• Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Solder iron not hot enough for PC board

audioguru

Well-Known Member
Most Helpful Member
#22
The Chinese amplifier or your observations have many errors:

1) The 10k input volume control feeding the 4.7k resistor and parallel with the 50k input resistance of the amplifier produces a total input resistance of 3k which is too low for anything. They probably added the unneeded 4.7k resistor to convert a wrong linear variable resistor into a logarithmic volume control.
2) They used a 100pF input capacitance that is completely wrong since a simple calculation shows that it cuts all audio frequencies.
3) The 100pF capacitor at the output pin 5 should be 0.05uF (500 times more!).
4) The 100uF output capacitor value is too low for a speaker but is fine for attenuated earphones.

I would replace the 10k variable resistor and remove the 4.7k input resistor then replace them with a 100k log volume control. Then a simple calculation shows that an input capacitor of 150nF (0.15uF) would pass most audio frequencies. I would also replace the 100pF capacitor at pin 5 with 47nF (0.047uF).

The output power is too high for earphones and will damage them and your hearing. If they are 32 ohms each then connect them in parallel and add a 47 ohm resistor in series with them as an attenuator.

Your calculation of 4.7k for the LED results in a very dim light, I changed it.

A linear variable resistor is marked "B" and an audio logarithmic volume control is marked "A".
 

Attachments

Nigel Goodwin

Super Moderator
Most Helpful Member
#23
1) The 10k input volume control feeding the 4.7k resistor and parallel with the 50k input resistance of the amplifier produces a total input resistance of 3k which is too low for anything. They probably added the unneeded 4.7k resistor to convert a wrong linear variable resistor into a logarithmic volume control.
2) They used a 100pF input capacitance that is completely wrong since a simple calculation shows that it cuts all audio frequencies.
3) The 100pF capacitor at the output pin 5 should be 0.05uF (500 times more!).
Perhaps you should try reading previous posts?, I already explained all those points in the post above yours, and that Gary has put the wrong values on the diagram. The zobel cap should also be 0.1uF, in my opinion - but isn't in any way critical.

As for the speaker coupling capacitor, I knew you'd comment on that - as you always do - and as always you're 'wrong' to some extent, the amplifier is intended for feeding tiny little speakers, which as they have little bass response doesn't need a large coupling capacitor, and will use less power because of the smaller one.

As for input impedance, it's obviously perfectly fine for feeding from an opamp or similar, but I would imagine it's intended for feeding from a headphone socket (hence the 3.5mm input socket), so is perfectly suited.
 

audioguru

Well-Known Member
Most Helpful Member
#24
Sorry Nigel.
I started my reply then ate lunch after your reply that I did not see.

I never use lousy little speakers because they are squeakers. My cheap clock radio has an amplifier similar to a 0.45W LM386, I replaced its coupling capacitors and now it produces sounds down to 40Hz a ported 6.5" woofer and dome tweeter speaker.

The input resistor values are probably for a low impedance dynamic mic instead of a much higher impedance electret mic.
 

Nigel Goodwin

Super Moderator
Most Helpful Member
#25
The input resistor values are probably for a low impedance dynamic mic instead of a much higher impedance electret mic.
Not for a mike at all I wouldn't have thought? - an LM386 is a (much misused) small power amplifier, you really need a preamp to feed a mike to it.

In fact AG (as I'm sure you know) your electret mike preamp in the sticky would be perfect for the purpose.
 

gary350

Well-Known Member
Thread starter #29
Sound is not good. I did not put the 47 ohm resistor in the output there is no easy way to solder it in place. I have 9vdc inverter power supplies but they all have the wrong plug. I have a 12vdc .75ma that I am using. I did not change 10k volume control because I don't have a replacement. With volume turns all the way up I have to talk directly into the electric mic to get sound from the ear phones. Sound is good quality and loud. Mic is not sensitive enough to pick up sound from the TV or me talking 1 inch from the mic. I double checked mic is connected the correct way. 10K resistor is soldered between +9v and the .1uf cap & + side of mic. The .1uf capacitor is ceramic not electrolytic I wonder if that makes a difference. The cap in series with 10 ohm resistor is ceramic too? Sound at ear phones is same volume as me talking.

 
Last edited:

Reloadron

Well-Known Member
Most Helpful Member
#30
So we are done with:
Solder iron not hot enough for PC board
We have now moved along to LM386 Power Amplifier circuits? Just as a channel check.

Ron
 

audioguru

Well-Known Member
Most Helpful Member
#31
Maybe your mic and/or LM386 and/or capacitors is/are Chinese so they are garbage?
Maybe you do not have an electret mic (you said electric mic)? Maybe you connected the electret mic backwards?

A 0.1uF capacitor feeding the 10k volume control parallel to the 50k input of the LM386 cuts low audio frequencies below 193Hz so it cuts a man's voice level but passes voices from squeaky little kids.

Also the low value of the 10k variable resistor "loads down" and reduces the output from the fairly high impedance electret mic. Try the amplifier with the 10k variable resistor removed.
 

gary350

Well-Known Member
Thread starter #32
The electret mics work good I have them on TL071 amp it picks up every sound in the room 25 feet away..

I replaced the volume control .1uf ceramic caps with .1uf electrolytic caps amp is much louder but sound quality it terrible and mic picks up better but still not good. Volume is too loud for ear phones it hurts my ears it needs a speaker. Mic is not sensitive will not pick up any sound more than 6" away.

I replaced the .1uf ceramic cap in series with 10 ohm resistor with .1uf electrolytic.

I removed 4.7K resistor between pin 3 and 4 no change to sound quality or volume.
 

audioguru

Well-Known Member
Most Helpful Member
#36
The electret mics work good I have them on TL071 amp it picks up every sound in the room 25 feet away.
Then please post the schematic of the TL071 preamp so we can see why is sensitivity is much more than your LM386.

I replaced the volume control .1uf ceramic caps with .1uf electrolytic caps amp is much louder but sound quality it terrible.
Please describe terrible sound. Distortion? No bass? Boomy bass? No highs? Shrieking highs? Acoustical feedback howling? Hiss noises?
Please describe the volume control resistance and is it "A" logarithmic for audio or "B" linear for a voltage setting.?

Volume is too loud for ear phones it hurts my ears it needs a speaker.
The loudness in earphones depends on the impedance of them. Are they 32 ohms for each ear?

I replaced the .1uf ceramic cap in series with 10 ohm resistor with .1uf electrolytic.
The capacitor at pin 5 should be ceramic which works well at the high frequency where the LM386 tries to oscillate without the capacitor. An electrolytic capacitor works poorly at high frequencies.

I removed 4.7K resistor between pin 3 and 4 no change to sound quality or volume.
The 4.7k resistor should not be there when a proper volume control is used. I calculated a 0.15uF input capacitor when the 4.7k resistor is removed and the volume control is 100k.
 

gary350

Well-Known Member
Thread starter #37
I built this 4 or 5 years about I am pretty sure this is the circuit.

Sound is garbled hard to under stand the words. I removed the ear phones and attached a small speaker it sounds really bad. Too much power for earphones above 70% volume. Sound was a lot better with he ceramic capacitors but not very loud for ear phones at full volume. I will need to put ceramic cap on pin 5 again.

I soldered 4.7K resistor on bottom side of PC board for quick test. It sounds no different and volume is no different with or without the 4.7K resistor.

 

audioguru

Well-Known Member
Most Helpful Member
#38
I suspect that ebay sold you a Chinese defective or fake LM386. Maybe your capacitors are also Chinese and are defective. All my audio parts are bought from Digikey or Newark and are genuine. They all sound excellent and clear.

Your LM386 circuit has a gain of 200 times. Your TL071 circuit has a maximum gain of 1+ (220k/4.7k)= 47.8 times which makes it a little less sensitive than your LM386 circuit, but its load on the mic is only 50k while the load on the mic with the LM386 is a heavy load reducing its sensitivity.

Yes, the metal case of the electret mic connects to ground.
 

gary350

Well-Known Member
Thread starter #39
I tried an experiment & swapped the chinese LM386 m-53 with a 40 year old Radio Shack LM386 n-1 there is no difference in the sound. All circuits I find including data sheet says 10K variable resistor.

I know the LM386 n-1 from radio shack work excellent in the circuit I built 40 years ago. This circuit might not be correct and maybe it could be made better but it works amazing well. The big 8 inch diameter speaker from a 50 year old TV acts like a dish if I aim it a boat on the lake that is so far away it is just a spot I can hear the people talking like they are standing right next to me. I have not tested this circuit in years the electrolytic capacitors are probably bad by now and I'm not sure I still have a speaker for the input. 10:30pm at night I can rotate the 8" speaker around soon I pick up people talking I have no clue where they are or how far away they are but they sound like they are setting right next to me. This old thing has 40 years of dirt and dust on it. It might be fun to clean it up and get it working again it has always needed to be built into a portable speaker box with battery inside. NEW 9v battery works good for 20 minutes then it won't work it needs another NEW battery. I should probably use 12v or 12v with 9v regulator.



 
Last edited:

Latest threads

EE World Online Articles

Loading

 
Top