Mini amp circuit

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: 1760
Joined: Sat Nov 08, 2003 1:01 am
Location: Phoenixville, PA USA
Contact:

Re: Mini amp circuit

Post by CeaSaR » Wed Sep 09, 2009 10:16 am

Thanks Robert,

I switched C5 to 0.0001 uF and R2 to 3.3 k ohms and re-ran the sims at 10 mV:

Code: Select all

Hz      output (mW at the speaker)
50      441.6
500     518.0
1k      524.8
2k      530.8
5k      533.4
10k     533.0
16k     529.4
20k     526.7
The addition of a 20K potentiometer (R9) in line between the source and C1 allows
a range of 10 mV to 200 mV at the input . Not too terrible after all, and the Sine
waves look really nice. A nice "little" amp to play around with, at least on paper.
Now to see if I have all the components...

BTW, here's the updated schematic:

Schematic #2 - latest schematic on page 6, all the way at the bottom.
Image

CeaSaR
Hey, what do I know?

User avatar
CeaSaR
Posts: 1760
Joined: Sat Nov 08, 2003 1:01 am
Location: Phoenixville, PA USA
Contact:

Re: Mini amp circuit

Post by CeaSaR » Wed Sep 09, 2009 10:30 am

Bob,

I have no idea where he (the "hacker") got/came up with that schematic. I just wanted
something to base it off of and my copy of CircuitMaker doesn't have other preferred
transistors like the BC series et al., so I needed something with real generics - like the
2n3904/06's. I tried to mod the circuit by adding the 2n4401's in the Quasi-Complementary
configuration because I have some of those and so does CircuitMaker. As you can see,
although I have been around electronics all my life, I am really a neophyte.

What would you do to this design, with the caveat of staying with these transistors and
trying to use standard values for the other passives (standard as in what you might get
in a value pack at a place like Radio Shack - oops, The Shack...).

Thanks,
CeaSaR
Hey, what do I know?

User avatar
sofaspud
Posts: 531
Joined: Wed Feb 13, 2002 1:01 am
Location: San Antonio, TX
Contact:

Re: Mini amp circuit

Post by sofaspud » Wed Sep 09, 2009 12:32 pm

Why the 2200uF loudspeaker coupling capacitor? In my mind, that part would be better utilized
across the power supply + and -, and substitute something like a 470uF 25V capacitor at the
loudspeaker. I don't think that will change the sim'd output levels significantly, if at all. At mW
output audio levels, single-digit frequency response is irrelevant except for the most unusual
apps. fc = 1/2*pi*R*C

User avatar
CeaSaR
Posts: 1760
Joined: Sat Nov 08, 2003 1:01 am
Location: Phoenixville, PA USA
Contact:

Re: Mini amp circuit

Post by CeaSaR » Wed Sep 09, 2009 2:08 pm

2200 uF vs 470 uF as the output capacitor:

Code: Select all

Output (mW at the speaker)
Hz     2200 uF     470 uF
-------------------------
50      441.6      271.4
500     518.0      513.2
1k      524.8      523.8
2k      530.8      529.0
5k      533.4      530.3
10k     533.0      532.4
16k     529.4      529.0
20k     526.7      526.4
All sims run with R9 removed - input connected straight to C1 - and the input set to 10 mV.
As you can see, the bass frequencies are affected by the size of the output cap. Originally,
2200 uF was selected because it shaped the output best, lesser caps showed a flattened
waveform (positive 1/2). I kept it through the rest of the sims because it worked so well
early on in the "meddling around" with this circuit.

That's it so far. Any more thoughts, questions, quips...?

CeaSaR
Hey, what do I know?

User avatar
CeaSaR
Posts: 1760
Joined: Sat Nov 08, 2003 1:01 am
Location: Phoenixville, PA USA
Contact:

Re: Mini amp circuit

Post by CeaSaR » Wed Sep 09, 2009 7:19 pm

Robert,

You had asked about R8 and R10. I re-ran the sims without them, just having
Q3 connected to Q5's base and Q4 connected to Q6's base. There was a 0.2 mW
drop in the output at 1 kHz. The drop is -2.7 mW if I left them in but reduced the
values by "/10" to 5k ohms. Are they necessary? Not really. Maybe I'll play around
with it later.

CeaSaR
Hey, what do I know?

Robert Reed
Posts: 2276
Joined: Wed Nov 24, 2004 1:01 am
Location: ASHTABULA,OHIO
Contact:

Re: Mini amp circuit

Post by Robert Reed » Wed Sep 09, 2009 8:11 pm

Ceasar

Actually R8 & R10 do very little in this circuit although they may have some effect on Class B/AB operation at idle affecting crossover distortion. Q3 & Q4 will supply about the same bias current to the output stages with or without them (acting as variable resistors), but while you are "simming", you may want to look at those effects along with dynamic range vs. those values.

Also I notice you connect R9 in series rather than parrelled to ground as would be with a traditional volume control. This will limit range of attenuation as you have already found out. Placing it to ground will give you infinate range for any input level, however the control would be severely loaded by Q1's input impedance. A simple follower transistor inserted between the volume control and Q1's input would make a much better match. Maybe a 1 K emitter resistor on the follower biased at 1 VDC would do the trick and only add about 1 ma more load on the power supply. A 1K series resistor -follower out put to Q1 input- may be needed to preserve feedback from R1, however R1's value is so high that there probably isn't a whole lot of feedback going on there. Remember that the feed back path is composed of a divider network of R1 and whatever value the input impedance is at its lower junction point. Oh well, just some ideas to play around with on your simulator.

User avatar
CeaSaR
Posts: 1760
Joined: Sat Nov 08, 2003 1:01 am
Location: Phoenixville, PA USA
Contact:

Re: Mini amp circuit

Post by CeaSaR » Wed Sep 09, 2009 8:46 pm

As I stated in this thread and elsewhere around the board, I use CircuitMaker.
The version I have is Student Edition and it was free when I picked it up oh so
long ago. Unfortunately, the company that developed it was sold and it is no
longer available for download. Since it was freeware, if anyone would like a copy
just PM me and you can have a copy (3.24 MB).

As for how to set up the program to check for dynamic range or any other type
of sweep, I am not quite sure how to make it do that.

I just want something to putz with. Maybe feed from my MP3 player or maybe
plug in a guitar for travel purposes - something that wasn't so generic as a
single black square of plastic that does everything. That'll be later. :grin:

CeaSaR
Hey, what do I know?

User avatar
sofaspud
Posts: 531
Joined: Wed Feb 13, 2002 1:01 am
Location: San Antonio, TX
Contact:

Re: Mini amp circuit

Post by sofaspud » Thu Sep 10, 2009 7:49 am

I have to confess that my judgment was a bit skewed, because when I deal with mW audio amps it's
almost always in a headphone-load context. I am just a little bit surprised that the 50Hz response was
reduced by roughly 40%. But my point is that at 1/2 watt there is a lack of power to drive a speaker
at usable 50Hz levels, practically speaking. And that it seemed like a big honker cap to afix as an
output coupler in a mini amp.

User avatar
CeaSaR
Posts: 1760
Joined: Sat Nov 08, 2003 1:01 am
Location: Phoenixville, PA USA
Contact:

Re: Mini amp circuit

Post by CeaSaR » Thu Sep 10, 2009 2:16 pm

Robert,

I don't know what I was thinking when I put R9 in that way. I must have been falling asleep
or something. Connected as you explained I get an input range from 10 mV up to 1 V for full
power.

Sofaspud,

As much as the calcs say one thing, the real world is usually different. The really simple amp
that I used to start this thread had some amazing low end, even though it simmed at ~5 mW!
Of course there isn't any output cap on that design...

Almost all of the higher end "designers" of discrete amps say to put as large as possible output
cap before the speaker out to reduce the amount of distortion and give a flat bass response
down to 10-20 HZ. I saw one where the original design called for 1000 uF and the reviewer /
redesigner said to put in 2200 uF or better. As proof he cited another guy's findings about electrolytics
and distortion in audio amps. It's somewhere out there and I don't have time to look for it now.

Anywho (yes, who), I will try this approximate schematic out, hopefully this weekend.

Thanks,
CeaSaR
Hey, what do I know?

User avatar
sofaspud
Posts: 531
Joined: Wed Feb 13, 2002 1:01 am
Location: San Antonio, TX
Contact:

Re: Mini amp circuit

Post by sofaspud » Fri Sep 11, 2009 6:54 am

I wouldn't take issue with any of that, and certainly don't want to tread any "wire with gain" waters.
I was thinking more of the overall system as small mp3 portable amplifier. The sim results show what
is across the loudspeaker. Versus what is actually audible from the loudspeaker.

User avatar
Bob Scott
Posts: 1192
Joined: Wed Nov 20, 2002 1:01 am
Location: Vancouver, BC
Contact:

Re: Mini amp circuit

Post by Bob Scott » Fri Sep 11, 2009 7:14 am

CeaSaR wrote:Almost all of the higher end "designers" of discreet amps say to put as large as possible output cap before the speaker out to reduce the amount of distortion and give a flat bass response down to 10-20 HZ. I saw one where the original design called for 1000 uF and the reviewer / redesigner said to put in 2200 uF or better.
Funny you should mention.......A few weeks ago I was rebulbing a Sony STR110 reciever and noticed that it had 470uF DC blocking caps in series with the output. I was saying to myself "I really should change those to 2,200uF."

Just use that equation Xc = 1 / (2 * Pi * f * C) and you'll see that the Xc (reactance, impedance) of a 1,000 uF cap is ~8 Ohms at 20 Hz. So you have an 8 ohm capacitor in series with your 8 ohm speaker down at 20 Hz. At this point on the frequency graph, where the cap impedance is equal to the speaker resistance, the bass will be down 3dB.
As proof he cited another guy's findings about electrolytics and distortion in audio amps. It's somewhere out there and I don't have time to look for it now.
The distortion of electrolytics really does not relate directly to the bass dropoff. Maybe just distortion at that point in the reponse curve. The distortion would be caused by electrolytics not being perfectly linear. They have some kind of minor hysteresis loop like inductors that have a iron core. Hifi fans and high-end speaker manufacturers use air cored inductors and film capacitors in the crossover networks in order to avoid distortion. I wouldn't worry about electrolytic capacitor caused distortion in your amp. There are a LOT of amps out there that sound just fine with them. Plus, my little Wharfedale speakers have iron powder core inductors and non-polarized electrolytic caps for their factory made crossovers. They sound awesome.

Bob
-=VA7KOR=- My solar system includes Pluto.

User avatar
CeaSaR
Posts: 1760
Joined: Sat Nov 08, 2003 1:01 am
Location: Phoenixville, PA USA
Contact:

Re: Mini amp circuit

Post by CeaSaR » Fri Sep 11, 2009 2:30 pm

Sofaspud,

Please don't worry about the "wire with gain" crowd here. I've seen quite a bit of the snake oil
that the super hi-fi elitists have tried to espouse - 00 gauge super oxygen free wire recommended
for runs of 20’ or less, $1000 IEC power cords (6’ no less!), cable elevators to reduce stray
capacitance in your speaker runs, etc. As long as it costs more, it’ll make it sound better/more pure.
You are probably close enough with the 470uF output cap that it really doesn’t make THAT much
of a difference, especially at those levels (as you said).

Bob,

What the guy was saying about the electrolytics is that there is a (paraphrase) “sweet spot” in the
value lineup where this so-called distortion is lowest. I think it was in the 2200 – 6800 uF range.
I’ll have to find that link for you again.

What improvements would you suggest for my last schematic? I’m open to quite a bit, but my
current stock of actives is limited to what I show. That and my situation does not allow for much,
if any, discretionary spending.

Here’s a link to another possible low power amp:
http://www.techlib.com/electronics/audi ... transistor
What do you think of this one? I'd love to hear your thoughts - all who have contributed to my threads,
as I highly value the combined knowledge that has been shared.

Thanks,
CeaSaR
Hey, what do I know?

User avatar
sofaspud
Posts: 531
Joined: Wed Feb 13, 2002 1:01 am
Location: San Antonio, TX
Contact:

Re: Mini amp circuit

Post by sofaspud » Fri Sep 11, 2009 5:26 pm

Took me a while to find again, but have you seen this page?
Some Useful Audio Circuits
He also shows some discrete low-power amps with fairly decent specs.
That capacitor explanation seems a bit familiar to me... can't seem to recall who what or where though.
Among my favorite audio websites is Elliott Sound Products from "down under." We share a no-nonsense
perspective. His cap info page is here.

User avatar
Bob Scott
Posts: 1192
Joined: Wed Nov 20, 2002 1:01 am
Location: Vancouver, BC
Contact:

Re: Mini amp circuit

Post by Bob Scott » Fri Sep 11, 2009 11:26 pm

sofaspud wrote:That capacitor explanation seems a bit familiar to me... can't seem to recall who what or where though. Among my favorite audio websites is Elliott Sound Products from "down under." We share a no-nonsense perspective. His cap info page is here.
Sofaspud,

I agree with the author of the above link when he talks about the nonsense that goes into audiophile snobbery. Just search PartsExpress.com for AC outlets to find the $120 wall receptacle. :mrgreen: However I disagree strongly with his low opinion of ceramic capacitors. He doesn't indicate that he even knows that there is more than one flavor of ceramic caps. There is the excellent and stable COG(NPO) type, the vary-with-temperature X7R, and the reaallly bad (use only for power supply bypass)Z5U. I would use the COG ceramic cap in any critical precision application. I actually prefer them over plastic film. They are solid. The don't age. Film caps do change value over time.

I do agree strongly when he points out that hi-fi audio is first processed heavily with opamps and non-magic long cables at the studio and record company. A few years ago, I heard of a development effort to build an RIAA preamp with no phase shift, to make it sound better. However, they forgot that the point of an RIAA preamp is to decode the RIAA encoding that the record company has added at the record factory, including the phase shifts. The preamp needs to undo the phase shift encoded at the factory, and any garden variety magnetic cartridge phono preamp already does that.

I'm not a complete atheist when it comes to the golden eared types. I realize that the human ear can hear a much larger dynamic range of levels than the human eye can see. Video frame stores only need 8 bits. Digital audio needs at least 20 because you can hear sounds that are 120dB down from full volume. That is 1 part in 1,000,000.

Bob
-=VA7KOR=- My solar system includes Pluto.

User avatar
Bob Scott
Posts: 1192
Joined: Wed Nov 20, 2002 1:01 am
Location: Vancouver, BC
Contact:

Re: Mini amp circuit

Post by Bob Scott » Fri Sep 11, 2009 11:41 pm

CeaSaR wrote:Here’s a link to another possible low power amp:
http://www.techlib.com/electronics/audi ... transistor
What do you think of this one? I'd love to hear your thoughts - all who have contributed to my threads, as I highly value the combined knowledge that has been shared.
That one is similar to the previous, but better. Go ahead and breadboard it. I already know how to add a couple of improvments, like a current mirror, more gain and negative feedback. It won't be for a couple of days though. I have stuff to do like replace the flexible front-end brake lines on the wife's car, the rubber coated ones that connect to the front calipers. I have to do it now, before the monsoons come for the winter. Non-stop rain is expected to start any day now and not stop until next spring. That, and the dishwasher is leaking.

Would you like a really short course in the basics of transistor amplifier design? They really are simple devices. I went YEARS trying to figure them out by myself because nobody told me the key properties and how simple they are to apply, not the transistor manual nor the electronics magazines. Looking at those rudimentary amateur designs, they don't know much either.

Bob
-=VA7KOR=- My solar system includes Pluto.

Post Reply

Who is online

Users browsing this forum: Google [Bot] and 44 guests