60 watt troubles

[rtill]

recently built figure 1a on this page...
http://sound.westhost.com/project03.htm

My power transistors blew because the heatsink wasn't large enough. This also blew the bd140 and bd139 transistors. Before it blew it worked great for a few hours. I replaced all of the transistors and am now using a larger heatsink that barely gets warm now - yeehaw!. The problem is now it powers the speakers but the sound is fizzy and cuts out. What other components could have "blown" in the amp?

[Leolabs]

output resistors???output snubber???Check them first.

[TheMG]

Hopefully this is not the case, but it is possible for the speakers to have been damaged if the amp does not have DC protection when it failed.

[rtill]

changed out the resistors....worked great for about 10 minutes.....then a burning smell arose....and it started doing the fizzy thing again.

I wonder if the resistance is too little??

[jaycee]

Sounds more like it could be oscillating... do the transistors get very hot with no load?

[richie00boy]

Maybe VAS transistor BD139 is faulty.

[Andy L. Francis]

First, check all BJT's! If they're fine then turn on your amp, BUT WITH A 60W MAINS BULB IN SERIES WITH THE PRIMER COIL OF SUPPLY TRANSFORMER. The bulb will protect your amp module in case of oscillation or short circuit.

What PCB layout are you using? With bad layout a CFP amp likes oscillating.
What's your idle voltage?
Have you got the 10Ohm/100nF Zobel on the output?
Any rail bypass? (100nF/100uF)

[rtill]

Originally posted by Andy L. Francis
First, check all BJT's! If they're fine then turn on your amp, BUT WITH A 60W MAINS BULB IN SERIES WITH THE PRIMER COIL OF SUPPLY TRANSFORMER. The bulb will protect your amp module in case of oscillation or short circuit.

What PCB layout are you using? With bad layout a CFP amp likes oscillating.
What's your idle voltage?
Have you got the 10Ohm/100nF Zobel on the output?
Any rail bypass? (100nF/100uF)

layout is exactly how the schematic looks. Using perf board.
using a 25-0-25, 2 amp hammond transformer....getting 34.4v+/- on the rails.
yes, have the cap and resistor on the output
100 uf rail bypass

changed the resistors again.....worked for 15 minutes.....then fizzy again. I just realized that I'm using .47 Ohm 5 watt resistors rather than a .5 Ohm....could this be the problem?

[EWorkshop1708]

The 0.5 is hardly different from 0.47. That's not the problem.

What may be the problem..........
The "Fizzy" sound you are referring to sounds like crossover distortion.

With the amp on, test to see if you have about 1.3V-1.4V between the base leads of the drivers. If you get 0 volts, it means your VBE transistor is dead and you have no bias.

An amp will play music with no bias, but it gets distorted because the amp plays in pure Class B operation.

[fotios]

Originally posted by rtill
recently built figure 1a on this page...
http://sound.westhost.com/project03.htm

My power transistors blew because the heatsink wasn't large enough. This also blew the bd140 and bd139 transistors. Before it blew it worked great for a few hours. I replaced all of the transistors and am now using a larger heatsink that barely gets warm now - yeehaw!. The problem is now it powers the speakers but the sound is fizzy and cuts out. What other components could have "blown" in the amp?

RTILL

I have only one but meaningfull remark (the same in another thread - which i don't remember - before few days for the same circuit you are using). The CFP output produce much more heat in comparisson with the EF output. Thus you need accordingly bigger heatsinks. Also in supply levels above +/-35Vdc this kind of output (although it has the great beneffit of the better bandwidth in comparisson with EF) it is some delicate. And this is obvious because the emitters (which are more negligible from the colectors) are connected in the supply rails. By some way in clasic electronics theory, the direction of this connection it is unorthodox.

Fotios