Another simple amp question

This is the place for any magazine-related discussions that don't fit in any of the column discussion boards below.
User avatar
CeaSaR
Posts: 1624
Joined: Sat Nov 08, 2003 1:01 am
Location: Phoenixville, PA USA
Contact:

Another simple amp question

Post by CeaSaR » Sun Nov 19, 2017 2:33 pm

Got a new project I'd like to complete before Christmas this year. Rather simple and old school, but I need a bit of advice. Here goes:

The Dollar Store sells these dirt cheap amplified speakers that use a LM4871 Boomer audio amp. The supplied driver (speaker) sucks for audio quality, but the amp board is actually quite nice, and with a real driver has amazing sound. 3 watts @ 4 ohms @ 5volts. Not bad for a buck. It's mono, so no problems there, just use 2 for stereo if needed.

Now, I want to make a small portable/wearable guitar amp for my oldest progeny. I figure it needs a preamp due to the possibility of extremely low input signal possible from an electric guitar, so why not a standard Common-Emitter amplifier, with maybe a Diode fuzz between the two, almost acting like a gain distortion - the harder you play, the more distortion you get, but that is ancillary.

Biggest problem I get is trying to figure out the biasing in order to get it to work well without draining the battery too quick. I've been using: http://www.falstad.com/circuit/e-ceamp.html since my computers with spice on them are down for a while.

I don't have the exact figures I ended up with, but know that the supply voltage is limited to 5 volts, and I'd like to get about 1.25 - 2 volt swing, if possible. The only way this webpage let me come close is by biasing with a 660/330 ohm input network, a 1k/470 ohm output, and a minimum 25k output/gnd. Sounds like it'll gobble up a battery pretty quick. I plan on using a. 2N3904 as that is what I have most of.

Any suggestions?

-CeaSaR
Hey, what do I know?

dyarker
Posts: 1640
Joined: Fri Aug 22, 2003 1:01 am
Location: (somewhere), Afghanistan; from Rochester, NY
Contact:

Re: Another simple amp question

Post by dyarker » Sun Nov 19, 2017 4:51 pm

Why 660/330 Ohm input? Is guitar output impedence that low? (really is a question, because I don't know) That would mean input bias would be drawing more current than collector side.

How about 600 Ohm resistor to common for low input impedence, capacitor to base, and 4K/2K bias resistors.

Maybe a 50 Ohm emitter resistor to decrease the load on the bias resistors, and help stabilize circuit gain better than raw Hfe. (emitter/collector resistor ratio)

If you raise collector current from 100uA to almost 1mA the minimum Hfe of the 2N3904 goes from 40 to 70. We saved more current than that on input side.

This will all need calculator whipping with actual imedence and level out and actual input needs of the power amp. Still ... a couple of ideas I hope.

Cheers,
Dale Y

sghioto
Posts: 319
Joined: Wed Dec 26, 2007 5:16 pm
Location: N. Florida
Contact:

Re: Another simple amp question

Post by sghioto » Sun Nov 19, 2017 5:14 pm

A common emitter is not an amplifier. Why not just use a simple op amp circuit like a MCP6281?

Steve G
Attachments

[The extension jpg has been deactivated and can no longer be displayed.]


User avatar
CeaSaR
Posts: 1624
Joined: Sat Nov 08, 2003 1:01 am
Location: Phoenixville, PA USA
Contact:

Re: Another simple amp question

Post by CeaSaR » Sun Nov 19, 2017 7:35 pm

Could be that I am not looking at the correct topology. I want to use a single transistor because:
1. That is what I have on hand
2. It must run off of 5 volts
3. I don't want to have to buy another op-amp to work within these constraints as I have no way to etch a board and don't want to send out for one (p-p construction is fine)

I need to bring up, to a useable level, from as low as 20 mV up to about 200 mV, minimum. So a gain of 10 would be required.

So really, I am looking for is voltage gain. The inputs on the actual amp board are already capacitively coupled.

-CeaSaR
Attachments

[The extension jpg has been deactivated and can no longer be displayed.]

Hey, what do I know?

User avatar
CeaSaR
Posts: 1624
Joined: Sat Nov 08, 2003 1:01 am
Location: Phoenixville, PA USA
Contact:

Re: Another simple amp question

Post by CeaSaR » Sun Nov 19, 2017 9:53 pm

So, no clue as to why it didn't work so well last time, but here's another screenshot that should look much better.
-CeaSaR
Attachments

[The extension png has been deactivated and can no longer be displayed.]

Hey, what do I know?

sghioto
Posts: 319
Joined: Wed Dec 26, 2007 5:16 pm
Location: N. Florida
Contact:

Re: Another simple amp question

Post by sghioto » Sun Nov 19, 2017 10:05 pm

The problem with that circuit is the low input impedance. Guitar pickups are high impedance devices that need a high impedance input amp to work correctly which is why I suggested the op amp circuit. You don't have to etch a board, you can assemble the circuit on a small "perf" board.

Steve G

dyarker
Posts: 1640
Joined: Fri Aug 22, 2003 1:01 am
Location: (somewhere), Afghanistan; from Rochester, NY
Contact:

Re: Another simple amp question

Post by dyarker » Mon Nov 20, 2017 5:05 am

A common emitter is not an amplifier.
Yes, it is.

The Emitter follower (AKA common collector) does not amplify the voltage. It does amplify the current (lower output impedence), so could still be called an amplifier.

Since the guitar pickup is high imopedence (which I didn't know before) I like sghioto's idea of a low Vcc op amp. If you want to stay with discrete transistors put an FET in front of the 2N3904. Direct coupling of FET and base of 2N3904 saves adding bias resistors. A second 2N3904 wouldn't help much because gain decreases at low collector current.

Cheers,
Dale Y

sghioto
Posts: 319
Joined: Wed Dec 26, 2007 5:16 pm
Location: N. Florida
Contact:

Re: Another simple amp question

Post by sghioto » Mon Nov 20, 2017 5:09 am

Yes it is, I was thinking of the emitter follower when I made that statement so stand corrected.

Steve G

dyarker
Posts: 1640
Joined: Fri Aug 22, 2003 1:01 am
Location: (somewhere), Afghanistan; from Rochester, NY
Contact:

Re: Another simple amp question

Post by dyarker » Mon Nov 20, 2017 5:13 am

I suspected that. Which is why I added the next paragraph in my previous post.

No problem and cheers,
Dale Y

User avatar
CeaSaR
Posts: 1624
Joined: Sat Nov 08, 2003 1:01 am
Location: Phoenixville, PA USA
Contact:

Re: Another simple amp question

Post by CeaSaR » Mon Nov 20, 2017 8:56 am

But that doesn't fit the constraints I have.

So...

Here's another one I found and a Sim thereof.

Granted, Rs is arbitrary, but without knowing, this is a middle ground according to the text. Also not worried about super fidelity, since it's for guitar and I want distortion at harder hits and hotter pickups. Also already have level and extra fuzz ready if this works out. Should know Friday.

-CeaSaR

**BTW, the B/W below is from http://guitarscience.net/calcs/ce.htm
***The Sim from Falstad shows a 200mV +- P-P swing with a 20mV input
Attachments

[The extension png has been deactivated and can no longer be displayed.]

[The extension png has been deactivated and can no longer be displayed.]

Hey, what do I know?

User avatar
Joseph
Posts: 635
Joined: Mon Dec 10, 2001 1:01 am
Location: USA,World
Contact:

Re: Another simple amp question

Post by Joseph » Mon Nov 20, 2017 5:51 pm

I think adding another 2N3904 as an emmitter follower for the input would be good to try.

dyarker
Posts: 1640
Joined: Fri Aug 22, 2003 1:01 am
Location: (somewhere), Afghanistan; from Rochester, NY
Contact:

Re: Another simple amp question

Post by dyarker » Tue Nov 21, 2017 4:35 am

The Sim from Falstad shows a 200mV +- P-P swing with a 20mV input
Yes, because of the 10 to 1 ratio of the collector and emitter resistors. As long as the ratio is less than "raw" gain of the transistor at the bias point (collector current), the ratio controls the signal gain. The resistance at the base is the emitter resistor times Hfe. Then that in parallel with what ever bias resistors gives circuit input resistance.
Dale Y

User avatar
MrAl
Posts: 3854
Joined: Fri Jan 11, 2002 1:01 am
Location: NewJersey
Contact:

Re: Another simple amp question

Post by MrAl » Tue Nov 21, 2017 6:40 am

Hello there,

Any circuit you build for this that is linear will always waste a substantial part of the input power and thus will drain the battery quicker. Since it is all about current mostly with battery operated stuff, you could study the current drain itself and see if you can improve something that is already designed though.

For best efficiency and longest battery life you want to get away form the linear amplifiers and move to the Class D type amps. They use PWM just like a buck circuit and are made to be high efficiency. They basically came after the industry realized that a buck circuit can be very efficient, but audio amps up to that point were not, so using a fast buck with AC modulation allowed for a new class of amplifier that could be high efficiency too. It had to wait though for the cheaper MOSFETs to be most effective, but now there are chips that do this.

So the best bet for battery operated stuff is to go to the Class D amplifier by finding a chip that does that. The battery life could be much longer like 2 times.

The reason why the linears waste so much energy is because there is always a partially conducting transistor in series with the speaker. In Class D the transistors are either all the way on or all the way off so they dont dissipate a lot of energy. This makes a huge difference in run time for battery operated stuff.

Note the actual circuits are buck circuits, but driven typically with an H bridge. That means you get more power from a given battery voltage too. Since power is V^2/R, that could mean 4 times the power output from the same battery voltage.
LEDs vs Bulbs, LEDs are winning.

User avatar
CeaSaR
Posts: 1624
Joined: Sat Nov 08, 2003 1:01 am
Location: Phoenixville, PA USA
Contact:

Re: Another simple amp question

Post by CeaSaR » Tue Nov 21, 2017 7:50 pm

Well, the board with the actual power amp is set to use heaphone levels, presumably from a digital player such as an MP3 player or a cell phone's heaphone output, which, in this case are rather low on all accounts. Besides, just based on the resistance shown, it should only draw a few milliamps.

And yes. I know of the advantages of the "new classes" of amps (not really new, just much more refined and more widely accepted). But, like I said in my OP, it's a bit old-school, and since it is aimed at a guitar, analog is the preferred sound, at least for the person in question. Besides, this is a one, maybe two, - off project, just for the sheer joy of making and giving.

As to the actual power amp, here's a video from a guy I've followed for a long time on Youtbe:
https://youtu.be/L08pd5l1aT0

-CeaSaR
Hey, what do I know?

User avatar
Joseph
Posts: 635
Joined: Mon Dec 10, 2001 1:01 am
Location: USA,World
Contact:

Re: Another simple amp question

Post by Joseph » Thu Nov 23, 2017 5:25 pm

I refined my circuit today on LTspice. I added the emitter follower on the input like I was thinking earlier, and added another one to buffer the pull-up resistor.

I was thinking about lowing the value of the 47ohm resistor since I noticed little or no actual effect by including it in the circuit at all. I included it in the first place to try to raise the input impedance a bit more. Maybe lower is as much as down to 22ohm, or maybe try omitting it altogether.
Attachments

[The extension png has been deactivated and can no longer be displayed.]


Post Reply

Who is online

Users browsing this forum: CeaSaR and 2 guests