They are usually employed when VI limiting of the output stage is used as a protection measure, as the sudden cutting off of the output stage and inductive loads cause flyback spikes which have to be absorbed somehow. With the output stage being high impedance at this point the diodes return the current back to the power supply and clamp the voltage at one diode drop above the rails thus not exceeding the Vce rating of the transistors.
Leolabs said:
Left alone these diodes are useless. In order to comply with their claimed nominal function, they should be helped *at least* by another pair of reverse biased diodes which protect the base-emitter junctions of output devices.
However their protective function is just limited in the occurrence of hard-clipping on inductive load. If protection have to be assured also in occurrence of sudden speaker cable disconnection with high but not full drive, other circuits and remedies are to be implemented.
Hi
Piercarlo
it's also generally accepted design practice to put the clamping diodes across any BJT that might possibly drive an inductive load. they're not needed with most MOSFETs, since the diode is part of the transistor substrate already. with some designs where a driver section might be used with BJTs in one model and MOSFETs in another model with almost the same circuitry and board layout, the diodes might be included with the MOSFET amp, just because they're there. that way the production line doesn't need additional setup time.
Re: Re: Re: Output clamping diodes:When do we need them???
Is simple: imagine a power amp throttling about to half maximum output voltage and playng consistent bass content which usually require current beefing from output devices. In case of sudden disconnection of output cables, especially if long and connected to complex crossover, a voltage spike that may occur may be insufficient to activate the usual clamping C-E diodes ma enough to reverse breakdown the B-E junction of output devices and damage it (usually shorting it).
I've meet, many years ago, a mobile P.A. amplifier which mounted *in series*with the emitters of output devices (two old 2N3055) a 1N5401 power diode just for preventing this occurrence (worsened, in that circustamces, by the presence of an output transformer).
Hi
Piercarlo
PS - However this kind of fault is not "mandatory"
Many is related to the path available for discharging the spike's energy. A 100 Ohm resistor placed in parallel to the output often may be enough in preventing serious damage to the active devices. In the most of cases the accident don't cause any damage and the amplifier can continue his work blissfully... But in the case were is not too recommendable counting on luck (professional amplification at first) some kind of preventing measures must be taken in account.
AndrewT said:
Is simple: imagine a power amp throttling about to half maximum output voltage and playng consistent bass content which usually require current beefing from output devices. In case of sudden disconnection of output cables, especially if long and connected to complex crossover, a voltage spike that may occur may be insufficient to activate the usual clamping C-E diodes ma enough to reverse breakdown the B-E junction of output devices and damage it (usually shorting it).
I've meet, many years ago, a mobile P.A. amplifier which mounted *in series*with the emitters of output devices (two old 2N3055) a 1N5401 power diode just for preventing this occurrence (worsened, in that circustamces, by the presence of an output transformer).
Hi
Piercarlo
PS - However this kind of fault is not "mandatory"
Many is related to the path available for discharging the spike's energy. A 100 Ohm resistor placed in parallel to the output often may be enough in preventing serious damage to the active devices. In the most of cases the accident don't cause any damage and the amplifier can continue his work blissfully... But in the case were is not too recommendable counting on luck (professional amplification at first) some kind of preventing measures must be taken in account.Re: Re: Re: Re: Output clamping diodes:When do we need them???
I remember reading somewhere that even though a transistor would survive reverse biasing the junction each such occurence will weaken it's structure and contribute to addition of noise. The process is so slow that it's difficult to perceive. So, damage may have happened after all. After several such occurences the transistor will eventually die. Better be safe than sorry.
Piercarlo said:
I remember reading somewhere that even though a transistor would survive reverse biasing the junction each such occurence will weaken it's structure and contribute to addition of noise. The process is so slow that it's difficult to perceive. So, damage may have happened after all. After several such occurences the transistor will eventually die. Better be safe than sorry.
Re: Re: Re: Re: Re: Re: Output clamping diodes:When do we need them???
Because the voltage spike occurs on the inductance, when it is disconnected , and by definition, at that time, it is.... not anymore connected to the amplifier, which is therefore isolated from any consequences.
And if the spike is high enough to generate a spark that closes the circuit again, the current finds the way it followed before the break, and the overvoltage is quenched automatically.
LV
Piercarlo said:
Because the voltage spike occurs on the inductance, when it is disconnected , and by definition, at that time, it is.... not anymore connected to the amplifier, which is therefore isolated from any consequences.
And if the spike is high enough to generate a spark that closes the circuit again, the current finds the way it followed before the break, and the overvoltage is quenched automatically.
LV
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.