Carvin B1500 problem

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

CJ

Well-known member
GDIY Supporter
Joined
Jun 3, 2004
Messages
15,658
Location
California
this monster amp is causing headaches,

puts out 1900 watts into 2 ohms! 

all the output transistors and drivers have been pulled and checked, but when power is applied the transistors start to get hot in a hurry til the fuse blows,

so we are working backwards, no voltages on the schemo does not help,

the TIP31 (red arrow) was not stuffed into the board, and the resistor (circled in red) is 0 ohms (jumper)
this is apparently an option on the board,  we tried putting in a TIP31 and taking out the resistor but no joy,

my question is <what is the function of the TIP31 transistors and what kind of voltages should be present?>


thanks for any help!  the 340 and 350 surface mount transistors have been replaced already,
 

Attachments

  • Carvin B1500 pwr.jpg
    Carvin B1500 pwr.jpg
    178.1 KB · Views: 37
CJ said:
this monster amp is causing headaches,

puts out 1900 watts into 2 ohms! 

all the output transistors and drivers have been pulled and checked, but when power is applied the transistors start to get hot in a hurry til the fuse blows,
If transistors get hot, that sounds like too much class A bias.

the Tip31 actually the two Tip31s form a Vbe multiplier. So the bottom tip31 is diode connected so should measure 0.5-0.6V base-emitter.  The top tip 31 has it's emitter connected to the diode tip 31 so should measure 1V-1.2V from base to bottom of bottom tip 31 emitter.  The collector voltage of the the top tip31 will vary with trimpot p101... the ratio of top R to bottom R defines collector voltage, and class A bias current. Since that output stage has 4 forward junction drops nominal on voltage should be around 4x0.5 or 2V (base of Q107 to base of Q111.

You can measure the voltage drop across any one of the 0.33 Ohm enitter degeneration resistors to set class A current. About 8mV would be 25 mA. 
so we are working backwards, no voltages on the schemo does not help,

the TIP31 (red arrow) was not stuffed into the board, and the resistor (circled in red) is 0 ohms (jumper)
this is apparently an option on the board,  we tried putting in a TIP31 and taking out the resistor but no joy,


Shorting the bottom tip 31 is a crude way to under bias the amp to run cooler. Crossover distortion will suffer.
my question is what is the function of the TIP31 transistors and what kind of voltages should be present,
The bottom tip31 may be mounted to the heat sink to provide some thermal tracking to the output devices (when transistors heat up the Vbe drops, so constant bias voltage means more current when hot. )Thermal coupling from outputs to the Vbe multiplier could help keep the amp from thermal runaway.

Can you bring the amp up on a variac? look for normal drive voltage feeding the output stage drivers (2V +/-). 
thanks for any help!  the 340 and 350 surface mount transistors have been replaced already,

if you temporarily short between base Q107 and base Q111 does the output stage stay cold? If not you still have some other broken components that need replacement. 

JR
 
Looks like H9.1 and H9.2 are test points to check the output idle current in one output device

As posted by John  reads like the output transistors are set to too high a idle current

What does the wire that goes to the right from R251 connect to?  It is cut off
 
> 1900 watts into 2 ohms!

+/-90V supply. (180V total.) (Plus un-sag.)

As the MJL21194 is a 250V(!) part, this seems reasonable.

> bring the amp up on a variac

Yes, but I like resistors. A dead-short will go to infinite current at teeny voltage; a resistor sets some limit on current *and* power.

I used to use 10r 10W on baby 60W amps. Break the rail(s) from the amp, insert series resistor. You could get 30V, or 3 Amps, but only 15V 1.5A (22W) at once.

And the node which "dont make sense" at the reduced voltage is the Clue. Assuming you do get enough voltage for the circuit to bias-up kinda-nearly-normal.

As this amp is 4X lower load impedance we might be thinking 2 or 3 Ohms. But this leads to 2,700(!) Watts possible resistor dissipation. As your amp should work-right with 8 Ohm load (or higher!), let's keep to 10 Ohms. At 90V that could be 9 Amps 900 Watts. You *will* be watching the sagged rails as they do (or don't) come-up, so it won't be 900 Watts for long.

Especially since you work a shop, you should invest in an amp limiter. Not lamp-limiter (you should already have one!!), but big resistors to hobble big transistor amps' power supplies so they don't smoke too fast. Two 10r 400 Watts should cover about anything they drag in the door (if you act-fast when a big-amp is near-short). Dale (and the Chinese) make a "100 Watt Aluminum case" resistor. These have to be on a BIG chassis to dissipate 100 Watts (can you get old VW heads?).
 
PRR said:
> 1900 watts into 2 ohms!

+/-90V supply. (180V total.) (Plus un-sag.)

As the MJL21194 is a 250V(!) part, this seems reasonable.

> bring the amp up on a variac

Yes, but I like resistors. A dead-short will go to infinite current at teeny voltage; a resistor sets some limit on current *and* power.
we used to make "load boxes" for working on amps... Variac with a 100W incandescent light bulb in series. any problem and the lamp current limits to about 1A on the mains.  After the amp comes up Ok, the light bulb gets bypassed for full current testing. Not sure why we called it a load box, but that's the name other people gave them.

JR
I used to use 10r 10W on baby 60W amps. Break the rail(s) from the amp, insert series resistor. You could get 30V, or 3 Amps, but only 15V 1.5A (22W) at once.

And the node which "dont make sense" at the reduced voltage is the Clue. Assuming you do get enough voltage for the circuit to bias-up kinda-nearly-normal.

As this amp is 4X lower load impedance we might be thinking 2 or 3 Ohms. But this leads to 2,700(!) Watts possible resistor dissipation. As your amp should work-right with 8 Ohm load (or higher!), let's keep to 10 Ohms. At 90V that could be 9 Amps 900 Watts. You *will* be watching the sagged rails as they do (or don't) come-up, so it won't be 900 Watts for long.

Especially since you work a shop, you should invest in an amp limiter. Not lamp-limiter (you should already have one!!), but big resistors to hobble big transistor amps' power supplies so they don't smoke too fast. Two 10r 400 Watts should cover about anything they drag in the door (if you act-fast when a big-amp is near-short). Dale (and the Chinese) make a "100 Watt Aluminum case" resistor. These have to be on a BIG chassis to dissipate 100 Watts (can you get old VW heads?).
 
this thing still,

got it passing audio,  got some voltages,  do they look alright?

not much class A current, like zero, but it still works,

 

Attachments

  • 1500-trouble.jpg
    1500-trouble.jpg
    195.7 KB · Views: 15
Nothing noted looks like A Problem.

> not much class A current, like zero

No, but it will do. And it's a gol-durn stage amp, not for playing string quartets off virgin vinyl.
 
got rid of the clipping noise at higher volumes by changing a 680 ohm to 2 K,

now pwr transistors are biased up a bit, about  23 mv on emitters = about 70 ma per device,

gets hot without fan, heads towards runaway, almost impossible to work on without heat sink,

have to make a pwr jumper cord to fix outside the box,

here is resistor change, tried messin with the 150 ohm above the pot, and zero ohm jumper resistor but that did not do the trick
 

Attachments

  • carvin.png
    carvin.png
    205.6 KB · Views: 8
amp works! sounds awesome, sepecially the sub bass control, thanks for all the help,

just needed a resistor change, 680>2K, and all new transistors,  :eek:

here is the final voltage chart, class A current not as high as stated before, forgot to check DC offset, which is a bit high at 30 mv but just glad to get rid of this thing,

 

Attachments

  • 1500-trouble.jpg
    1500-trouble.jpg
    205.5 KB · Views: 5

Latest posts

Back
Top