Clamp Diodes

Status
Not open for further replies.
Jump to Latest
east electronics

east electronics

Member
Joined 2006
eastelectronics.gr
#1
a few days we had a very creative post about zobel networks and some many nice arguments were placed there .....

lets talk about output clamp diodes ......

it is a form of protection

against short circuit ????
against reactive loads ????
why yes ???? or why no ????

and finally why big comercial amps are always equiped with this diodes and almost none of the amplifiers shown in the forum hasn't any of these diodes ????

except a protection do they also have some effect in sound ????

regards sakis
 
#2
If the output stage has some form of current limiting, then the fast switch off of output current creates a back EMF from an inductive load and this back EMF can overvoltage the output transistors.
The clamp diodes limit the back EMF voltage to supply voltage + 0.7V and save the output devices.
The distortion created by triggering the protection is still present.
Prior to triggering the amp should sound the same with/without the clamp diodes.
An additional pair of diodes (making 4 in all) from supply rails to ground are a further useful protection to help prevent reverse voltage building up on electrolytic capacitors during power down.
 
#3
sakis said:
a few days we had a very creative post about zobel networks and some many nice arguments were placed there .....

lets talk about output clamp diodes ......

against reactive loads ????
Click to expand...
Yes

sakis said:
...and finally why big comercial amps are always equiped with this diodes and almost none of the amplifiers shown in the forum hasn't any of these diodes ????
Click to expand...
Lack of design experience plus the amps work in light duty probably.
 
#6
true..

AndrewT said:
the ESP site has a couple of articles about this.
Click to expand...


i will take look ...thanks to rod there is a lot of nice things arround there

i was actually looking at P68 that looks like one of the designs i like very much ( i have to listen though ) but eventhough is a design that is a high power thing ....there is a double zobel but no clamp diodes ......i wonder why
 
#7
Re: true..

sakis said:
looking at P68 that looks like one of the designs i like very much ( i have to listen though ) but even though is a design that is a high power thing ....there is a double Zobel but no clamp diodes ......i wonder why
Click to expand...
does the p68 have limiting?
If not, then clamping diodes are not mandatory.
But those 4 clamping diodes are nice to have, cheap to fit and gives peace of mind with no deleterious sound effects to worry about.
 
#8
sakis said:
and finally why big comercial amps are always equiped with this diodes and almost none of the amplifiers shown in the forum hasn't any of these diodes ????
Click to expand...

It could be, as was mentioned, lack of design experience or the idea about this protection method even existing. I also have couple of alternative theories: In some cases those diodes are already integrated inside the output devices, e.g. such as in TIP142/TIP147 darlingtons. It has been said that trusting that these internal diodes exist or are even robust enough is bad practice, though. Some people simply do not care to make full-detailed schematics and trust that builders are wise enough to include all the neccessary protective or stabilizing elements to the design by themselves.
 
#9
Re: Re: Clamp Diodes

teemuk said:


It could be, as was mentioned, lack of design experience or the idea about this protection method even existing. I also have couple of alternative theories: In some cases those diodes are already integrated inside the output devices, e.g. such as in TIP142/TIP147 darlingtons. It has been said that trusting that these internal diodes exist or are even robust enough is bad practice, though. Some people simply do not care to make full-detailed schematics and trust that builders are wise enough to include all the neccessary protective or stabilizing elements to the design by themselves.
Click to expand...


hmmmmm

"""designers that they dont include things like that in the schematics presuming that the constructor will any way ......""""

i will buy that ..... i kind of like it a lot .....

the truth is that i did try to dig some old story to find out if inductor was allready existing inside the citation 12 when modified by Nelson Pass to a mosfet amp cause in the schematic posted by NP there is no one .....

i wonder if this is simply a printing mistake ???? was inductor located some where else inside the amp so is no part of the amplifier schematic .....

or just simply missed it ???

asked some related people in the forum that seem to constructed citation 12 but no luck .....
 
#11
rods input

this is a copy of rod's answer (ESP) to a simular post existing to ESP forum


Clamping diodes (i.e. from the output to each supply rail) are needed when the circuit has full protection (VI or Voltage - Current limiting). Audio signals don't change quickly enough with normal programme material to cause any reactive impulse voltages.

Any amp that has VI limiting must use clamping diodes, because the output can be effectively disconnected very quickly, leading to inductive spike voltages that can damage the amp if the diodes are omitted.
 
#13
No, Baker clamps are normally used to prevent or limit saturation of BJTs, for instance during overdrive or clipping, also during protection activation as effectively, if an NFB loop is present, when the protection kicks in, it is disconencted, because the output stage turns off.

In particular amps with foldback VI limiting MUST use clamping diodes or the output semiconductors get destroyed by back EMF, usually reverse polarizing the C-E pins, in effect what happens is that the back EMF pulse gets dissipated through reverse polarized B-E junctions of the output stage, working as (not very sturdy) zener diodes. Typically the output transistors get destroyed, also very often the drivers, but in some cases the drivers will measure fine on a meter but in fact suffer damage (usually resulting in sudden death of the outputs with no apaprent at some later date after the amp has been repaired).
I've seen an interesting implementation of what Andrew mentioned, usimg a rectifier bridge. + and - terminal to their respective power supplies, one AC terminal do GND, the other to the amp output.
 
#15
nigelwright7557 said:


But surely the back emf is absorbed by the complimentary transistor ?

I dont use clamp diodes in my amp designs but certainly do where relays are concerned.
Click to expand...


no the opposite transistor does not clamp inductive spikes across the output devices. the clamp diodes protect the output devices from inductive spikes that may exceed the rail voltage (and therefore reverse bias the output transistor and damage it). the clamp (more accurately "flyback snubber") diodes can be pretty much ignored in any discussion of how an amplifier normally works. but when it comes to building an amp (especially one for sale to the public) you want to add as many "reliability enhancers" as possible. you don't have any idea what load your customer intends on using with your amplifier (believe me, working in the business for many years, i have heard some real "doozies" from customers, as well as tried a few of my own). clamp diodes are a reliability enhancer, and is so inexpensive to include, that it should be a "no-brainer" (for an additional $0.50 you can keep the amp from blowing up, even if the customer uses it to drive an ignition coil). they may want to use it (thank a Sony Application Note for this one) as a variable frequency AC line source for an electric drill. as you can imagine, inductive spikes from a drill are pretty high in voltage, as well as being almost continuouslly generated while the drill is in operation try. this with an amp that has no clamp diodes, and you will need a constant supply of output transistors
 
Status
Not open for further replies.
Top Bottom