chipamp.com kit problem.

[sberube]

Hello everyone,

I recently purchased a LM3886 kit from chipamp.com and I'm having a weird issue I hope someone could help me with. I've read through the documentation for the kit completion, searched many forums, but alas, my electronics skills are not all that great (I've been building speakers for a long while, but this is my first attempt at an amp).

My inquiry is triggered by building the whole kit and measuring the output voltage of both amps and getting 2.2 vdc when turning the power on. It reads in the documentation that it should measure <100mv.

Now I was weary of maybe having a resistor that went bad (would be odd since both channels produce a 2.2vdc read out), so I tested every resistor on both amplifier PCBs, and they read proper values (odd thing is I tested them while soldered on board, and they all gave good values - I was told afterwards I should have removed them from the board because I might end up getting skewed values). Then I went back to the power supply to measure output values and noticed the following:

the V- to PGND- produces 30vdc (out of 22VAC input)
but the V+ to PGND+ produces 1.2vdc.

Going back to the documentation, it reads to check the voltage between the V+ and PGND+ but never mentions how much it should read ?!?.

Also, between my V+ and V-, I get 30 VDC.

Does this make any sort of sens to anybody here? I'm sorry for what is most likely a very crude and improperly articulated question, but as I mentioned earlier, my skills are quite basic when it comes to electronics and I've been trying to look for answers for 3 days now, but I feel my lack of proper electronics vocabulary (and understanding) is inhibiting me to find the answer on my own.

Thanks.

[guitvinny]

Hi,

It looks like you got some issues with your power supply : V+ to ground should read the opposite of V- to ground. In your case you should have around +30V and -30V. Between V+ and V- your should read around 60V. I suppose that both of your amplifiers are supplied by the same power supply board. To test if it's the power supply : Unplug both amplifiers board and measure the voltage of both rail (V+ to ground and V- to ground). You should see a bit more than 30V, since you got no load. Just be sure to have a correct power supply before re-connecting the amplifiers.

[mullardel34]

It sounds like you need to diagnose an issue in your power-supply. Does your kit allow you to separate the power-supply (power transformer, rectifiers, power-supply filter capacitors) from the power-amp board? The site, http://www.chipamp.com./, isn't currently responding, so I can't consult the schematics associated with your kit...

If you can isolate the power-supply, check the AC-volts output of the power-transformer output taps. You want to confirm that you're actually seeing the specified AC-voltage at the output taps of the transformer. For example, if you've got a transformer specified as providing 22VAC, measure the AC-voltage presented at both of the pairs of transformer output secondary taps (see the attached sample power-supply schematic; measure across the pair of "AC1" transformer output taps, then measure across the pair of "AC2" transformer output taps).

If this measurement shows that both halves of the transformer outputs are presenting equal AC-voltage levels, then you need to measure the rectified DC voltage presented across the power-supply capacitors ("+V" and "+PGND" for the positive supply, "-V" and "-PGND" for the negative supply). Again, assuming a 22VAC secondary voltage, then each power-supply capacitor bank should be presenting approximately 30VDC (22VAC x 1.414 to convert the RMS AC-voltage to the resulting peak DC voltage (minus about 1.4VDC dropped across the rectifier diodes). All indications are that you'll likely find a fault somewhere in the power-supply portion of the chip-amp kit, but keep us posted...

[sberube]

Thanks for the quick answer! I've measured both my secondaries, and they both give me 22 VAC - which is good. But when it comes to PGND+ and V+ the rectified current is not good.

Mullard, your diagram is quite similar to what I've implemented.

I guess my next step will be to test the components on the side of the power supply that's not providing 30vdc of output.

Any suggestion as to which component to test first? The Resistors are quite easy to test, but can anyone point me to a link where I can find easy to read documentation on testing diodes and capacitors?

Thanks for your help. I'll update the thread when I find something as it may be usefull for someone else.

[mullardel34]

When performing rudimentary measurements of diodes (unpowered, using the ohm-meter function of your multimeter), you should see a significantly lower impedance when the "-" or "COM" lead is attached to the anode end of the diode (on the schematic, that's the end of the diode pointed-to by the "arrow" within the diode's schematic symbol). When you invert the multimeter lead connections, the ohm-meter function should show a sizeable jump in the impedance reading. Confirm that the alignment of your diodes (as indicated by your multimeter readings) matches the orientation of the diodes when compared to the schematic and/or the PC-board markings.

If you locate an obvious inconsistency in diode orientation, then just correct the improperly installed diodes. If not, you may need to unsolder the power-supply capacitors in order to isolate the fault.

[john blackburn]

Hi sberube.

It would be worth posting a photo or two if you can.

My first amp was a 3875 kit from Chipamp.com about 4 years ago. It has played at least 14 hours a day, virtually every day since then.

My guess would be a soldering issue, give the boards a good clean with isopropyl and a toothbrush and inspect every joint with a good magnifying glass. Check for whiskers of solder causing a short and that no joints look dry.

Did you shorten your transformer secondaries before soldering them to the board? They have a clear insulating lacquer on all but the very end of the lead. I tested my first transformer from the box and it was fine. I tried it again later in the build and had zero volts on both secondaries. I thought I must have burned it somehow but couldn't understand how. All I had done was trim to length and twist the secondaries, with no power applied obviously.

Once I was told about the invisible insulation I scraped it off with a knife blade, 22 volts were restored and we were back on track. If it's not cleaned off you will never get a good solder joint.

Also read forum member AndrewT on the subject of "lightbulb tester" Use it every time you power a new or altered thing up.

John

[sberube]

That's a great idea. Here are a few pics of my PS. It is devoid of its diodes as I removed them all last night to test them. Most of them didn't seem to work (either that or I really can't test a diode for s***). The ones that didn't seem to work seemed to be opened.

I've re-tested my secondaries and they do provide the proper voltage (I measured using the screws on the backside of the PCB that pinches the secondaries into place)

I've ordered another bunch of diodes, and we'll see what happens when I hook it all up together. I'll hook up ONLY the power supply and the transformer and capture the voltages between the PGNDs and the Vs to make sure all works. I'll let you know how things turn out!

Thanks for all your help... and more to follow...

[mullardel34]

Sebastien,

Glad to hear that the power-transformer secondaries are still providing their specified voltage levels. Before installing the replacement rectifier diodes, be certain to check the power-supply capacitors in order to ascertain if any of them are shorted and confirm proper orientation of the power-supply capacitors ("+" and "-" terminals soldered to the appropriate PC-board locations).

Good luck in bringing the power-amp to life! ;-)

[sberube]

I HAVE checked the orientation and they are good. But is there an easy way to test the capacitors? they are 50v 10,000uF. I've read through some website and what I could find is basically that I should start by discharing the caps (ie. shorting them). I'm sure I can do that Then, with my multimeter on Ohm, I should connect the leads of my multimeter while respecting polarity and I should have a 0 reading that will grow closer and closer to infinite as the cap is being charged by the battery of my multimeter.

Now should I remove the cap from the circuit first? I have a feeling that if I'm going to be shorting a cap, I should be unsoldering it from the pcb first.

Thanks,

[john blackburn]

Hi sberube

I take it you know about the white lines on the pcb indicating the tab side for the diodes?

Measure your AC voltages on the pcb rather than on the terminal screws, I am a bit unsure about the pinched on connectors. I think it would be a more "solid" setup with the secondaries soldered directly to the board.

John