haven't had time to crank it up and see if the sound is okay. I used 110R. I did get some info on the mixer, input Z is 1.2k. They don't know what the sensitivity is.

Live gigs my speaker is micd but this is just for practice recordings.

 

[PB2]

Consider that power = V^2/R and therefore if your amp is 12W into 8 ohms,
then it will provide 12 * 8/110 = .87W into the series resistors you plan to
use. The 100 ohm should be 1W rating if you are going to crank the amp.

Nothing wrong with using a 110 ohm load, should work fine. The 10 ohm
shunt provides a low output impedance, good for driving long cables.

 

[PB2]

If you try the mic inside the speaker box you could try using acoustic foam, or even
mattress pad to line the inside and seal off the back to keep out the external sounds.
Seems like a simple solution. The foam lining the inside will help to reduce cavity
resonance sounds that you wouldn't have on the outside.

 

[famousmockingbird]

Hi, I used 100-10 divider, worked great. Sound from speaker appears to be the same to my ear. Sound through the direct out isn't as good as the speaker but works great for our practice.

Thanks everyone for the all the advice.

 

[Gnobuddy]

Sound through the direct out isn't as good as the speaker but works great for our practice.

Try adding a graphic EQ to the direct-out signal, and tweak to taste.

I think this can get you closer to the acoustic sound in front of the speaker, as you can mimic some of the characteristic frequency response peaks and troughs of the speaker (or, maybe, find a better-sounding EQ curve.)

But it won't get you all the way, as speaker harmonic distortion won't be present in the direct signal, EQ or otherwise. Whether this is important or not has a lot to do with how you play - if you normally drive the speaker hard (loud), it's contribution will matter more to your sound.

Personally, I'm quite skeptical of the speaker-impedance-emulating attenuators. At least some are based on total nonsense - and still sell well.

(If you immobilize a speaker voice coil with wax or epoxy or whatever, it turns into a plain ordinary lossy inductor as far as it's impedance goes. Once immobilized, it behaves nothing like an actual speaker does, down in the bass frequency region where the guitar amp is supposedly influenced by the actual speaker impedance. So: buyer beware of expensive commercial attenuators that contain immobilized speaker motors as part of their "mojo".)

-Gnobuddy

 

[rongon]

I once made a voltage divider line out from the speaker output of a Fender Deluxe Reverb amp (about 15 watts output). I chose a capacitor to shunt high frequencies to ground, which should give a -6dB/octave roll-off. Since most guitar speakers' responses take a steep dive above 5kHz, I chose the cap's value to be -3dB at 5kHz. I used it a couple of times into people's PA systems, and much to my surprise it sounded very good! Now, this was for a 'clean' tube amp sound, not overdriven to smithereens. But it did get enough of the amp's warmth into the mix.

You could use that low-tech direct signal going to one mixer channel in combination with a mic on the speaker into a different mixer channel. That way you could use the mic as a way to mix in some 'air' without introducing too much bleed from other instruments. That kind of thing is done all the time with acoustic bass players and acoustic guitars using piezo pickups. You use the piezo pickup to get the 'body' of the sound, but mix in some of the instrument's miked up sound to mellow out that pinched 'piezo sound.'

--

 

[MitchMev]

It looks like I'm late into the conversation but I figured I'd offer my two cents.

We practice "live" in our facility but he wants to mix each instrument separately for a demo CD and claims that there is too much bleed through from other instruments into the mic he uses on my speaker. I am not a sound engineer but I told him hanging an SM57 directly down in front of my speaker isn't the way to do it. He refuses to use my Sennheiser e609 which is what I have always used in other bands and feel it's a better tool for the job.

First, it's a demo CD and he's worried about bleed between instruments? Hell, most professional live recordings have bleed between the instruments.

Second, an SM57 should do the job just fine, although the Sennheiser will also work fine. The SM57 is a cardioid meaning it picks up no sound along the axis behind the mic and little sound from other directions behind the mic. The Sennheiser is a supercardioid which means it does pick up signal along the axis behind the mic. Therefore if you're worried about bleed, the SM57 may be a better option, but only if it's pointed directly at the cone (NOT down towards the ground).

 

[famousmockingbird]

Hi guys!

The EQ on the output could help a lot "dial" in the tone, I do have a small Alnico driver that does add to the mojo when brought to it's threshold. I was thinking of adding a cap to smooth things out but the EQ could do the same.

And looking at the patterns the SM57 should work perfectly if it is pointed at the speaker. I kinda liked the tone of the Sennheiser I have and I have never had a problem with bleed through. Oh well the direct out is working good so the next step is to tame the highs with a cap or EQ, I think the board I am going into has the standard tone controls.

 

[bruce egnater]

You do know there are many good "cabinet simulating" direct boxes to do what you wish. you can't just use a padded output direct because you are bypassing the speaker. It will sound horrible.

 

[famousmockingbird]

You do know there are many good "cabinet simulating" direct boxes to do what you wish. you can't just use a padded output direct because you are bypassing the speaker. It will sound horrible.

Hi Bruce, I know it won't sound as good as through the speaker. This direct out will only be used for band practice evaluation purposes, but I still would like to get the best tone if possible without having to buy anything. As for now it serves it's purpose good enough but I will say I won't be using it for a professional recording or gigs.

I will try a cap first and shoot for a similar range and roll-off as Rongon suggested. My amp has a 8k plate to plate OPT and I run push pull class AB pentode mode with 6V6's, zero global feedback or any sort of localized feedback in the output section. I am running Va=350. The output impedance is most likely pretty high, ~20 so I am looking for a cap value of 1uF-2uF shunted across the output? I guess I will start there and measure response on the scope at the output. Like I said before The board has adjustments for lows, mids and highs so I can roughly dial in the tone with that too. I think the cap is still a good idea though.

 

[Gnobuddy]

...I am looking for a cap value of 1uF-2uF shunted across the output?

Danger, Will Robinson! I wouldn't put a cap directly across the amp output (speaker), I think that may harm the amp. Sometimes amplifiers will become unstable and burst into oscillation if the output is loaded with a capacitor.

What will work, is to put a two-resistor voltage divider (R1,R2) across the guitar speaker. The two resistors are chosen to divide the voltage down, so you get a reasonable line level across one of them (R1). Then put a capacitor (C1) across that resistor, to create a low-pass filter. The amp itself won't "see" the capacitor, because the other resistor (R2) is in between the amp's output, and the capacitor.

The attached schematic shows the plan. The 330 ohm resistor (R1), along with the 0.1 uF capacitor, forms a low-pass filter with a -3 dB corner at about 5 KHz. The 3.3k resistor (R2), along with the 330 ohm one, divides down the speaker voltage, to generate a line-level signal across the 330 ohm resistor.

I'm guessing a little as to the value of the 3.3k resistor - its value depends on how much signal the mixer wants at it's line-level input, and how loud you like to play through your guitar amp. But if 3.3k turns out to be the wrong value, you can easily change this to suit you, or replace it with a 5k trimpot in series with a 1k resistor, so you can adjust to suit.

If you need to change the 3.3 k resistor, here's how to do it: play through the amp at your intended practice volume, with the line-out signal (across the 330 ohm resistor) running to a line-level input on the mixer. If the mixer gain needs to be turned way up to get adequate signal, then *reduce* the 3.3k resistor in the attenuator, so you get a hotter signal to the mixer. I would try halving it, say 1.5k, which will give you roughly 6 dB more signal.

Obviously, do the opposite if the signal to the mixer is too "hot" - in that case, *increase* the 3.3k resistor. I would try 6.8k, which will halve the signal (-6 dB) to the mixer.

I hope this works for you - I've had good results with the two resistors followed by a Danelectro Fish-n-Chips graphic EQ pedal, but I've never tried the single first-order low-pass filter at 5 KHz.

Oh yeah - half watt resistors will be just fine, unless you plan on using a 100 watt guitar amp.

The capacitor (C1) only sees a line-level signal, so use any inexpensive non-polarized one you have lying around in the junk box. Polyester, Mylar, film, ceramic, whatever. As long as it's rated for at least 10 volts, you have a nice big safety margin.

One last thing, just to make sure: the guitar speaker must remain connected to the guitar amp the whole time. It provides the 8 ohm load that the amp must have at all times. Tube amps can get seriously unhappy if the load is removed from their output. You probably know this, so I apologize if I'm telling you something you already know. I just wanted your amp to stay safe and healthy.

This is a simple circuit which only provides an unbalanced signal to the mixer. To avoid hum, make sure the guitar amp and mixer are plugged into the same mains circuit (adjacent wall outlets, or adjacent outlets on the same power strip).

-Gnobuddy