Overload protection circuit. Does it work correctly?

[Fritzell]

Ok here's a circuit of an amplifier with a overload protection circuit by monitoring the Re of the output transistors.

http://sound.westhost.com/p27b-f2.gif

Im thinking of adding one of these to my amp but in simulation it shows strange results.

If i put a DC signal on the output and short-circuit the output, the current in the upper output transistors is limited, and the current through the diode and the current-limiting transistor is about 18mA (the VAS current). This is correct.

But when i simulate it with a negative voltage on the output, the curent IS limited in the output transistor BUT the current through the diode and protection-transistor inreases to 450mA

Is this really correct? Why can this happen? Is it because the current is amplified by the VAS? Shouldn't it have the same current as the positive transistor?

Thanks alot
//Fritzell

 

[darkfenriz]

I think it is correct.
It is because single ended VAS together with feedback.
Feedback is doing its best to force output to zero Volts.

 

[Fritzell]

Ah yes i think i understand. The feedback tells the LTP that the VAS needs some more base current to establish the voltage. But if the BC556 is going to sink 400mA it would be hot Maybe it's better to put a more powerful transistor there..

Thanks for your reply
other people please fill in if you wish!

 

[CBS240]

I have noticed that this type of DC protection circuit is used when using paralelled output transistors. My question is, is it prudent to use this circuit when using just a single pair of outputs?

 

[amplifierguru]

Hi Fritzell,

The Vas transistor needs to be clamped under overdrive - it needs a 39R in the emitter leg and the LTP 560R -> 820. Then put two series diodes pointing down across this 820R. Sim that and it should limit symmetrically.

Cheers,
greg

 

[powerbecker]

Ok here's a circuit of an amplifier with a overload protection circuit by monitoring the Re of the output transistors.

http://sound.westhost.com/p27b-f2.gif

Im thinking of adding one of these to my amp but in simulation it shows strange results.

If i put a DC signal on the output and short-circuit the output, the current in the upper output transistors is limited, and the current through the diode and the current-limiting transistor is about 18mA (the VAS current). This is correct.

But when i simulate it with a negative voltage on the output, the curent IS limited in the output transistor BUT the current through the diode and protection-transistor inreases to 450mA

Is this really correct? Why can this happen? Is it because the current is amplified by the VAS? Shouldn't it have the same current as the positive transistor?

Thanks alot
//Fritzell

It is not correct, current through the BD139 (and so also through the BC556) must be limited!
Put a small resistor between the emitter and -35V and two diodes in series over the 560E.
Even than the current through the BC556 is greater than in the
Bc546 but you don`t kill both!

 

[Fritzell]

Hi all thank you very much for your replies.

Im using a current-mirror in my simulation instead of the resistor. But adding a 39 ohm resistor on emitter of VAS didn't really make a difference. If i increase it to something like 390 ohm things start happening and current goes down to 70mA but i will loose too much output swing with this.

 

[Eva]

As other people has already mentioned, that circuit is not reliable in overcurrent conditions because the current limiting protection is incomplete.

In order to fix that I prefer to use a emitter resistor, and a current limiting transistor, connected in such a way that it shunts the base drive of the BD139 when the voltage drop across the emitter resistor reaches the Vbe thresold of the limiter transistor.

 

[darkfenriz]

Re: Re: Overload protection circuit. Does it work correctly?

It is not correct,...

Yes, the protection circut works incorrect. I meant simulator shows correct (life-like) situation.

 

[jan.didden]

Ok here's a circuit of an amplifier with a overload protection circuit by monitoring the Re of the output transistors.

http://sound.westhost.com/p27b-f2.gif

Im thinking of adding one of these to my amp but in simulation it shows strange results.

If i put a DC signal on the output and short-circuit the output, the current in the upper output transistors is limited, and the current through the diode and the current-limiting transistor is about 18mA (the VAS current). This is correct.

But when i simulate it with a negative voltage on the output, the curent IS limited in the output transistor BUT the current through the diode and protection-transistor inreases to 450mA

Is this really correct? Why can this happen? Is it because the current is amplified by the VAS? Shouldn't it have the same current as the positive transistor?

Thanks alot
//Fritzell

Fritz,

Sorry for coming in so late, but are you putting a DC on the output and then shorting it, or are you putting a DC on the input that gives the DC out and then shorting it in the sim?

Jan Didden

 

[Fritzell]

Hey all thanks for so many replies.

janneman: I've omitted the dc-blocking cap on the feedback and put a DC signal on the input to get the DC signal on the output. This is because then i can test positive and negative voltage overload protection separate and see how it behaves.

I agree this circuit is somewhat incomplete and doesn't really work properly

The problem seems to be that on positive signal, the current is limited by the current-source and there's to much resistance to take any current from the VAS, but on negative swing the output protection has direct contact with the VAS collector and can sink so much current as the LTP input pair can deliver out * hFE of vas = too much Right now i don't see how to work this out

 

[Eva]

Fritz:

Add another emitter resistor and current limiting transistor to the VAS, like the ones you are using for the output devices. Clipping behaviour will also get improved.

 

[jan.didden]

Fritzell,

Strange, you would think that whatever is above the Vas, the current limit xsistor should be able to short the drive to the driver. I think the 12k is too large to sustain the base current for full shorting. Keeping the same current limit, try to change the divider from 1k - 12k to say 100 ohm - 1.2k. That should do it.

Jan Didden

 

[Fritzell]

Hey thanks for all ideas. Here's the current circuit im been working with. I put a 22 ohm emitter resistor for vas and a transistor on vas base. When current exceeds something like 30mA voltage increases so it opens the transistor and leads away all base-current to the vas and this is how it works. In simulation this config seem to work very well.
I dont know how it will work in real life though. Will it cause oscillation or some other bad effect? Please comment this circuit

Im thinking of maybe adding an inductor on the output also..

Scheme:
http://upl.silentwhisper.net/uplfolders/upload8/ampschematic.GIF

Thanks!

 

[Eva]

I have never seen such a VAS current limiting arrangement oscillating in practice, but the classic output device current limiting arrangement may oscillate because it is closing a feedback loop across two more transistors. You should build a prototype and test it carefully.

 

[Fritzell]

Okay i will try. Thanks all for your answers!

 

[jan.didden]

Hey thanks for all ideas. Here's the current circuit im been working with. I put a 22 ohm emitter resistor for vas and a transistor on vas base. When current exceeds something like 30mA voltage increases so it opens the transistor and leads away all base-current to the vas and this is how it works. In simulation this config seem to work very well.
I dont know how it will work in real life though. Will it cause oscillation or some other bad effect? Please comment this circuit

Im thinking of maybe adding an inductor on the output also..

Scheme:
http://upl.silentwhisper.net/uplfolders/upload8/ampschematic.GIF

Thanks!

Hi,

I see you changed one of the resistors across Re for current limiting, now the limit is MUCH lower. You sure you want to do that?

Jan Didden