Tweed Champ Line Out

[mcgruff]

Some "reactive" dummy loads seem to turn into resistive loads at high levels of attenuation. Beyond a certain point, the guitar tone gets flatter and flatter (a reactive load is more dynamic).

People I've spoken with Webers say you lose some tone at extreme amounts of attenuation - that might be the reason. However, I don't understand all the circuit details and I could be wrong. I should have said: "AFAIK The Weber mass series are pretty good but only for small amounts of attentuation".

[Tom Phillips]

That's the ongoing issue with attenuators in general. There are so many other things in play that the goal of perfect cranked sound at low volume levels can't seem to be duplicated.

[Steve Conner]

Some "reactive" dummy loads seem to turn into resistive loads at high levels of attenuation.

If the load is too inductive, it might not load the OT heavily enough at high frequencies. This could result in destruction from voltage spikes when heavily overdriven. So, I guess many makers of attenuators add a bit of resistive load as a precaution.

I've got pretty good results using a simple dummy load with a line output feeding into a digital parametric EQ, digital reverb and headphone amp. I adjusted the EQ by ear to sound like my favourite guitar speaker, and the reverb put it in a "room". For recording, you can get software that'll do the same thing: google "cabinet impulse".

WRT the original topic: The speaker is a huge part of the amp's "original tone". Connect the best guitar amp in the world to a hi-fi speaker, it'll sound dreadful. Connect a little transistor amp to a Marshall 4x12, it'll sound surprisingly good.

So it really makes no sense to speak of "preserving the original tone" with a line-out circuit, unless you took an impulse response of your speaker and used a DSP to apply it to the line-out.

[mcgruff]

So it really makes no sense to speak of "preserving the original tone" with a line-out circuit, unless you took an impulse response of your speaker and used a DSP to apply it to the line-out.

The problem I'm specifically trying to solve is how to take the speaker completely out of the equation but preserve all the tone normally fed into the speaker. You're right to say that the speaker will have to be added back in at some point.

For example, suppose an amp which you want to play cranked is too loud for a small venue, or for home practice. If a line out is fed to a slave or PA the volume can be turned down as low as you like. At home you could plug straight in to a DAW and monitor either through your studio monitors or on headphones. As you say, you'd want to use a cab sim; the reactive load just mimics the speaker's impedance curve not its frequency response.

IMO a reactive load can sound noticeably better than a simple resistor. For example, here's a clip originally posted in the Aiken reactive load thread (I've normalised the three different samples so they can be compared properly). The pure resistive load does provide a "fair" sound but it's flatter and a bit compressed-sounding. The reactive load is very close to the original.

[overtone]

Living in a city apartment I stumbled along many winding paths like the OP, line out mods, a Weber type speaker motor thingy for headphones, blankets, huge iso-boxes under the roof with cables "everywhere".

Fed up with all this nonsense my girlfriend went out and bought me a Palmer PDI03 Speaker Simulator.


The PDI03 gets sent to the DAW for night time use with EQ rips + time based effects, similar to Steve's method.
Note: the PDI03 is not an attenuator, it is either silence or only the inductor line-out in parallel with your speakers.

Getting a signal out before the speaker is no big issue. What happens after that is.
Now I just accept that it sounds and feels different and deal with it.

By the way, around five years ago I built that champ kit in the OP and it sounded really good.

Best,
tony

[Steve A.]

The problem I'm specifically trying to solve is how to take the speaker completely out of the equation but preserve all the tone normally fed into the speaker. You're right to say that the speaker will have to be added back in at some point.

As several people here have already said the speaker is an integral part of the tweed Champ sound. The amp is so basic that if you remove one element it just ain't the same. We've had people here looking to put a master volume on the amp- yes, you can do it but with the MV turned down you've lost the tweed Champ vibe.

Take away the speaker and the sound from the amp circuit is not that great. You can filter it and attenuate it all you want but you are not going to get a good line out signal from it. When Clapton used these amps recording Layla they used mikes- with all of the different angles and all of the different mikes you have a lot of control over the sound.

Mesa Boogie amps usually have a fairly good line out signal which is possible since a large portion of their sound is in the preamp. Add some filter circuitry and you are good to go!

Another idea- build an isolation box for your tweed champ and use a microphone.

Just my own 2 cents...

Living in a city apartment I stumbled along many winding paths like the OP, line out mods, a Weber type speaker motor thingy for headphones, blankets, huge iso-boxes under the roof with cables "everywhere". Fed up with all this nonsense my girlfriend went out and bought me a Palmer PDI03 Speaker Simulator...

By the way, around five years ago I built that champ kit in the OP and it sounded really good.

By "it sounded really good" do you mean with the PDI03 or with the cabinet and speaker (the cabinet is very important part of the sound as well.) I'm sure it sounded decent with the PDI03 but I suspect that the sound was not as full as the combo amp. Correct me if I'm wrong.

Steve Ahola

[mcgruff]

By the way, around five years ago I built that champ kit in the OP and it sounded really good.

Good to know If you've got any clips of the WF-55 through the Palmer I'd love to hear it.

The Palmer boxes look very good. I thought about buying one but the parts for the Aiken load will only cost me about ｣100 (｣300 for a PDI-03). Plus it's kind of cool to make things yourself. You learn so much more than if you just go out and buy something.

On the other hand, the risk of electrocution goes up... I guess everybody reading this will know to be careful though.

[mcgruff]

Take away the speaker and the sound from the amp circuit is not that great. You can filter it and attenuate it all you want but you are not going to get a good line out signal from it. When Clapton used these amps recording Layla they used mikes- with all of the different angles and all of the different mikes you have a lot of control over the sound

With the Aiken design, it looks like you can match the impedance curve of different cabs by tweaking the values of various components. That's something I maybe need to look at for an authentic Champ tone.

It should also (in theory) be possible to get fairly close to the frequency response of the speaker using an impulse response and/or EQ. I do have a Tweed Champ IR which sounds quite mid-heavy and Champy. IR's can even simulate different mic positions.

I don't know. Usually I'm dead against modelling because it can never capture the rich, responsiveness of the real amp, just two-dimensional snapshots of a frozen point in time. This is an experiment - when it's finally built I'll post some clips so you can judge for yourself.

[dai h.]

Living in a city apartment I stumbled along many winding paths like the OP, line out mods, a Weber type speaker motor thingy for headphones, blankets, huge iso-boxes under the roof with cables "everywhere".

Fed up with all this nonsense my girlfriend went out and bought me a Palmer PDI03 Speaker Simulator.
[ATTACH=CONFIG]23131[/ATTACH]

The PDI03 gets sent to the DAW for night time use with EQ rips + time based effects, similar to Steve's method.
Note: the PDI03 is not an attenuator, it is either silence or only the inductor line-out in parallel with your speakers.

Getting a signal out before the speaker is no big issue. What happens after that is.
Now I just accept that it sounds and feels different and deal with it.

(excuse my interjection if you're already aware of this, but) the Palmer output (since they are transformer coupled) should be affected by the input impedance they see, so you might be able to shape the sound more favorably by tweaking the load with a resistor. If a transformer out "sees" an output higher than it's designed to, they can "ring" (have a resonant peak in towards the highs which usually gets larger the less the load (and more undamped) they see). The undamped resonance (from my understanding) can make the sound thinner and the excess signal voltage where the resonance is will distort and intermodulate with the signal which (I think) is especially bad when the preamp the Palmer out is seeing is a typical modern preamp topology where it sounds fine as long it's not overloaded but sounds nasty when it is. I have a PDI-03, and I remember plugging the balanced filtered out into a balanced mic pre input (2.8k input impedance) and sounding bad (so I just used the unbalanced filtered out (10k output Z) into the insert in (10k input Z), and that sounded more correct (but at the time I had no clue why and just went w/the insert in since it simply sounded better).

Also (from what I've read) it's fairly typical for recorded guitar to have a bit of high end rolled off (LPRF'ed--or low pass filtered, i.e. a high cut), so using an EQ which can provide that could also help make things sound more "correct".

[Steve A.]

It should also (in theory) be possible to get fairly close to the frequency response of the speaker using an impulse response and/or EQ. I do have a Tweed Champ IR which sounds quite mid-heavy and Champy. IR's can even simulate different mic positions.

I don't know. Usually I'm dead against modelling because it can never capture the rich, responsiveness of the real amp, just two-dimensional snapshots of a frozen point in time. This is an experiment - when it's finally built I'll post some clips so you can judge for yourself.

I assume that by "IR" you are referring to an iRig. The computing power of an iPad is nothing like a real computer. For direct to digital recording the Behringer UCG102 is really great for $40. It comes with 3 Guitar Rig amp models but I use it with the full Native Instruments Guitar Rig program which is so much more powerful. The UCG102 plugs into a USB port and has a guitar input jack and a headphone output jack. I suspect that digital modeling is used a LOT in recording studios these days- they can record a track with just the unprocessed guitar signal and then experiment with plug-ins until they get the sound they want.

Behringer Guitar Link UCG102 USB Interface | Musician's Friend

All of the electronic gadgets mentioned so far do not deal with the ambience of the tweed champ in which the cabinet is a very important part (I think of those old amps as being like music boxes.) Perhaps you heard the story about the Fender Custom Shop trying to duplicate EC's favorite tweed twin:

"Eric had what he considered to be the Holy Grail of amps (his favorite 1958 Fender tweed Twin amp). But since he only had one he was afraid of damaging it on tour" says Del Breckenfeld Fender's director of artists relations. "Our mission was to clone that amp, which was a quite difficult task. We first analysed Eric's vintage amp to the specs then we searched for old parts. After all that, it still didn't sound right. At that point, John Page suggested old pine wood. We found some that came from an old church's floor and that made the difference. We nailed* it and we built a total of three. We built them and then subjected them to the acid test… Eric's ear. He loved them and declared them to be the exact replicas and gave one to BB King as a gift.

* I hope that Mr. Breckenfeld is speaking figuratively since I think that glue would work a lot better than nails.

Eric Clapton - ClaptonWeb.com - E.C. Mainline Florida

If you need to practice at low or no volumes all of the solutions mentioned so far should work. But for pristine 24bit/96khz digital recordings of your future tweed champ I really doubt that you can beat a high quality mike (or even an SM57.) If you can get it work with attenuators, etc., more power to you- I will stand corrected. Perhaps you can post your recordings with whatever rig you end up using so that we can hear the results.

Good luck!

Steve Ahola

[mcgruff]

I assume that by "IR" you are referring to an iRig.

Sorry that wasn't clear. I meant "impulse response".

Good point about the cabinet. I could stand a little deviation from a strictly authentic Tweed tone so long as it's still a responsive, dynamic and "musical" sound.

[overtone]

By "it sounded really good" do you mean with the PDI03 or with the cabinet and speaker (the cabinet is very important part of the sound as well.) I'm sure it sounded decent with the PDI03 but I suspect that the sound was not as full as the combo amp. Correct me if I'm wrong.

Steve Ahola

Steve, the little champ kit sounded great straight into an open-back 2x12 on firing it up. I remember my girlfriend was in the bath and said "Wow, that sounds just like a recording" or words to that effect, which I thought was sweet from someone who has no idea of music. She then bought a Fender tweed 1 x 12 cab for it. I don't recall ever playing that amp through the Palmer.

In answer to dai h.: we discussed Z loads and the Palmer in 2011 but your point about varying the amp output load never occurred to me, thanks for that. I recently rigged up a 200W 16 Ohm load for the Palmer + an amp that only has 16R, but never thought to use the new load for a mismatch! Good tip.

With regard to the EQ: I push into the Palmer with an unusually bright amp setting to try to get the feel of the amp "punched through" the unit and then make multiple narrow band cuts around 6kHz. Seems to me that a little of the direct inject hash is always there, so yes I always prefer a cab - but the neighbors tend not to so much.

Now back to the thread,

best, tony

[dai h.]

In answer to dai h.: we discussed Z loads and the Palmer in 2011 but your point about varying the amp output load never occurred to me, thanks for that. I recently rigged up a 200W 16 Ohm load for the Palmer + an amp that only has 16R, but never thought to use the new load for a mismatch! Good tip.

With regard to the EQ: I push into the Palmer with an unusually bright amp setting to try to get the feel of the amp "punched through" the unit and then make multiple narrow band cuts around 6kHz. Seems to me that a little of the direct inject hash is always there, so yes I always prefer a cab - but the neighbors tend not to so much.

tony(overtone),

(sorry I wasn't clear) I meant varying the load the *signal* output sees, not the amplifier speaker output.

This is an example (600 ohm to 600 ohm small signal transformer with source impedance and terminating impedance varied) :

http://fixerhpa.web.fc2.com/sokutei/...71_inout_z.gif

If you look at the upper two curves where the secondary is lightly loaded, a rise in the response can be seen. That is what I'm referring to (my understanding is that a steep rise is to be generally avoided for the aforementioned reasons). My thinking is that getting the high freq. response to roll off a bit more (as with the orange curve which is 40ohms driving impedance, 600 ohm terminated) would be better esp. for trying to achieve a speaker-like sound direct.

[overtone]

dai h., I could not access the image but get the gist of it.

Back when I adjusted the loading I was using the balanced output.
Target was Z-source 532 Ohms / Z-load of 2k4 by using a 4k7 resistor.
With the decade box and a lot of careful listening I found 4k2 did the trick.

I will look again at why I did not use the unbalanced filtered out. Maybe i missed something.
The mismatched amp output load intriges me still Worth a bash too.

Best, tony

[dai h.]

hmm... you're right I get an error message for some reason. FWIW (hope this works), this is the web page the image is from (basically a documentation of experiments done w/a low-cost 600:600 small signal transformer) :

(4th image from top. Title of page in English: "Freq. response and dist. characteristics and such of Sansui ST-71 600Ω:600Ω")

サンスイ　ST-71 600Ω:600Ωトランスのf特とか歪とか

Back when I adjusted the loading I was using the balanced output.
Target was Z-source 532 Ohms / Z-load of 2k4 by using a 4k7 resistor.
With the decade box and a lot of careful listening I found 4k2 did the trick.

ah, so you've already tinkered w/the loading. Well, hopefully the info is of use to someone else who has never seen it. (The issue of loading affecting the freq. response when a transformer is in the signal path may have been "general knowledge" that was lost somewhat as transformers became less common. I think it can be very useful when dealing w/anything w/a transformer for example such as the common SM57/58. From what I gather some modern mic pres have a rather high input Z (5k-10k) and this can affect the sound of them for the worse (can be very simply amended w/a resistor to load and damp the output more correctly).)