Necessary to use speaker protection in LM3886 gainclone?

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Everything will fail eventually, doesn't matter how cheap or how good it is, it will fail.

It depends on how you think ?

Purists will argue that you don't want anything in the signal path, ie relays.

Built well, the time between failures will be high, but so will the cost of such a failure.

Even the protection circuitry can fail - there's no protection against that.
 
Yes Minion, you can bridge three LM3886 without problems. My PCB is similar to this:
Shine7 BPA300

I think I'll go for the speaker protection circuit. Actually, once, with my old TDA7294 based amp, one of the wires for the power slipped out of its socket, and made the Chip blow, and the speaker go crazy. I want to prevent that scenario form happening again.

But how does this speaker protection circuits work? Does it cut when the voltage becomes too high? Or does it break the circuit when it observes DC?
 
The speaker protection is usually a relay in normal open, connecting the amplifier output with the speaker.
When you turn the equipment on the protection circuit will close the relay after a few seconds accounting for any start bump, and will open the relay at any time if DC normally above 2V is present at the amplifier output.
 
It seems like there are two types of speaker protection kits at ebay. One is using the UPC1237, which is a dedicated speaker protection IC, and the other circuits is just using transistors and some other basic componentes.
Here are the links:
UPC1237 Speaker Protection IC Datasheet
Speaker Protection Without UPC1237
Speaker Protection With UPC1237



Offtopic:
I've been using this forum for a while, basically just for reading. Still there are one thing at this forum that I truly don't understand; How in the world can a single forum post have 200+ pages (!!)? Many of these posts are many years old! It really makes is hard to read the whole post (If you have the same probles as the post describes) because several members "hijacks" the thread, and taking it in a new direction.
Can someone please tell me why this is done on this forum, but never on other forums? :)
 
Last edited:
well, a simple protection from DC is...
a capacitor.
Yess, many will consider this as something that will ruin the sonic quality of the whole thing.
mm , got news, if the cap is large enough, it won't be noticeable by any means for a human.
and will surely block any DC from the speakers.
 
Actually is at least another one type of protection for speakers, output shorting (with a triac) when DC is present at output following to blows the output fuse.
This type of protection it is used for high power amplifiers and generate minimum problems comparing with other solution.
 
Last edited:
I was also thinking about adding a fuse to the output, so if I accidentally shorted the output, I wouldn't blow the whole amp up. However, the LM3886 datasheet says that the chip have an internal short circuit protection to ground. Have anyone tested this feature?
I want my amp so sound nice, bout robust enough to let other people handle the connections. :) I've heard that adding a fuse may influence the sound. Anyone got any experience on this?
 
...
I've heard that adding a fuse may influence the sound. Anyone got any experience on this?

It does influence the sound...everything influences the sound.

But most of these changes I can't tell so I don't worry about it.

My ears get "broken in" to the sound of whatever amp I use.

Build your amp the way you want to and listen to it.

If you like it, great.

If you want to change it, change it.

IMHO

:)
 
I was also thinking about adding a fuse to the output, so if I accidentally shorted the output, I wouldn't blow the whole amp up. However, the LM3886 datasheet says that the chip have an internal short circuit protection to ground. Have anyone tested this feature?
I want my amp so sound nice, bout robust enough to let other people handle the connections. :) I've heard that adding a fuse may influence the sound. Anyone got any experience on this?


Actually, this may be the practical solution to the thorny question of what to do for protection. It's on thin data, but I've has an A60 (indeed my favourite A60) blow up in front of me due to a huge over-voltage on the supply (monitored by a friend's UPS down the road) and it did no harm to my speakers. Nor have I ever heard of an A60 - and I've used and abused dozens of them - ever damage a pair of speakers. And their protection goes no further than a pair of 3.15A fuses on each rail and 1.6A on the outputs to each speaker. Slow blow in both cases, I think.

Of course you can hear them, as someone says above, but if you didn't know they were there, what else would you blame? And, in any case, if you can do that in one of the best sounding amps in its era, then it's hardly for us to decide otherwise in a cheap chip amp. They might even soften up the sound, which is never less than penetrating with a 3886, so you might even prefer it.

I have never had one of these amps blow a speaker, though I have had a few of them blow, but mine are slightly more peculiar circumstances since I would dick around with them while they were turned on, or forget that a capacitor was fully charged. Not really the same thing as domestic use.

By and large I think they are pretty robust, but I'm not above thinking that they get damaged if abused - even if they don't blow. The answer I would have given to Helmut would probably have been to put a current limiter on the supply and so protect both the chip and your speakers (we hope). In any case, at least the chip will usually go before the speakers. But simple protection, however primitive it seems in the light of today's audio principles, may be the best way to go. I may even take my own advice.
 
I have a micro controller using the ACS712 hall effect current sensor to protect the speakers using a solid state relay. When the current reached the tripping point the program in the micro open the relay. This is using a micro. I think that you can use the output from the acs712 to feed the input of a comparator op-amp and by adjusting the comparator voltage you can adjust the tripping point to a given current and open the speaker output relay. It is a simple circuit that any one can built.

Now I am working a micro controller that will eliminate the output relays to the speaker by controlling the regulating the rails voltage output. When the output current reached the tripping point it will removed the regulated rails voltage output consequently protecting the speaker without relays . By doing it in this way I will eliminated the output relays. Right now it is in development. I still will used the ACS712 for monitor the speaker current.
 
I came over some speaker protection circuits at ebay, and wonder if this is necessary to use!

If I were you, it would depend on what I want to protect:
(1) The chip (2) The Transformer (3) The speaker

(1) The chip is too cheap for the low probability of failure. I will not used the protector even if I have one ready to be used.

(2) The transformer can be protected with other means. I will not use speaker protector.

(3) My speaker is expensive, and I have invested a lot of time to get the best out of the drivers. Still, I dare to omit speaker protector when I know the condition of the amp, and I am the only one who operate the system.

Every new amps should not be used with my speaker. At least one day is necessary to pair the amp with cheap speaker before it can be considered safe.

Problem may arise from accidental input disconnect (my wife or my son did this!) when the power is high enough. The tweeter can easily be fried. This happens many times with me, that's why I don't want to use a Gainclone, instead I use lower power amp, such as TDA2030A.

Amplifier stability (oscillation) is the main issue. If I know the amp is rock stable, I will take the risk. Instability may come from capacitive cable, but depends on the build, such as PCB trace and whether some measures have been taken to increase stability.

To make sure the Gainclone is stable, check:
(1) The gain should not be lower than the recommended example from the manufacturer.
(2) Commercial kits are usually more stable than necessary. You may find a small capacitor (100pF-220pF).

With such measures, I will run the Gainclone for a few days with cheap speakers, try to use difficult load (subwoofers or big 3-way) and check that there is no critical temperature increase, then I will use it with my main speaker. But I will not leave it unattended because of the possible accidental input disconnect issue.
 
Thanks! I'll consider using a fuse on the output! :)

But about the speaker protection kits, should I choose a transisor driven or UPC1237 driven protection? Is one more reliable that the other?
I've also seen some kits that uses Omron relays. These are usually twice as expencive as the models that's using other relay brands.

I'n which way may the Omron relay be better? More copper? larger surface that the signal can flow trough?

EDIT:
Why can't you use 3xLM3886 in BTL mode? I posted a link at the first page showing a guy who did this.
 
Last edited:
................I've also seen some kits that uses Omron relays. These are usually twice as expencive as the models that's using other relay brands.

I'n which way may the Omron relay be better? More copper? larger surface that the signal can flow trough?...............
More probably because OMRON and other reputable manufacturers test their product to PROVE they meet specification and warranty their product for fair use within the limits of the specification.

I wonder how many "cheap" relays are fully up to the specification characteristics and how many offer a warranty?
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.