Hey diyaudio pro builders, masters and wise ones, the worst amp builder in the whole world requires the assistance once again. So the thing is I got a scooter, and thought I would add some cool mods to it and the first thing that came to my mind is an audio system. So i scoured the internet and found a nice amp schematic (uses TDA2005), made it, and ofcourse it doesn't work. There isn't even any "power on plop", it makes absolutely no sound what so ever. My multimeter(which i think is broken) tells that there is no voltage going out of the amp (those connectors that connect to speaker). I've attached the schematic to the attachments part. I think that there are no cold joints. The only thing that i had to do is bend some chips legs to fit the holes and use a (i think it was R3) 2.4 kOhm resistor instead of 2 kOhm one, because my local electronics store doesn't have one. Here's some pictures of the amp.
[image]http://i56.tinypic.com/1roxtk.jpg[/image]
[image]http://i52.tinypic.com/t99rhs.jpg[/image]
[image]http://i55.tinypic.com/nclhtt.jpg[/image]
[image]http://i55.tinypic.com/2lxxwch.jpg[/image]
I hope these help.
PS. Please don't use any of the "Hard To Understand" electrical terms, and technical jargon.
[image]http://i56.tinypic.com/1roxtk.jpg[/image]
[image]http://i52.tinypic.com/t99rhs.jpg[/image]
[image]http://i55.tinypic.com/nclhtt.jpg[/image]
[image]http://i55.tinypic.com/2lxxwch.jpg[/image]
I hope these help.
PS. Please don't use any of the "Hard To Understand" electrical terms, and technical jargon.
OK - the fact that your meter dosent show anything on the output dosent mean that the thing is blown up, I am concerned that you say that the meter is maby broken. Lets start with the model and brand on the meter, and mabe a picture of it. You can find the problem fast with a meter, but perhaps not at all without it. We can also help with some simple tests to tell if the meter is broken.
It's a Mastech M-833, I'm pretty sure it's dead, i tried mesuring resistance and it showed random numbers. The probes are plugged in correctly and i wasn't touching the resistor leads or probes.
How about volts, just measure the power source, or a battery. Also, it has a battery in it, so you should check that next. If its internal battery is low it will cause ohms not to read correctly.
You were right, battery was dead, it could barely power the built in speaker for testing continuity. Measuread the amp and it showed < 1 volt.
You have shown us a schematic but what is the power supply to it, exactly (in circuitry details if possible)?
The schematic for your project is incomplete. What I mean is, it shows only an amp subcircuit which assumes (near enough) perfectly clean power up to that point in the circuit. It only has a power rail decoupling capacitor of 0.1uF (= 100nF). On the same PCB as the amp chip you should have far more than 0.1uF. The datasheet for TDA2005 suggests 100uF for "Figure 23 : Bridge Amplifier without Boostrap" (they need to proofread more!) which seems reasonable if your power supply itself has at least 1000uF capacitance in it (even more is better).
Your C4 polarity is reversed! Your PDF shows the wrong capacitor orientation on the component outline picture, look at the etching picture and the schematic, negative polarity should connect to ground and positive to pin #3 of the amp chip.
Also, you will get better performance if you increase the capacitance of C4 higher than 10uF (100uF or a little more) and also if you were using a lower ESR capacitor. If you want to stick with 10uF, try using a tantalum capacitor for C4 but MAKE SURE the polarity is right as they can pop quickly if polarity is reversed. Would a reversed C4 cause your specific problem? I do not know, but doubt that it would work properly with it reversed so I suggest correcting that polarity problem as the first step then retesting.
In this picture, http://i55.tinypic.com/nclhtt.jpg of the back of the amp chip, I see a couple little whiskers or hairs or something around the chip pins. If they are metal they could potentially short chip pins. Take a toothbrush and clean those out of the chip pin area. Otherwise (from what I can see of it in the picture), the pins look ok, I do not see a problem resulting from bending them. The picture of the bottom of the PCB also shows some kind of whisker-like fragment on the lower left side between one of the traces from a chip pin and the wide left-most trace (probably ground) which should be cleared off.
Otherwise, is this lead containing or lead free solder you used? Good leaded solder joints usually look shinier. It might just be an optical illusion from the picture but the solder joints on the lower row of chip pins look a bit granular and questionable. Ideally you would apply flux only to these joints and reheat them (no more solder is needed, just to rewet and flow the solder that is there)... but as mentioned above I would reverse polarity on the capacitor and retest before doing anything else.
As for your multimeter, maybe it just needs a fuse replaced?
The schematic for your project is incomplete. What I mean is, it shows only an amp subcircuit which assumes (near enough) perfectly clean power up to that point in the circuit. It only has a power rail decoupling capacitor of 0.1uF (= 100nF). On the same PCB as the amp chip you should have far more than 0.1uF. The datasheet for TDA2005 suggests 100uF for "Figure 23 : Bridge Amplifier without Boostrap" (they need to proofread more!) which seems reasonable if your power supply itself has at least 1000uF capacitance in it (even more is better).
Your C4 polarity is reversed! Your PDF shows the wrong capacitor orientation on the component outline picture, look at the etching picture and the schematic, negative polarity should connect to ground and positive to pin #3 of the amp chip.
Also, you will get better performance if you increase the capacitance of C4 higher than 10uF (100uF or a little more) and also if you were using a lower ESR capacitor. If you want to stick with 10uF, try using a tantalum capacitor for C4 but MAKE SURE the polarity is right as they can pop quickly if polarity is reversed. Would a reversed C4 cause your specific problem? I do not know, but doubt that it would work properly with it reversed so I suggest correcting that polarity problem as the first step then retesting.
In this picture, http://i55.tinypic.com/nclhtt.jpg of the back of the amp chip, I see a couple little whiskers or hairs or something around the chip pins. If they are metal they could potentially short chip pins. Take a toothbrush and clean those out of the chip pin area. Otherwise (from what I can see of it in the picture), the pins look ok, I do not see a problem resulting from bending them. The picture of the bottom of the PCB also shows some kind of whisker-like fragment on the lower left side between one of the traces from a chip pin and the wide left-most trace (probably ground) which should be cleared off.
Otherwise, is this lead containing or lead free solder you used? Good leaded solder joints usually look shinier. It might just be an optical illusion from the picture but the solder joints on the lower row of chip pins look a bit granular and questionable. Ideally you would apply flux only to these joints and reheat them (no more solder is needed, just to rewet and flow the solder that is there)... but as mentioned above I would reverse polarity on the capacitor and retest before doing anything else.
As for your multimeter, maybe it just needs a fuse replaced?
Last edited:
!!!!!Dude!!!!! Somehow you made a mirror image of the tracks on your board. It will not work unless you put the chip on the back.
^ I agree the mirror image of the traces is confusing but it seems to have been pre-mirrored in the PDF picture so the amp chip does appear oriented correctly as soldered in on the PCB (as far as I can see looking at the camera picture).
For power I use a 12v battery, that's all. Also, those whisker-hair things are my cat's, she likes to mess around in on my table. I noticed that it was flipped while soldering it, but i just flipped the image and soldered according to the flipped version of the image. Also I have one 2200 uF capacitor laying around, can I use that instead of 1000uf capacitor?
I used solder with lead, and it's not that shiny, because it's been laying in my drawer for some weeks now, I think it's because of that.
I used solder with lead, and it's not that shiny, because it's been laying in my drawer for some weeks now, I think it's because of that.
You should put the highest uF capacitor that will physically fit on the PCB itself. Yes the 2200uF would be even better than 1000uF.
Leaded solder lying in a drawer for weeks will not affect how shiny the joint is. It "might" just be the exposure of your camera but the joints don't look ideal. Does it mean they are problem? No way to know until the amp is proven to work properly and stays that way (temperature cycling can make poor solder joints fail).
Last edited:
Ok, tommorow i will try to flip the chip and the C4 capacitor and use the 2200uF one, I will post the results.
For now leave the resistor. How do you charge the battery, and measure the actual voltage on the battery for us. Im glad others are chiming in. It makes it more fun.🙂
With the nice cuts around that backwards board I think you should submit you pics to some of the other treads arould here. LOL
You "really" need a heatsink. Without one, you will cook the chip amp. No BS. At the least find a piece of metal to screw the chip down to.
When i get home from school i will measure the battery, i charge it using a 0.5 A charger, the battery is 7.2 Ah. Those "nice" cuts are from a very dull hacksaw, i couldn't find a new one at home. I know about the heatsink, i will put it on on later.
I followed pin 1 and it looks to be connected to pin 2 through an electrolytic cap.
2k2 ohms is a standard resistor value that shouldn't be too hard to find. From what I see, R3 helps set the voltage gain. A higher value equals higher gain. 2k4 ohms should be OK; it appears a lower value than 2k would be a concern as the closed-loop gain must be >32dB.
Those are pretty dull solder joints. You say you used leaded solder; was it rosin core? I would get some flux and "pretty them up" a little. When you reorient the TDA2005.
2k2 ohms is a standard resistor value that shouldn't be too hard to find. From what I see, R3 helps set the voltage gain. A higher value equals higher gain. 2k4 ohms should be OK; it appears a lower value than 2k would be a concern as the closed-loop gain must be >32dB.
Those are pretty dull solder joints. You say you used leaded solder; was it rosin core? I would get some flux and "pretty them up" a little. When you reorient the TDA2005.
The orientation of the TDA is correct.
The solder joints are definitely a mess. It looks as if you used plumbers solder without flux. Add flux and once you are at it, tin the traces to avoid corrosion. The copper is already going green in many places.
The solder joints are definitely a mess. It looks as if you used plumbers solder without flux. Add flux and once you are at it, tin the traces to avoid corrosion. The copper is already going green in many places.
- Status
- Not open for further replies.