Loudspeaker protection kit

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
i think there are such kits on lamington road in mumbai. normally all you need todo is protect the tweeter. what amp -speaker combo are u using?

tweeters usually blow if your amp is a bit too small for your needs. you push the amp a bit to hard it overloads and clips. creates all sorts of HF noise which the XO directs towards the tweeter and wham.

i find all protection ckts change the sound in a way i dont like. in a pro - pa enviroment they are useful but for audiophile use i doubt it.

woofers usualy complain so you get a chance to turn the volume down.
 
Hi Navin,
I made an amp published in Electronics For You in 1991. It has also appeared in their Electronics Projects No.13. It is a 70 W amp designed by C Sanjay. You might have seen the article.
The amp has fuses in series with the outputs to protect the speaker in case of DC appearing at the speakers. But I do not know how reliable it will be.
The speakers are DIY with Peerless drivers. Naturally, I do not want my efforts to go up in smoke. What do you suggest?

Vivek
 
fuses? that is odd. there are better methods. however i have not seen the design you are talking about so i cant tell what would suit the design best. the easiest is a DC blocking cap. however that affects sounds quality most. there are other current managing/sensing circuits that can also work. most fuses are not fast enough to protect the speaker.
 
Fuses just for DC blocking?

Navin,

Is it possible that this protection circuit that Vivek refers to uses fuses only for DC protection? I would have thought that a speaker can handle a few seconds of DC, in which case the fuse might be good enough? Can't speak from experience...

I'd have thought that a proper speaker protection circuit would protect against transients, turn-on "thumps", etc. I've seen some; they use relays. Randy Slone's book has one. I've wondered about the sonic effects of a fat relay in the poweramp output path, but I haven't yet built one so can't say from experience. In any case, I would have thought that if the amp is really high-powered, a speaker protection circuit would be advisable. (Really high powered = 200W/ch or more). Randy Slone's take on this is that it is reasonable to expect that some goof-up somewhere will expose your speaker to some sort of transient or DC at least once in its life, even in the home. :)

Tarun
 
navin said:
Most protection ckts have a sonic character. the digital ones are a bit beter.

What is a digital protection ckt? You mean something which has sensors triggering a microcontroller? I guess such things can have more complex analyses and responses, no? Are there any designs/kits for DIY? These PIC and Atmel microcontrollers are quite incredible... both from their processing power and inexpensive price points of view. If I wasn't fiddling with audio, I'd be fiddling with those, I guess. :)

Incidentally, you're the only person I know of in Bombay who talks of Jordan full range drivers. I have a pair in small boxes. They are probably the best speaker drivers I've heard, which is not saying much because I've not heard much. :)

Tarun
 
My JX92s... ah well

navin said:
you got a pair. are they for sale? actually the smallest application i got needs 5. eventually it'd need 20. 4 per channel. 5 channel.

I'm totally in love with the drivers myself. And I paid a big markup for them (I paid full courier charges for the pair from the UK to India, which came to UKP 50+.) So you wouldn't like to buy them from me... Anyone who travels as much as you do can probably pick them up from the UK and avoid the exorbitant courier charges, I presume. If you don't mind the price and you don't mind the fact that I'll be removing them from enclosures where they're mounted (in effect they're "used" units), then you can have them, I guess. Email me.

I have made the small single-driver bookshelf speakers using those JX92s, and this has encouraged me to try building something bigger to fill in the lows and highs. I'll add a tweeter and a large bass unit to each of these, someday. Maybe even try open baffle for the JX92s, instead of ported box. But currently, no time or budget to build that augmented system, so the little JX92 boxes are lying unused.

When are you making your full-range based speakers? Can I come listen? :) I'll be real quiet. :)

Tarun
 
A lovely way to handle clipping

navin said:
tweeters usually blow if your amp is a bit too small for your needs. you push the amp a bit to hard it overloads and clips. creates all sorts of HF noise which the XO directs towards the tweeter and wham.

i find all protection ckts change the sound in a way i dont like. in a pro - pa enviroment they are useful but for audiophile use i doubt it.

Do you know of this lovely approach to clipping which one of Randy Slone's amps has?

Randy claims that one of the main reasons tube amps sound more pleasant than SS (solid state, not single sided) at high volumes is because they do "soft clipping", i.e. when they clip, the waveform does not have sharply-sliced-off tops and bottoms. They have rounded corners. In frequency terms, this "soft clipping" will introduce less high-order harmonics than the sharply sliced off waveforms of SS amps. And he goes on to say that we often drive our amps to points where they are occasionally clipping, but we don't realise it, so we just ascribe the final sound to the "characteristics" of the amp, where actually it's clipping-induced distortion.

So, to address this problem, he designed a totally symmetrical amp topology, with MOSFET output stages, and then introduced some protection circuitry which actually mimics the soft clipping. The protection circuitry part is very simple, and uses (probably) four transistors to control the biasing of the OPS driver transistors. But to get that rounded waveform from a SS amp... it's quite something else. And like all Randy Slone amps, this one has distortion and transient characteristics which are super hi-fi anyway.

I was reminded of this because of your post about clippings, high-order harmonics, and tweeter blowups. I guess this soft clipping will not only give the amp a more euphonic sound when over-driven, it'll also help protect the tweeter, won't it? Simply because it'll generate much less higher order harmonics?

Tarun
 
Re: A lovely way to handle clipping

tcpip said:

Anyone who travels as much as you do can probably pick them up

When are you making your full-range based speakers? Can I come listen? :) I'll be real quiet. :)

Tarun

Hold your breath, I have not even decided on the drivers. but you are welcome to listen to my existing speakers/amp/preamp. there are photos elsewhere on this forum.

tcpip said:

Randy claims that one of the main reasons tube amps sound more pleasant than SS (solid state, not single sided) at high volumes is because they do "soft clipping", i.e. when they clip, the waveform does not have sharply-sliced-off tops and bottoms.

So, to address this problem, he designed a totally symmetrical amp topology, with MOSFET output stages...
I was reminded of this because of your post about clippings, high-order harmonics, and tweeter blowups.
Tarun

Yes this is much nicer if expensive solution. A cheaper solution is to Tri amp and use a tube amp for the tweeter there are many 5-10W tube amps out there to be built.

Maybe if I get really nutty I will do this...

5 channel HT system using 11-12 mono amps.

Amp 1 & 12: 200-300W Class D for the sub(s) (upto 100Hz)
Amp 2, 4 ,6, 8, 10: 10 W SE or 50W Pass labs for mids
Amp 3, 5, 7, 9, 11: 10W triode tube amp

first let me get the speakers done. :)
 
The easiest way to achieve soft clipping is to use back to back zeners in the feedback path. But I will be bit reluctant to do that because the zeners may introduce non linearities in the non clip mode of the amp.

Fuse or relay in series with the speaker is the common method to protection, but again they will leave their sonic signature.

If you want to maintain the sonic purity and not put any contacts in series with the speaker, the best thing is to sense the current in the speaker and some how shut the power supply down. This may save the speakers from frying but still one has to think of finding an elegant solution as how to eliminate the switch-on thump without series relays.

Small bulbs are also used in series with tweeters, but again for audiophile use I don't think anyone will prefer this however bright the idea might seem...or sound.

I am still puzzled about the Digital protection kit. What is it. Does it fire a '1' for protection and '0' for 'keep your fingers crossed'...

Tarun the Jondans that you ordered from UK, what % of duty did you have to pay. As an individual can one import speakers even if they are under OGL.

Angshu
 
Re: Re: A lovely way to handle clipping

navin said:

Yes this is much nicer if expensive solution. A cheaper solution is to Tri amp and use a tube amp for the tweeter there are many 5-10W tube amps out there to be built.

Actually, Navin, I don't know why this should be expensive, if you want to build your own amp anyway. The price difference between a good amp without soft clipping and Randy Slone's amp will be insignificant... he has just added a few transistors to his design, that's all. Of course, you may still need to add other, more severe kinds of protection, because this soft clipping thing will not guarantee you protection the way a more thorough protection circuit will. But I seem to have missed the point about this being expensive...

And yes, tri-amping (not cheap, please!!) will be absolutely luvverly. Actually, ever since I built the Jordan bookshelf boxes, I've been dreaming of building a three-way, bi-amped, with active crossovers, crossing over at, say 300 Hz, using 4th order LR filters. The low-pass output would drive a 200W Randy Slone L-MOSFET design (the one with the soft clipping, incidentally) which would drive a (or two, depending on budget) big woofers in an open baffle.

The high-pass would drive an ultra-low-distortion Randy Slone amp with BJT outputs, 80W/channel. This would feed the Jordan full-range driver on an open baffle. In parallel with the Jordan driver, I'll connect a good (depending again on budget) dome tweeter with just a single cap. (Should I think of ribbon? Are they all shockingly expensive?) I've been day-dreaming of this system ever since I heard the Jordans.

Will open baffles be okay for deep, realistic bass, do you think? Should I think of twin-opposed bass drivers in push-push configuration, like a lot of people have been talking about on your thread recently? Have you read Lynn Olson's article about push-push subwoofers in sealed enclosures? I found it very interesting: http://www.aloha-audio.com/ME2txt.html

I guess the Jordan FR drivers can be an easier job with open baffle, and I shouldn't even need a particularly large baffle for it. I listen to music; I'm not a HT fan. But yes, I do listen to hard rock like Deep Purple's "Wring that neck" too. :)

Tarun
 
Angshu,

angshudas said:
Fuse or relay in series with the speaker is the common method to protection, but again they will leave their sonic signature.

Do you know of any actual listening tests done with and without high-quality relays in the speaker signal path? When a relay's NC connectors are being used to pass the current, there's not even any current in the coil. Will just the passive relay contacts affect the sound of the system? Not disagreeing, just curious.

I had long read that interconnects sharply affect the sound of a system. I had home-made interconnects made out of Monster RCA connectors and Gotham coax cable. I then went and bought Van den Hull cables. I tried comparisons. The difference in sound was so small that I'm still not sure it was not a figment of my imagination. Since then, I've been skeptical about sonic effects of good quality, well made passive components.

As a counterpoint, I've heard speaker cables making a difference. When I first bought my current system, I didn't get speaker cables with it. I just bought 15amp electrical copper cable and hooked it up, for the time being. After a week, when the speaker cable came (300-strand OFC, unbranded, imported), and I hooked it up, I immediately saw that there were more details, the highs were smoother, cleaner, more "open", and the system sounded better. I believe that the electrical wiring I'd used earlier was not "good enough" to compare with any good quality speaker cable. I'm not sure I'll hear any difference if I compare my unbranded 300-strand OFC to a super-duper uranium-enriched silicone-impregnated plutonium cable. But then, I've not done any comparisons after that one experience... don't have the money for experiments unless I believe I'll get something out of it. :)

If it shows up in THD measurements, I believe it'll have an impact on the sound. If it doesn't show up in THD or dynamic IMD measurements, and it's well designed and well made, i.e. no huge capacitances or loose connections, then I'm not sure there will be sonic impact. What do I know, I am just an engineer who likes to listen to music on his cheap audio system. :)

I accept that systems more revealing than mine can show up sonic impact of passive components more. My system is not the world's most revealing.


If you want to maintain the sonic purity and not put any contacts in series with the speaker, the best thing is to sense the current in the speaker and some how shut the power supply down. This may save the speakers from frying but still one has to think of finding an elegant solution as how to eliminate the switch-on thump without series relays.

What about using relays in the power supply path of the poweramp, and bringing in current flow to it in steps, first through resistors, then through less resistors, and finally by bypassing all resistors? Put this stepped-resistance circuit before the smoothing capacitor, so that the inrush current into the caps can themselves be "slowed down" a bit. Surely relays in the power supply rails will not affect the sound of the amp?

I am still puzzled about the Digital protection kit. What is it. Does it fire a '1' for protection and '0' for 'keep your fingers crossed'...

:)
I guess the current sensors, DC sensors, etc, feed inputs to a microprocessor, which can then control relays (maybe solid-state ones) to the power supply if needed.

Tarun the Jondans that you ordered from UK, what % of duty did you have to pay. As an individual can one import speakers even if they are under OGL.

An importer can import practically anything (which is not on the banned list) legally by paying duty as per the books. An individual can also import practically anything, I think, provided it is in small quantities and the Customs chaps do not feel that the quantities appear to be too big for personal use, or else they'll believe this is a commercial import and begin to ask for import licence and all sorts of other documents. For you or me to import a pair of drivers, these headaches are not there.

When I imported my Jordans, the courier just brought the drivers through Customs, and said that they had decided not to charge any duty, believing that this small quantity was "sample". Samples can be imported without duty, provided you can convince Customs that it is a sample. The Customs chaps are unpredictable.

Tarun
 
Re: Re: Re: A lovely way to handle clipping

tcpip said:

Actually, Navin, I don't know why this should be expensive, if you want to build your own amp anyway. The price difference between a good amp without soft clipping and Randy Slone's amp will be insignificant... he has just added a few transistors to his design, that's all.

I guess the Jordan FR drivers can be an easier job with open baffle, and I shouldn't even need a particularly large baffle for it. I listen to music; I'm not a HT fan. But yes, I do listen to hard rock like Deep Purple's "Wring that neck" too. :)
Tarun

well i did not know his ckt. i assume he was using a current sensing device and then drop the primary voltage to reduce the amps level. in any case i usually biamp. so triamp is only one more amp for me. so far most speaker i have built are 3 way with a woofer going to 100Hz a midbass covering 100-2/3k and a tweeter doing the rest.

one amp below 100Hz and one above helps a lot.

however i am tiring from these systems. also i have run out of relatives who will baby sit my creations. so my next system will be for me.
 
Re: Re: Re: Re: A lovely way to handle clipping

navin said:

in any case i usually biamp. so triamp is only one more amp for me. so far most speaker i have built are 3 way with a woofer going to 100Hz a midbass covering 100-2/3k and a tweeter doing the rest.

Do you use active filters before your (bi)amps and do away with passive XO totally, or do you just connect the amps to the separate terminals of the speakers, but allow the passive crossovers inside the speakers to remain? I guess if you use commercially made speakers, you have no choice but to do the latter, right?

one amp below 100Hz and one above helps a lot.

Yes, I'm sure.

However, ever since I read that about 50% of the energy content of "normal" music is below about 350 Hz, I was wondering about using a higher XO freq between bass and mid+high. Basically, if I can split the full 10-octaves at a point where about 50% of the energy goes into each amp, then I can use about equal-sized amps. With a lower XO frequency, I'll have to make a bigger amp for the mid+highs, won't I?

If I had my way, I'd use one amp to cover the full range of high sensitivity of human hearing (about 300 - 3K or so, I believe) plus one octave on either side, and make this a really high definition amp and use a FR driver to handle this range in the speaker. But if I do this, then I'll have to XO the bass-mid at about 150-200Hz, in which case I'll need a bigger power amp for the mids+highs than the bass. Which is fine, except that the most high definition amps I've seen have lower power outputs. It's easier to crank up the power with slightly higher distortion, and moreover (at least for solid state) this distortion is mostly in the higher frequencies. This means that it's easy to get a big-muscle 200-400W amp for the bass, and its distortions won't be audible because it's only working in the sub-300Hz region anyway. I'd like to choose the highest quality amp for the mids, sacrificing power rating if needed. (I don't necessarily want to do Class A, but even top-end Class B amps have sub-100W ratings.) Once I do that, I'll probably get only 80-100W amps of the ultra high quality. And if this amp has to be bigger than my bass amp, I can use not more than about 60-80W effectively for the bass. Where do I get the SPLs then?

Basically, I need more amplifying power for the mids+highs because that's how the energy distribution in music is. And I also need the highest definition amp for the mids+highs, because that's where the ear is most sensitive. Unfortunately, the highest-definition amps are not the most powerful. That's the long and short of it.

Compromises, compromises. And they say it's fun. :)

What amps do you use, BTW? Do you build your own?

however i am tiring from these systems. also i have run out of relatives who will baby sit my creations. so my next system will be for me.

I know what you mean. :) I haven't yet reached your stage. I have one numbers father and one numbers father-in-law who would love to get an amp+speaker combo from me, if I can find the time and money. :)

Tarun
 
Tarun, I have not heard of any tests about the sonic performance of relays, but less the number of contacts between your amp and speaker the better.

There is an interesting debate going on in this forum about speaker cables. The original author did some double blind tests and his participants got fooled. Personally I feel, there is lot of hype about high end/high priced cables and interconnect. I still feel the speaker is the weakest link in the chain. Once you have a reasonable cable, the contribution made by the cables are overshadowed by the deficiencies of the speakers. In the same breath I also admit that our hearing is still much more sophisticated then the test signals devised so far to actually make an objective measurement of what we hear. So the ear can still pick up things we cannot measure. SO if you hear the difference go for it.

For the time being I will be satisfied with the DIY interconnects and speaker cables as suggested by John Risch. http://www.geocities.com/jonrisch/cables.htm. I have also generated his test signals, but have not tried it on any cables yet.

As for tweeter protection, in the absence of soft clipping amps, one can try a passive low pass filter to cut-off at 20KHz or which ever highest frequency one can hear. I have not tried it, but this is worth experimenting. I have not blown that many tweeters....as I listen to music for personal enjoyment and not to test the elasticity of my eardrums.

Angshu
 
Re: Re: Re: Re: Re: A lovely way to handle clipping

tcpip said:

Do you use active filters before your (bi)amps and do away with passive XO totally...However, ever since I read that about 50% of the energy content of "normal" music is below about 350 Hz,
Compromises, compromises. And they say it's fun. :)
What amps do you use, BTW? Do you build your own?
I know what you mean. :) I haven't yet reached your stage. I have one numbers father and one numbers father-in-law who would love to get an amp+speaker combo from me, if I can find the time and money. :)
Tarun

1. I use a PLLXO
2. yes 300 Hz is an approx half way point but if you do that your woofer and mid have to be in phae as phase problems are very audible. my woofers are overgrown coffee tables. pics of my existing stereo are elsewhere on this forum. to reudce this i chose 80 Hz as the XO freq. since my mids are 2 x 8546 it does not stress them too much.
3. my father inspired, paid for and was even a bit involved in my earlier creations (1972-1980). After that I was on my own. so far both my brother in laws (wifes brother and sister hubby have got speakers from me)
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.