| Dominick22 |
I have been seeing alot lately about speaker protection and I do not have this in my LM4780 amp. Just wondering if this is a huge risk I am taking that I should remedy fast or not.
Also, Some of my speakers have built in protectors. Is this suffiecient or should I still be running pros on the amp itself??
Dominick |
|
|
| Stocker |
How many amplifiers have you fried by shorting the terminals or overloading the output stage? If = >1, then you might consider protection. If = >5, seriously consider it.
I for one, don't think it worth the effort, especially for a chipamp as economical to replace/repair as the LM-series. |
|
|
| Dominick22 |
Well, I have fried one chip although I am not %100 sure why. So the speaker pro protects the chip??
I thought it protected the speaker?? Call me crazy!LOL
*not trying to be sarcastic-I am really trying to understand*
Dominick |
|
|
| Peter Daniel |
My chip amps are constantly powered on and so far I didn't experience any problems (for last two, three years). However, one needs to be certain that the source equipment does not produce any DC offset.
Whenever I change something with the equipment, making mods to the amps, connecting new sources or change PS, I always disconnect speakers from the amp first, and before connecting them back measure the DC offset.
At one time I was experimenting with a preamp and that produced 24V DC at the speakers, but I acted quicly and it didn't damage anything. I got only once in trouble blowing two of my Triangle midwoofers ($250 ea.) but it happened with a different amp: the A75.
The drivers I'm presently using are quite expensive ($800pc and 1000Euro/pc) but I still chose not to use any protection. You can check one of the older threads dealing with a subject: http://www.diyaudio.com/forums/show...t=&pagenumber=1
The chips are protected for output short and will not produce DC offset if one rail fails. So you don't really need a circuit to protect chips, if anything, you need something to protect the speakers. |
|
|
| carlosfm |
| quote: | Originally posted by Peter Daniel
The chips are protected for output short and will not produce DC offset if one rail fails. |
Peter, they will produce DC offset if one rail fails.
Otherwise you wouldn't need an output cap when using single supply. |
|
|
| davidsrsb |
| Another thing never mentioned in the data sheets is what will happen if one of the supply rails goes open (likely with fuse protection) or short to ground. |
|
|
| Peter Daniel |
| quote: | Originally posted by carlosfm
Peter, they will produce DC offset if one rail fails.
Otherwise you wouldn't need an output cap when using single supply. |
I checked it, and it does not produce offset. But you need a load connected, otherwise you will see the offset. |
|
|
| mastertech |
Dominick22 -
I have been seeing alot lately about speaker protection and I do not have this in my LM4780 amp. Just wondering if this is a huge risk I am taking that I should remedy fast or not.
Also, Some of my speakers have built in protectors. Is this suffiecient or should I still be running pros on the amp itself??
Dominick
-------------------------------------------------
those spks that have protection yes theyre protected
typical of spks protection is provided with polyswitches,fuses ...,etc
cheers |
|
|
| Bazukaz |
| Well , if we need to protect loadspeakers only , why not use a simple fuse ? It should blow much quicker than a loadspeaker , if current rating is chosen correctly... |
|
|
| Upupa Epops |
| Dominick, look at www.czechaudio.com, PA03 amplifier, there you can to see protection for 4780.... |
|
|
| Dominick22 |
Upupa,
That lead was dead, but it would be great if you found one that worked!
Bazukaz,
Are you suggesting that I put a fuse on the output wires from the LM4780 amp just before the speaker terminals??
Would they be the same current as the input fuses?
Wouldn't that cause some sort of degredation in the audio signal?
Thanks for the great replies!!!
Dominick |
|
|
| Bobken |
Hi Dominick,
Please see my post #21 on "Fuses impact on sonics of a supply", for some relevant experiences of what this will do to the sound.
Regards,:eek: |
|
|
| Bazukaz |
| Well , if you are afraid that fuses are bad for sound , there is a simple DC proctetion circtut at sound.westhost.com |
|
|
| Bobken |
Hi Bazukaz,
I know there is!;)
Regards, |
|
|
| Nordic |
| I can see how a fuse can protect speakers against an amp doing its job too well, but what about low DC output errors that can easily fry speakers, would those pop the fuse? |
|
|
| Bobken |
Hi Nordic,
Probably not, but I would rather not find out!:bawling:
Regards, |
|
|
| Bazukaz |
| A low DC output should not blow the speaker easy.Tweeters and midranges usually have a cap in filter and are not sensitive to this.If amplifier is designed well , it should have DC blocking on its input. |
|
|
| Nordic |
What I'm saying is if the amp fails enough DC could get to the speakers to destroy them, but it may be less than whats needed to blow the speakers in the conventional way...
More a question than a statement I guess. |
|
|
| Peter Daniel |
| I mentioned previously that once it happened to me to blow Triangle mdranges. At that time, I also had Focal Audiom woofers in that system (12" and 15" in parallel). It took me about a minute to figure out what's going on before I disconnected the power. The full rail supply voltage (40V DC) was delivered to speakers for approx 1min. and it didn't damage the woofers. After 10 years they still work fine. |
|
|
| Bobken |
Hi,
Few, if any, LF drivers will have a cap in series with it, and therefore they are vulnerable here to any DC voltages output by the amp.
Anything (series cap, or whatever) at the amp's input will not have any effect on this, whatsoever, except to prevent any DC from source components from being passed through and amplified by the power amp.
To quote Doug Self from his well-regarded book "Audio Power Amp Design":
"Fuses are sometimes recommended for DC offset protection, but their only merit is cheapness"
"Selection of the correct fuse size is not at all easy. If the fuse rating is small and fast enough to provide some real loudspeaker protection, then it is likely to be liable to nuisance blowing on large bass transients"
"Fuses running within sight of their nominal current generate distortion at LF"
"Fuses obviously have significant resistance... so putting one in series with the output will degrade the theoretical damping factor"
These explain the phenomen I noticed many years ago when I carried out my tests outlined in the thread I referred to before, so I would stay well clear of them, if I were you.:xeye:
Regards,:) |
|
|
| Bobken |
Hi,
I've just seen Peter's comments, and I can second what he says here.
I did the same with 75 volts DC into an ATC driver for maybe 30 seconds when an output mosfet failed on me, and the full rail voltage was output.
However, ATC drivers are some of the most robust drivers in the world, and are also used in some of the finest Pro speakers used in many recording studios.
Whether lesser speakers will take much punishment would depend on how well they are made, and what the voltage and duration is, of course.
Whatever, fuses in the outputs of amps are simply not a satisfactory way of dealing with these problems, if they arise, in most knowledgeable engineers' views.
Regards,:) |
|
|
| Upupa Epops |
| If you find membrane of speaker on oposite wall, maybe you will start thinking about fuses... :D |
|
|
| Bobken |
Hi,
Not if you have any sense!
You might think about a proper DC sensing circuit and a relay to protect the speakers, though.;)
Regards,:) |
|
|
| Upupa Epops |
| It is any problem for me... ;) |
|
|
| JensRasmussen |
My advice would be to use some sort of protection of the speakers. Never mind if the amp blows – that is easy to fix, the problem is often the speakers.
So use some sort of relay to remove connection to the speakers in event of a DC on the output of the amp. I can live with what ever sonic signature the relay may have as long as my speakers will not be destroyed by me making a stupid mistake meddling with the equipment.
EDIT: This also enables you to get rid of any turn on thumb that may be present either from amp or connected equepment by using the same relay as a turn on muting/delay devise.
\Jens |
|
|
| Eva |
| I suggest fuses in the supply rails and a DC-activated crowbaw in the output implemented with a large enough triac. This allows to use fuses rated at a higher current, as the short-circuit formed by the triac should always blow them. |
|
|
| Upupa Epops |
| It is not good solution, Eva ( if you mean shorting both rails against each other ) - theoretically one of fuses can interrupt faster and in this case you can get DC voltage at output. Better is shorting rails before fuses ( with thyristor connected in series with small ohmic value resistor - thyristor have better delta I / delta t ) and is clear, that in this case you need ANY fuses ( only at primare winding of transformer ). |
|
|
| Bazukaz |
Well , for me it is very strange Dominick22 could blow his LM based amp.
Generally , these chips fail so unoften , that there is very little probability to get DC rails at output(I have tried various types of short circuit , including bridged amp - ICs standed this well).Spike protection has output transistor temperature monitoring , and shuts them down if temp is exceeded.Maybe it isn't fast enough in some cases ?
I wonder if any commercial designs use additional protection for IC amplifiers ?
Lukas |
|
|
| mastertech |
Bazukaz -
Well , if we need to protect loadspeakers only , why not use a simple fuse ? It should blow much quicker than a loadspeaker , if current rating is chosen correctly...
--------------------------------------------------
would this calculation be correct Bazukaz ?
(Prms/Rload sqr-root) = Ifuse
fast or slow blow?
quality of fuse?
Kind Regards |
|
|
| Eva |
| quote: | Originally posted by Upupa Epops
It is not good solution, Eva ( if you mean shorting both rails against each other ) - theoretically one of fuses can interrupt faster and in this case you can get DC voltage at output. Better is shorting rails before fuses ( with thyristor connected in series with small ohmic value resistor - thyristor have better delta I / delta t ) and is clear, that in this case you need ANY fuses ( only at primare winding of transformer ). |
What the triac has to short are the output terminals of the faulty channel of the amplifier, thus protecting the speaker and blowing the rail fuses if the output stage insists in producing DC. I thought that the trick was quite obvious. |
|
|
| Upupa Epops |
| In this case you will blow only one rail's fuse, so amp will get only one polarity voltage and after while circuit will be activate again, at this time against oposite voltage - not all amps are closed by one polarity rail's voltage.... |
|
|
| anatech |
Hi Eva,
Yes, triac is very effective, and really fast. It needs to be a "turn off to reset" deal. I've seen it used in really large amplifiers. Impressive.
Upupa,
This will blow both rails. A high energy failure will typically short the triac (good). So now it's an expensive piece of wire too. Mind you, I've seen it blow traces 1/4" wide for 1/2" back. Confuses the heck out of some technicians! :D
All amplifiers can go DC, chip amps included. All amps should have some sort of speaker protection.
-Chris |
|
|
| JensRasmussen |
| quote: | Originally posted by mastertech
Bazukaz -
Well , if we need to protect loadspeakers only , why not use a simple fuse ? It should blow much quicker than a loadspeaker , if current rating is chosen correctly...
--------------------------------------------------
would this calculation be correct Bazukaz ?
(Prms/Rload sqr-root) = Ifuse
fast or slow blow?
quality of fuse?
Kind Regards |
No, you need to calculate the I^2't from your output current.
Use this so select you fusetype.
You calculation says only what current the fuse must be able to withstand in avarage, not the energy it needs to absorbe.
BTW fusedesign/selection is hard!
\Jens |
|
|
| davidsrsb |
Relay protection is the only way to go.
The sense circuit can monitor both dc and power and be made frequency sensitive to disconnect if too much power is being sent to the tweeter.
Repairing an amplifier is much cheaper than scrapping a pair of speakers because you have fried one driver and cannot get a spare. |
|
|
| Eva |
Relays are far from perfect because they will arc for several seconds when trying to interrupt a high DC current flowing into a highly inductive load. Repeating that event several times may damage the contacs and cause them to latch.
Relays are much more reliable in AC applications where the arcing event terminates very quickly as the current becomes zero several times a second. |
|
|
| anatech |
Hi Eva,
Yes, put the relay in series with the speaker output. One contact instead of two (for a fuse). When teh contacts are bad, replace the relay - no biggie.
-Chris |
|
|
| Upupa Epops |
| You must use correct relay - look at type Fider's 65... line and Nais's TV-15 line... They are similar and Nais specify correct function by 30 A DC... |
|
|
| Upupa Epops |
| ...Finder's... ;) |
|
|
| carlosfm |
Surely there are apropriate relays for the job.
Otherwise you couldn't turn on the lights on your car more than a couple of times without changing the relay. :D |
|
|
| Bobken |
Hi,
You are quite right, of course there are Carlos, and there is one 'audio' relay specially designed for this purpose, too.
It has twin contacts with one very HD contact taking the 'heat' and doing all the switching, and a secondary HQ contact which 'follows' the other, for maintaining the electrical integrity when the 'main' contacts become pitted etc., which of course is bad for the sound.
Probably because it was me who first mentioned relays, that would be enough for Eva to disagree, anyway, as this seems to have happened before.
We all know that nothing in electronics is 'perfect', and that most designers' choices are inevitably a compromise. However, using a relay here is about the most sensible way overall, and that is why most manufacturers will adopt this method of protection.
Personally, I don't like the slight sonic degradation which every relay I have tested, appears to cause, but for safety, this has to be accepted by those who put safety before ultimate sound.
Regards,:) |
|
|
| Eva |
I was quite serious in my previous post about DC and contacts. I found out by myself what massive contact arcing was the day that I tested my 2KW PFC converter, that has 440V DC output, with two 230V 600W slightly inductive (wirewound) electric heaters connected in series as a load.
Altough the current flowing was only 3A DC and the switches of the heaters were rated at 10A 250V AC with more than 2mm contact separation, the contacts arced happily for 5 seconds or so after the switch was turned off. It was quite impressive because arcing could be heard and seen through the otherwise opaque plastic of the switches.
Concerning automotive applications, arcing here is not so troublesome because the voltages and the inductances involved are very small. Car lamps doesn't have 2mH inductance nor are powered from +-90V rails as big woofers do. It should be noted that car relays end up failing, and that all car manufacturers have moved towards extensive solid state switching, supressing most of the classic mechanical contacts. If your car is less than 5 years old, most lights are probably switched through MOSFETs. Furthermore, reputable German manufacturers like BMW, Audi and Mercedes have been extensively employing solid state switches instead of mechanical contacts from the past 15 years.
|
|
|
| Upupa Epops |
| Eva, look a datasheets of relays which I posted...they have contact distance minimally 4 mm... |
|
|
| Eva |
This is from Finder's catalog:
The figure on the right side of the page shows DC voltage breaking capabilities versus current of your beloved Finder 65 relays with 4mm contact spacing. As you will see, the DC current rating becomes ridiculously small as the DC voltage is increased. Now put a 2mH inductor in series with relay contacts...
Relays are mostly AC devices and quite ineffective for DC protection. As the previous graph shows, an high-quality 30A 250V AC relay priced over $10 is only rated at 1A DC for 100V DC operation. Now figure out the DC rating of smaller and cheaper relays.
|
|
|
| Bgt |
| quote: | Originally posted by davidsrsb
Relay protection is the only way to go.
The sense circuit can monitor both dc and power and be made frequency sensitive to disconnect if too much power is being sent to the tweeter.
Repairing an amplifier is much cheaper than scrapping a pair of speakers because you have fried one driver and cannot get a spare. |
Absolutely agree. Proper DC suppl. line fuses together with an ultrafast relay board is the most comfortable protection 1 can think of to protect your expensive speakers. I used it for over 20 years and never had a speaker blown. And I have seen a lot of burned out amps. and ejected cones with so called fuse protected speaker outputs. Speed is everything in an uncontrolled situation. Will never switch on any amp. without sensitive relay prot. |
|
|
| Upupa Epops |
| Yes Eva, it is thruth, but in this case is problem little bit different, 'cos relay work like urepeated circuit breaker and will not switch DC voltage continually. Distance of contact must be larger than by for AC switching relay, relay must be able arc rupture and it is single-use process... Probably best solution will be combination of both protection, relay + triac ( or thyristor ;) )... |
|
|
| anatech |
One must realize there is a difference between high power sound reinforcement amps, and amplifiers that will be found in a living room. The magnitude of energy in normal use is much smaller.
I would tend to use a triac (limit dv/dt so it doesn't self trigger) in larger, high power applications.
-Chris |
|
|
| Eva |
| Self-triggering due to excessive dV/dt is another interesting feature of triacs. Any parasitistic oscillation present in the output with enough amplitude will trigger them. |
|
|
| anatech |
Hi Eva,
I used low impedance gate to ground resistors bypassed with 0.01 uF.
If the amp is oscillating, it would be good to trigger it and just blow the fuses, rather than the output stage and whatever tweeter used to be there.
-Chris |
|
|
| mastertech |
JensRasmussen []
Flux Inverter
diyAudio Member
Offline
Registered: Jul 2002
Location: Denmark - Jutland
Post #34
quote:
Originally posted by mastertech
Bazukaz -
Well , if we need to protect loadspeakers only , why not use a simple fuse ? It should blow much quicker than a loadspeaker , if current rating is chosen correctly...
--------------------------------------------------
would this calculation be correct Bazukaz ?
(Prms/Rload sqr-root) = Ifuse
fast or slow blow?
quality of fuse?
Kind Regards
No, you need to calculate the I^2't from your output current.
Use this so select you fusetype.
You calculation says only what current the fuse must be able to withstand in avarage, not the energy it needs to absorbe.
BTW fusedesign/selection is hard!
\Jens
----------------------------------------------------
Jens you holy cow one the best designers here to correct
my calculations you put time in the equation amazing
"BTW fusedesign/selection is hard!"
absolutely
thanks my good friend
kind regards |
|
|
| Dominick22 |
OK-
It has happened again, you all are talking at a level way above my head on a thread that I started!!
LOL:D
I love this site,
Dominick |
|
|
| davidsrsb |
Replying to Eva
The relay would normally go open when the amp is switched off and the music should be muted.
Faults should not happen too often.
Properly rated relays are reliable devices, some of the relays in a car are switching very inductive loads like window motors. The trend to solid state now is too cut cost, space, weight and power consumption.
If you try to run a speaker through one of the telecoms style 2A pcb relays of course the contacts will weld. |
|
|
| Eva |
This is a large scale version of what happens when a relay tries to stop the current flowing through a very inductive load. It's even worse when the current is DC instead of AC.
http://www.arcfault.org/video
Note how the arc takes several seconds to extinguish as I described in my 440V DC experience with a few milimeters of contact separation :D:D:D (The video shows the same effect but for 500KV and a few *meters* contact separation).
davidsrsb:
In a car you have only 12V so the arcing capability is ridiculously small, and the inductance of any 12V motor is far below the 1mH figures commonly found in big speakers. In one of my previous posts there is a figure ilustrating that: a high quality relay rated well over 30A DC at 12V is just rated below 1A DC at 100V :eek: Things get quite funny as voltage is increased.
|
|
|
| carlosfm |
The higher the voltage, the more distant the contacts have to be from each other. This is for every type of switch.
If using a high voltage with a small switch, an arc will be produced, high voltage jumps to the nearest path.
An engine spark plug just works because of the very high voltage that goes through it. |
|
|
| forr |
What about an RC series circuit across the relay contacts ? I've seen that in a Bang and Olufsen amp. Will it minimise the arcing ?
And what about "polyswitches" ?
~~~~~ Forr
§§§ |
|
|
| anatech |
Hi Forr,| quote: | | What about an RC series circuit across the relay contacts ? Will it minimise the arcing ? |
Yes, it will extend the contact life in normal operation.| quote: | | And what about "polyswitches" ? |
They have a higher on resistance after they have operated onee or times. Stick with a standard speaker relay.
-Chris |
|
|
|
