Ruby amp /Grace /Big Daddy question

[Matt420740]

Hi all,

My first post here.  I have a question regarding the LM386 based overdrives over at runoffgroove.   I've built a couple Rubys, for me and a friend.  Been looking at the overdrives, and the LM386 datasheets, and I can't figure out what is stopping the Grace or Big Daddy from being amps themselves instead of overdrives.  Am I missing something?   The output of the Ruby with a LM386N-3 should be about a 1/2 watt, so are the grace and big daddy also churning out around a 1/2 watt?  If not, could you please explain what it is about the circuit that makes the Ruby and Gem amplifiers, while the Grace and Big Daddy are overdrives.  Thanks.

[Gurner]

They are all amplifiers - in so much as they increase the incoming signal. What differentiates the Grace & Big Daddy from say the ruby...is they actively set out to distort/clip the incoming signal, whereas say the Ruby, can be played clean.

The other thing that differentiates, is that the ruby is audible audio amplifier - in that it's attached to a low impedance inductive load (high falutin term for a speaker!)

There'd be nothing to stop you connecting the Grace or big daddy to a speaker, becuase the LM386 is intended to drive speakers (vs the 100k pot you see on the likes of the Big Daddy).

*******Warning******Over zealous use of the phrase 'zobel network' about to be dropped in..

If you do decide to attach a speaker to the big daddy...just remember to use a zobel network on the output (high falutin term for the 10 ohm & 47nf cap attached to pin 5 of the LM386!)

[ItZaLLgOOd]

I would also use the Ruby's larger output cap.  100uf - 220uf  instead of 0.1uf

[Matt420740]

Thanks so much.  I do have a few more questions though.  Never heard of the "zobel network"  Some searches led me to what a zobel network does - "zobel networks can be used to make the impedance a loudspeaker presents to its amplifier output appear as a steady resistance."   

I guess my question is why does it need to be used in the ruby/little gem, but not in other circuits I've seen, and what would happen without it? 

Another question, as you guys probably know, the 386 can't handle much input signal without clipping.  When I first built a Little gem I found this out, as I had to turn the volume knob of my guitar down just to get a clean signal.  The Ruby on the other hand has a volume control before the 386, limiting the amount of signal to the chip, so even with the volume knob of my guitar wide open, I can get a clean signal.    I love the tone of the Ruby, but it is way too loud cranked through my 4x12 cab.  Would it be possible for me to build the ruby as is, and place the master volume of the little gem on the output?  If so, is there a more common value pot that would work instead of a 25ohm, as I don't enjoy paying $4 apiece for them. 

[Jhouse]

THIS schematic I was looking at uses a 10k linear pot for its volume. I'm not sure about moving the volume pot around. Haha, you could do something really silly and add a 100k audio pot at the end of the circuit right before the speaker, like a second volume pot.

[Jhouse]

Unfortunately, my crackberry won't allow me to edit my previous post. Er. On second thought, maybe the idea of a master volume isn't a bad idea. That way you could still overdrive it, but keep it at a lower volume. I don't know what size pot you would need but trying a 100k might not be a bad starting point.

[Matt420740]

Thanks Jhouse,

If you look at the Little Gem amp, which is very similar to the Ruby, you'll notice it has a 25ohm master volume pot on the output before the speaker.  I'm not wanting to move the Ruby's 10K volume, I'm wanting to add the Little Gem's Master Volume to the Ruby.   It seems like it would be fine to do, but I'm a noob and want to make sure.  Also, if it is fine to do that, is there any value pot besides a 25ohm that will work well?  ?

[Jhouse]

Try a 50k or a 100k audio. That might end up being overkill honestly, but at least it's something.

[Matt420740]

 I've tried a 1K, but it goes from being loud to whisper quite in about 1mm of pot travel.  With a 25ohm pot the level is cut nice and smooth, but damn they are expensive and hard to come by.   Since the 1K seems to be too sensitive for the volume, wouldn't a 100K be even worse?

I've also considered using resistors in place of the 25ohm pot, with a switch to choose between full output, or bedroom levels.  Am I right in thinking that I could measure the resistance between the pins of the 25ohm pot at my desired output level, and recreate that using 2 resistors, essentially making a voltage divider?  Or am I way off here

[Jhouse]

100k would definitely be overkill. I don't even know what I was thinking. Yeah, you can put a resistor or several resistors across the lugs of the potentiometer and that would lower its value. It's the equivalent of using resistors in parallel.

Someone correct me if i'm wrong, but I think you can just put a 270 ohm pot across lugs 1 and 3 on that pot and make it 25 ohm.

[PRR]

> Someone correct me if i'm wrong, but I think you can just put a 270 ohm pot across lugs 1 and 3 on that pot and make it 25 ohm.

Consider yourself corrected.

1) 1K||270 makes 212.6 ohms, not 25 ohms.... I think your zero-key got excited, because 1K||27 does make 26.3 which is equal to 25 for our purpose.

2) You've slogged the end-to-end resistance lower, but we want a low resistance to the load via the WIPER. When 1K pot is half-way, the wiper impedance is still 500||500 or 250 ohms. Driving an 8 ohm load, this gives half (for pot-setting) and 8/258 or 0.03, result is 0.015, which is VERY weak.

Whereas a true 25 ohm pot, set half-way, has 6.25 ohms wiper impedance. So the output is half of 8/14.24 or about 0.28 or about a quarter, a good mid-turn level.

Yes, 25 ohm pots have become rare.

Loudspeaker volume controls are readily available as in-wall accessories. They start at 25 Watts, which seems like a waste, but there is not much demand for reducing part-Watt amplifiers.

If you can stand the whole Watt, don't throw-away power after you make it. Control volume before the LM386. Here you can also use common 10K-100K pots.

However in this forum there is some call for OVER-driving the LM386 for strong flavor, but only allowing maybe a tenth-Watt into the speaker so the neighbors don't bang on the ceiling. You can try powering the LM386 with six to three 1.5V AA batteries. You can stick in 33 to 220 ohms in series between '386 output cap and speaker.

[Matt420740]

Thanks for the reply PRR.  I think jhouse misunderstood my post before his.  I wasn't asking if I could lower the value of the pot with resistors, as I know that won't work in this situation.   I was asking about actually replacing the pot with resistors.  Seems to me the 25ohm pot on the little gem is just a voltage divider.  I do have one 25 ohm pot on hand, so couldn't I set it to the level I desire, measure the resistance between legs 1 and 2, and 2 and 3 of the pot, and recreate that using 2 resistors?  I'd of course need to use 1 watt or higher resistors since its on the output I assume.  I don't really have room on my enclosure anywhere to add another pot without it looking funny, but I can add a switch.  My plan is to use 2 resistors to form a voltage divider instead of using the pot, and using a SPDT to switch from full output to the attenuated level.    Am I thinking right here?
 

[Gurner]

I tend to think it's weird to generate the power & then waste a percentage of it, but if you really do want to go this route, then just experiement with some low value resistors in series with the load.

For example, an 8 ohm resistor in seris with your speaker (assumingyour speaker is 8 Ohms) would see half the signal voltage dropped across the resistor & half by the speaker


A working example (assume 8 Ohms speaker)....

No resistor
signal voltage across speaker = 8V peak to peak = 4V peak = 2.828V RMS

current through speaker = 2.828/8 = 353mA

Power = 1W (give or take)


8 Ohm resistor in series...
signal voltage across speaker = 4V peak to peak = 2V peak = 1.414V RMS

current through speaker = 1.414/8 = 177mA

Power = 250mW

So by using an 8 ohm resistor in series with the load you've reduced your power output by a factor of 4 - a very inefficient way to lower the output volume though! (I shall forward your username to Greenpeace!)

.....you'd need to use at least 1/4W resistors. (but I'd err on the side of 1/2W resistors)