Amp4 -HV build

[sleepyhead]

Greetings

Building for use with battery.

The good news: Built the amp and have music playing through it. I'm at 24v using a variable power supply with variable current limit set at 180mA.

The bad news: Slowly turned up the supply voltage. At 42v the amp died. Current limit pegged and volts went to near zero. Raised current limit but to no avail. Troubleshooted it for a while and found the problem goes away with the removal of the sta517b chip. Installed another and adjust the output offset voltages and the amp plays music again. Raise voltage and again the power stage chip gives up the ghost same as before. I remove it and the amp4 resumes normal current draw. But...I'm out of chips unless I put a 2050 in there. Both sta517b chips have numbers 991D3 937 on them.

Did I get bad chips or am I a bad chip installer? Modulator fb total resistance is 24.9k. Input gain is -2.5. Vpp sense resistors are 357k. Bulk caps not installed. C102,c103 not installed because I bought 63v 1500uF and they are too big. I couldn't wait to test amp so I proceeded without them. I do however have some 63v 470uF caps on order with proper lead spacing. (live and learn)

Thoughts? Suggestions?

[gnome]

I'm just curious why you went to 42V?
seeing that it is only rated to 32V max and the output is only 50W

[makruzz]

The HV version is rated to 52V

If i remember correctly there is a number of sta517b chips out there that are faulty and cant take the rated voltage

[gnome]

ah ok I didn't realise there was a HV version



although output is still only 50W?

[makruzz]

I belive its about 2*100W 4 Ohm at 42V, 42V is max with 4 Ohm speakers and 52 with 8 ohms

[sleepyhead]

sta517b datasheet. Read the 2nd paragraph under "description" on page one and "electrical characteristics" in table 3 for VCC on page 5.

Also see AMP4-HV, it's alive!!

I didn't choose 42 volts. The chips died at 42 volts (or thereabouts) as I was on my way towards 52 volts. I installed all 100v decoupling caps, other parts, etc as described by V-bro. Music was from mp3 player on battery. (no strange ground loops) Amp4 on heatsink.

Anywho... if I get this amp cranking out some killer wattage I'll share the results, post some pics, etc.

[V-bro]

It's roughly 2*160W guys, it's twice the current and almost twice the voltage of the 'normal' output stage chip...

BTW, what's the 5V reg fed by? Are you sure it's not just the regs that gave up? With the voltage sensing resistors I advised to use in that thread there should be overvoltage shut down with no damage...

[sleepyhead]

Yeah I'm not expecting real killer wattage. <sorry> Maybe about 76 "clean" watts per side because 8 ohms.

On the first try I left r6 out and fed the 317 with half the supply voltage. At 24v it got 12. At 40 it got 20 and so on. For the 2nd attempt I put r6 in and am using 317hvt. In both cases amp4 went into massive current draw that didn't go away when I removed jumper 5. It didn't go away when I cut power and turned it back on at 24v. Problem DID go away when I removed the power stage chip. Upon removing jumper 5 the 5v partly came back and most of the abnormal current draw remained. My thoughts were that not having c102,c103 present might be the cause but I really don't know.

I figured the current limiting feature of the power supply would save my butt. I dialed up 180mA before connecting Amp4. Would by chance, starving for current cause damage? I wouldn't think so but with electronics anything's possible. Also the way I know there's a problem is the voltage dropped to near zero. (you know, that's how the supply limits the current) I briefly raised the current to 250-300mA but the voltage stayed near zero. Usually when a circuit needs more current the volts drop a little, I raise the current and the V comes back when circuit is good. See difference? That's how I know this one is bad.

I changed both r52 and r54 to 357k. I figured that to be okay for the 52v. I forced the relay closed when amp was at 24v so I could hear the music. (playing very very low) r100 is 1.5k. Relay stays closed, well, until things go wrong.

[makruzz]

Hello I believe you have gotten some of the bad chips floating around out there, check This thread. You have the same serial as the fakes

Hope you can get it working soon

//Makruzz

[V-bro]

Current starvation should not have caused this, it is very likely you had a bad STA517 chip and even though the voltage sensing resistors limited it at a proper voltage the chip couldn't handle it....