help with noisy power amplifier

[panfilero]

Hello,

This looks like a cool site, this is my first post, I found this site after struggling with making a little solid state amplifier for my guitar.

I wanted to make a quick amplifier so I went to the store and bought an amplifier kit: Velleman's K4003 kit which is based on a TDA2612 datasheet is here

http://pdf1.alldatasheet.com/datasheet-pdf/view/19262/PHILIPS/TDA2616.html

anyway, I hook up an mp3 player to this and no problem, sounds fine.... I hook up my guitar, and as soon as I turn up my volume knob.... noise! I can hear the guitar sound in all the noise, but the noise is terrible.... it's the same noise I hear if I leave nothing plugged up to the input of this amp...

so, I'm thinking:  My guitar is high impedance and this IC is made to use with low impedance signals... I tried putting an opamp buffer circuit between my guitar and amplifier to bring down the impedance, but that didn't work, I tried a non-inverting amplifier stage to boost my signal before the amplifier, that didn't work either....

any thoughts?

much thanks!

[tonyharker]

I have some questions for you.

1. Why did you choose a stereo Amplifier?
2. Have you speakers on both channels?
3. Is the guitar only connected to one channel or both?

Regards  Tony

[panfilero]

1. I just picked whatever kit they had at my local electronics store and it happened to be that one
2. I have two 4 ohm speakers, one hooked up on each channel
3. the guitar is tied in to both input channels

[Roly]

Welcome.

panfilero> it's the same noise I hear if I leave nothing plugged up to the input of this amp

Which suggests that you have an open ground, say possibly in your guitar lead.

Your guitar lead is screened lead I suppose?

[panfilero]

I've plugged in different guitars and basses, they all sounds the same, when I crank up the volume, it gets noisy... I don't know what a screened lead is.  I was thinking it was too high of an impedance for the amp that is causing the problem... I dunno I'm stumped

[DrGonz78]

Just to note about amps input impedance and the guitars output impedance. Your guitar's output impedance might be somewhere between... lets say... 9-17k ohms. In the manual of the amp that you have built it states the input impedance is 150k ohms. So, even a Fender G15 has a 1M ohms as its input impedance. So your amp is low comparably to many amps and not higher impedance. (Note: I hope that can help us get away from that being the cause of the hum.) I hope my information is accurate too.  :duh

One thing that I would want to know is the amp able to be hooked up to only one channel? I would want to know if the problems is still occurring if only channel of the amp is connected at a time.

Edit: One thing to note about that manual is that it is called "Input Sensitivity" and I took that as being input impedance. Please forgive my ignorance if that is not accurate, thanks.

[Roly]

Looking at the assembly instructions this little amp has quite respectable specs, but while the (maximum) sensitivity of 300mV should be more than enough for guitar, the input impedance of only 150k is quite low; however this should only result in a "thin" or "dull" guitar sound, not noise.  Don't worry about the impedance question at the moment, just concentrate on the noise problem first, then we can look at getting the amp input impedance up (your non-inverting op-amp preamp is the right idea BTW).

Could you be a bit more descriptive of what the "noise" sounds like please?  Hum?  Hiss?  A whine or particular note?  Can you hear local AM radio stations?

A screened lead is one that has an inner active conductor and a woven braid around the outside for ground.  All commercial guitar and mike leads are (or damn well should be) of this sort.  The question only arises because speaker leads often have the same connectors and often aren't shielded, just plain twin core, and these will pick up a heap of "noises off" and for that reason should never be used on the input side.

I notice that this kit doesn't come with an input connector, so the question arises, have you got the tip connection on the socket connected to signal in, and the sleeve connection to the associated ground point on the amp module?

Dumb question - if you are only using one channel of the stereo amp, are you sure you are feeding the guitar signal to the same side you have the speaker connected to?  Also, if you are only using one side at the moment it would be a good idea to short the input of the other (unused) side to its associated ground point with a short wire link. {sorry, I note you are using both channels in parallel - I have several irons in the fire ATM, and sometimes (often?) lose track)}

I'm still inclined to think that this is some sort of grounding problem on the input side, the ground isn't getting carried through to your instruments for some reason, or perhaps the signal and ground are crossed over.

HTH

[DrGonz78]

1. I just picked whatever kit they had at my local electronics store and it happened to be that one
2. I have two 4 ohm speakers, one hooked up on each channel
3. the guitar is tied in to both input channels

I think there is some kind of ground loop here between the two channels. Try one speaker and one channel at a time to make sure that it works. Which is where I think Roly is headed too. At least this needs to be tested channel 1, and one speaker. Good luck! :tu:

Late Edit: After re-reading this all over again I have to admit that I am probably wrong about ground loop between two channels so ignore that comment if you haven't already...  :lmao:

Two things strange about this noise thing... which we need more info about...
1.) "It's the same noise I hear if I leave nothing plugged up to the input of this amp"
2.) "I hook up an mp3 player to this and no problem, sounds fine"

Just had to add more to this post and to exclude my lost thought about ground loop between channels.

[Roly]

{see, I think DrGonz is wondering about the same thing I am; without anything connected to the input of an amp like this they are generally quite quiet, but if you connect a long unshielded wire (aka "antenna"  ) they will pick up all sorts of crud from house wiring, local radio stations, &c, and this is also what happens with a shielded lead, e.g. guitar lead, if the ground happens to go open circuit to the screen.

And no, I don't think this is a ground loop as such ... um ... you are putting the guitar signal in directly from the guitar, and not via a stomp box?  If not please go direct to start with until we get the noise sorted.}

[Kaz Kylheku]

This could also be a case of poor common mode rejection in the amplifier.  Even if a guitar is properly wired, grounded and shielded, with humbucking pickups, it's still a source of common mode noise: crap that appears on both the ground and signal wires in equal voltages.

If it's happening with numerous guitars that behave properly in other rigs, it points to the amplifier.

[tonyharker]

How have you connected your guitar to the amplifier - to which points?  Have you used a stereo jack plug or a mono one?

[J M Fahey]

Please hand draw a simple schematic showing what you built, if no scanner available (they are becoming rare these days) you make take a picture and post it.
Also a picture of the amp itself.
Maybe we see something we don't notice in a verbal description.

[panfilero]

So I got a TL072 OpAmp, and have it set up non-inverting with a 47uF bypass cap at the input going to ground.  This cap gets rid of most of the noise when the input is left unplugged or at high impedance (like my guitar).  This works ok when plugged into an mp3 player as my input, but I still get nothing from my guitar...  and to add my issues there's 8V DC sitting at the output of my output cap C4... apparently capacitors don't block DC at my house....

[Roly]

First up, go back and read the responses above, answer the questions, and post a pic as requested.


Secondly, in your preamp circuit you have shown a 47uF cap, C2, across your signal input.  It isn't surprising that this would shut up your noise, the only thing that is surprising is that any signal at all gets past this point as this cap forms an effective AC short circuit for any signal.  Replace C2 with a 1Meg resistor.

Thirdly, unless there is a DC path on the output side of your C4 to allow it to charge, then you will certainly measure some voltage there.  Try placing a resistor of around 10k-47k, not critical, from the output side to ground,and you should see the voltage at the output settle to zero.

[tonyharker]

The + input to the TL072 is going to need a reference to earth, you should connect the 1M resistor between this point and earth and scrap the 47uF Cap.