Building A Practice Amp

[MetalUpYerEye]

 
I want to build a dual Ruby with a 2 band tonestack. I've heard that if you add a tonestack to the Ruby you also need to add a small gain/signal boost before the tonestack. So far this is what i'm planning on doing from guitar to speaker.

Jack Orman's AMZ Mini-booster to a simple tonestack designed by Aron Nelson into the Dual Ruby.

Jack Orman's AMZ Mini-booster
http://www.generalguitargadgets.com/diagrams/joam_sc_amz.gif
to
Arons simple tonestack
http://www.lorenburton.com/images/ruby1.gif
to
Dual Ruby
http://home.comcast.net/~knowitallrocks/temp_images/dual_ruby1.pdf

Also the dual Ruby has something on its layout that I haven't seen as of yet, described on the layout as a "voltage regulator." On the layout it shares the same shape as the transistors, but offers no part name or number. What is this and how does it incorporate into the design of the dual Ruby as opposed to the Ruby (which doesn't have one). I'll probably be putting all of this inside a small cab and using it to drive a 12" Celestion Rocket50 (50watts, 8ohms).

Is it a good idea? Bad? Insane? Let me know your thoughts.

[mdh]

The voltage regulator would probably be a 7809. It comes in at least a couple of flavors (78L09 in TO-92 case for up to 100mA and 7809 in a TO-220 for up to 1A w/ heatsink, I believe, but check data sheets). I'm not sure how much current the Dual Ruby would draw. Voltage regulators are used to... well, regulate voltage.  If you're running the thing off of a DC wall-wart, you may wind up with more than 9V DC. The regulator is there to keep it at 9V, as long as the supply voltage is greater than or equal to 9V.

You might keep in mind that the Dual Ruby will probably be happiest with a 16 ohm load, due to the bridged configuration of the chips (if I'm understanding what this thing is). I built something like this a while back, and one of the chips ran very hot with 8 ohm and 12 ohm loads. Also, you might want to consider running the whole thing on 12V instead of 9V (in which case you would want a 7812 regulator if at all).

[pyrop]

Dual Ruby
http://home.comcast.net/~knowitallrocks/temp_images/dual_ruby1.pdf

That layout doesn't look quite right. It only has signal input to one chip.
The wiper of the volume pot (2) should goto pin 3 & pin 2 should goto gnd of the first/lefthand chip.
Leave the trace from l/h chip pin 3 to r/h chip 2.

pyrop

[MetalUpYerEye]

Thanks MDH, I was able to find a 7809 in a TO-220 case fairly easily on Mouser. I want to run it off 6 AA batteries in series, with a wall-wart option available in order to keep it portable but not waste thousands of batteries while enjoying it at home.

Pyrop, let me get this straight - the middle lug of the volume pot goes to pin 3 of the lefthand IC. Pin 2 of the lefthand IC goes to ground. Right?

I shouldn't have any problem running the Jack Orman Booster into the tonestack and then into the Dual Ruby should I? If my assumptions are correct, i'll end up with a Gain1, Gain2, Bass, Treble, Master Volume, is this correct? Reminds me the config on a Marshall practice amp I used to have...

[Mark Hammer]

The tonestack is passive, hence it works by bleeding signal.

A different way to alter the tone is to work with the compensation loop in the 386 itself, where the selective gain properties of the amp are altered, not the level of the signal reaching it.

You will note that the gain of the 386 can be changed by closing the loop between pins 1 and 8.  In the Ruby, a 1k pot is placed there, and gain is increased as the resistance gets smaller.  The 386 datasheets actually suggest a 10uf cap (to block stray DC out of that loop, I gather) in series with that pot.

What was drawn to my attention in an article in Nuts and Volts a couple of years ago was that the blocking cap doesn't HAVE to be one that lets all frequencies through.  For instance, if the 10uf cap was, say, .047uf, whatever bass rolloff was imposed by the 10uf cap value would now be moved upwards much higher, such that the maximum gain would only be applied to mids and highs.

Extending this a little further, imagine the following:

1) In the feedback loop is a 10uf cap and a 2k2 fixed resistor.

2) In parallel with this in the same loop is a 5k pot and a smaller-value cap (perhaps .01uf)

When the 5k pot is at maximum resistance, there is no advantage to high end following that path.  Since the 10uf cap lets *everything* through, the gain for the full spectrum is whatever the 2k2 resistor sets it too.  As the 5k pot goes lower and lower in resistance,whatever passes through the smaller-value cap now has more and more gain applied to it until high-freqs end up having much more gain than the rest of the spectrum.

Note that, with nothing in the loop, gain is 20.  With a 1k2 resistor, gain is 50.  With pins 1 and 8 linked by a cap with NO resistance, gain is x200 for whatever frequencies can pass.  This, in fact, provides probably more treble boost than a passive stack would.  What it doesn't do is provide cut.  But you know, since you're probably going to use a little speaker, turning down the treble is not what you're aiming for anyways.

What about bass?  The same datasheet indicates that bass boost may be achieved by linking the output (pin 5) back to pin 1, via a series resistor/capacitor network.  My guess is that the same thing applies here.  The range where the boost is applied rests on the cap value chosen, and the amount of boost applied rests on the resistance value.

Get hold of the datasheets and learn more.  You may be able to get more interesting sounds from this little chip than you thought.

Finally, consider a small bypass cap on the volume pot to get a "bright" function at lower volumes.

[mdh]

Pyrop, let me get this straight - the middle lug of the volume pot goes to pin 3 of the lefthand IC. Pin 2 of the lefthand IC goes to ground. Right?

Yes, and pin 3 of IC1 remains connected to pin 2 of IC2 (numbering left to right). If you can read a schematic, I would recommend comparing the connections in this layout with those in the Little Gem Mk II, paying particular attention to the pins of the ICs (bottom of the page at http://runoffgroove.com/littlegem.html). There will be some differences in the passive components and the location of the volume control, and the Little Gem Mk II leaves out the gain pots, but it should help you make sure you don't build a paperweight.

[MetalUpYerEye]

Ok, Mr. Hammer, lets test out my powers of understanding...

http://i40.photobucket.com/albums/e215/metalupyereye/fig1.jpg

This figure shows the setup you described for tone boost using a 10uf cap and 2k2 resistor in parallel with a 5k pot and .047uf cap from pins 1 to 8 of the opamp. Does it look correct?

http://i40.photobucket.com/albums/e215/metalupyereye/fig2.jpg

Figure 2 here is what I understand to be a bass boost, as described by you, using the same layout; a 10uf cap and 2k2 resistor in parallel with a 5k pot and bass rolloff cap from pins 1 to 5 of the opamp. Is it correct?

I also dont know what freq's are rolled off by what value caps, as I dont have much time now i'll have to look later to find what value will fit for a bass boost control. I'll be using a 12" 50 watt speaker so I want as much control over treble as I do bass.

Py, thanks for the suggestion on the LGMK2 schematic, usually i'll use a schematic to compare with the layout but there was no schematic for the Dual Ruby. I never even thought to use the LGMK2 schematic...

[Connoisseur of Distortion]

the regulator is probably going to need more than 9V...

probably closer to 12-13...

[Mark Hammer]

MUYE,

You've got it.

I'm not doubting you'll get *something* out of a 12" speaker, but generally the higher the power rating of a speaker, the more sluggish it is, and the more power it needs to get going.  Some speakers undoubtedly have higher power ratings because the voice coil is thicker wire, or because it can dissipate the heat buildup in the coil during higher power use more efficiently, but most will also use stiffer coils and speaker cones to some extent to achieve the higher ratings.  This means you may well get less audible output from your 50W speaker than you would from one rated much lower.

As for where the boost occurs, there I can't help you.  On the other hand, it is a safe assumption that the 10uf value was intended to pass at least 20hz (and probably a bit higher).  The rolloff will move up an octave for each halving of the cap value.  So, if the stock 10uf value cut bass at 20hz, then 5uf would get you 40hz, 2.5uf = 80hz, 1.25uf = 160hz, .625uf = 320hz, .3125uf = 640hz, and .156uf = 1280hz.  Remember, that assumes that 10uf gets you 20hz.  It may well get you something much lower.  With that in mind, let me revise the recommendation and suggest .047uf as a starting value, and work out from there accordingly.

[MetalUpYerEye]

So .047 is a good place to start for the Treble control, any idea where I should start for the Bass?

[Mark Hammer]

The datasheet shows a 10k resistor in series with a .033uf cap, cap on the output, resistor on pin 1.  My guess is that this ought to be an on-off thing, rather than variable control.  Keep in mind this baby doesn't *like* bass, because it has little headroom or oomph.  I suppose you could stick a .033uf cap in series with a .047uf cap.  Wire up a 3-position toggle (centre-off) so that the common goes between the two caps, one goes to the junction of the resistor and one of the caps, and the other goes to the junction of pin 5 and the other cap.  In the middle, the capacitance is .019uf, to one side .033, to the other side .047.  Whatever .033 was supposed to get you, this will get you a boost lower down, and a boost that extends somewhat higher up as well.  Essentially the same as a bass boost with 3 different turnover frequencies.