Building a discrete multi-stage guitar amp

tracecom

Joined Apr 16, 2010
3,944
The TDA2005 is made by an Italian company. Their English datasheet is missing a lot of information. Maybe the details are on the Italian datasheet but I have never seen it and I do not know Italian.
Thanks

This is the latest English datasheet (2012) I have found; is it the one you have?


I am currently using 6 AA batteries to achieve the 9V. I find that I can buy 6 AA batteries for $3.79 as compared to 1 9V battery for $3.15. If it turns out that AA batteries don't last long enough, I can go to C or D cells. And if 9V is insufficient, I can go to 8 or 10 cells.
 

Attachments

Last edited:

Audioguru

Joined Dec 20, 2007
11,249
The TDA2005 datasheet shows a graph of stereo (not bridged) output power vs supply voltage but when overdriven so the clipping distortion is a horrible 10%.

With an 8V supply the output per channel is 1.6W into 4 ohms with 10% distortion which will be about 1.2W at clipping. Bridged into 8 ohms might be about 2.4W at clipping.

An AA alkaline cell battery has a capacity of about 2500mAh but that is when the voltage has dropped and the maximum power of the amplifier will be fairly low. Figure on 830mAh.

A bridged TDA2005 has an idle current of about 100mA which is 0.8W which is much higher than the average output power plus heating when playing music, not playing continuous tones at full blast.
So the Alkaline AA cells battery should last about 8 hours when playing loudly with the peaks at clipping.
 

tracecom

Joined Apr 16, 2010
3,944
The TDA2005 looks good to me; it will provide significantly more power than an LM386. There are lots of youtube videos showing homebuilt amps that use it. I will order some today.
 

MrChips

Joined Oct 2, 2009
23,946
I have no objections to a bridged amplifier. You try out the TDA2005 while I do the TDA2040 which I already have and we can compare notes. I have to start working on the cabinet.
 

THE_RB

Joined Feb 11, 2008
5,438
Interesting choices!

I thought the fact the class-D amp would give many times longer battery life would have been a significant factor in a portable amp.
 

tracecom

Joined Apr 16, 2010
3,944
I thought the fact the class-D amp would give many times longer battery life would have been a significant factor in a portable amp.
I confess to not knowing or thinking about that. Of course, I have no formal education in electronics; what little I know, I have learned through trial and ... well, you know.
 

Thread Starter

blah2222

Joined May 3, 2010
582
Okay so this site was mentioned earlier, but I am just curious why certain values for resistances and capacitors were chosen for the following JFET audio preamp:



I understand these capacitors are used for coupling, but I am trying to understand why their values were chosen.

What is the reason for the input 0.1uF capacitor value choice? Is this because it sets up a high-pass filter 3dB @ 1.6 Hz for audio frequencies with the 1M resistor???

What is the reason for a 4.7uF capacitor value choice to couple the signal from JFET1 to JFET2? High-pass filter 3dB @ 0.03 Hz with 1M resistor???

What is the reason for a 10uF capacitor value choice for the output coupling?

What is the reason for the 100uF capacitor value choice with a 10ohm resistor on the supply?

Thank you,
JP
 

tracecom

Joined Apr 16, 2010
3,944
Interesting choices!

I thought the fact the class-D amp would give many times longer battery life would have been a significant factor in a portable amp.
In further searching, I did not find a class D amp chip that is both through hole and single voltage powered. Do you have a recommendation? Thanks.
 

Audioguru

Joined Dec 20, 2007
11,249
What is the reason for the input 0.1uF capacitor value choice? Is this because it sets up a high-pass filter 3dB @ 1.6 Hz for audio frequencies with the 1M resistor???
The first Jfet is the preamp for an electric guitar magnetic pickup. The input impedance of the circuit is very high to allow the innductance of the pickup to resonate at about 4kHz with the cable capacitance to add some "twang" to the sound.
Since the pickup conducts to 0V then the 0.1uF input coupling capacitor is not needed.

What is the reason for a 4.7uF capacitor value choice to couple the signal from JFET1 to JFET2? High-pass filter 3dB @ 0.03 Hz with 1M resistor???
Stupid isn't it? A very high input impedance is not needed for the 2nd Jfet so it should be an ordinary transistor. The capacitor value should be calculated for a cutoff at about 30Hz.

What is the reason for a 10uF capacitor value choice for the output coupling?
Another stupid wrong value. It should be about 0.47uF for a 30Hz cutoff frequency.

What is the reason for the 100uF capacitor value choice with a 10ohm resistor on the supply?
It keeps the tiny battery voltage from jumping up and down with the signal and reduces noise from the 12V from causing interference with the signal.
 

Attachments

Thread Starter

blah2222

Joined May 3, 2010
582
I am confused why you are saying not to have an input coupling capacitor. In the event of a DC input bias, this would affect the operation of the JFET, no?

How are you deciding on specific frequency cutoffs for filtering capacitors, like 30 Hz?
 

Thread Starter

blah2222

Joined May 3, 2010
582
Ever listen to a 20Hz signal? You feel it more than hear it. Almost no instrument produces it. It takes enormous amounts of power to reproduce it.
Silly question, but physically why does it require more power for low frequencies to sound?

Does it have to do with a longer wavelength needing to move more air to make a sound?
 

Audioguru

Joined Dec 20, 2007
11,249
The magnetic pickup of an electric guitar is a coil of wire to ground with no DC voltage so the Jfet does not need an input coupling capacitor.

An electric guitar does not produce deep low bass frequencies so a cutoff of 30Hz is fine.
 

wayneh

Joined Sep 9, 2010
17,153
Silly question, but physically why does it require more power for low frequencies to sound?
MrChips correctly suggested that it doesn't. (I think that's what he meant.)

But human hearing requires a much higher SPL for a low frequency to be perceived as the same loudness. Generating that higher SPL takes more power from the amplifier.

BTW, this is one reason that low frequencies are more dangerous to our hearing - they carry a lot more energy at the same perceived volume.



from here
 

Attachments

Last edited:

Thread Starter

blah2222

Joined May 3, 2010
582
Okay so I wired up this circuit (using an MPF102 instead of a J201) to my Gibson SG and it looked promising.



Gain was around -1 V/V.

Voltage measurements kinda perplexed me though because it would seem that I am operating in the triode region and not saturation, but this could be that I am assuming values for pinch-off voltage...

Vds = 0.60 V
Vgs = -2.05 V
Vdd = 9.75 V
Id = 1.05 mA
Vp = -3 V (assumed)

For saturation: Vds >= Vgs - Vp >= 0

0.60 >= -2.05 - (-3) >= 0
0.60 >= 0.95 >= 0 (WRONG!)

Do these operating voltages look correct and what values am I supposed to assume for Vp and Idss? Idss is listed on the datasheet but has a wide range (2 - 20 mA...)

Thanks
 

Audioguru

Joined Dec 20, 2007
11,249
Okay so I wired up this circuit (using an MPF102 instead of a J201)
It has some very wide ranges of spec's so some will work and others will not.
The IDSS of the original J201 Jfet is 0.2mA to 1.0mA. Your replacement is 2.0mA to 20.0mA.
The cutoff voltage for the original is -0.3V to -1.5V. Your replacement has no minimum voltage and has a max of -8.0V.

Your replacement Jfet is barely working since it is conducting too much.
A 2N5484 matches the original Jfet better but some will still be wrong.
 

Thread Starter

blah2222

Joined May 3, 2010
582
It has some very wide ranges of spec's so some will work and others will not.
The IDSS of the original J201 Jfet is 0.2mA to 1.0mA. Your replacement is 2.0mA to 20.0mA.
The cutoff voltage for the original is -0.3V to -1.5V. Your replacement has no minimum voltage and has a max of -8.0V.

Your replacement Jfet is barely working since it is conducting too much.
A 2N5484 matches the original Jfet better but some will still be wrong.
Okay, I'm going to play around with resistor values and biasing to see if I can get my Vds a bit higher.
 

THE_RB

Joined Feb 11, 2008
5,438
The magnetic pickup of an electric guitar is a coil of wire to ground with no DC voltage so the Jfet does not need an input coupling capacitor.
...
The first part of your sentence is correct, but are you sure about the second part?

The guitar pickup waveform is AC +/- in reference to ground. If we are talking about the same JFET circuit it is a single supply circuit which requires its input to always be above ground? Or have I missed something?
 
Top