Hi all,
I have a multi-way system, several amplifiers, one for frequency range. A collection of chipamps TI LM3875, LM3886, LM4766 and LM4780.
Multiple systems with electronic crossover, provides more dynamic with quality and low power amplifiers (except bass & sub-bass). And in the search for more perfection in music reproduction have read recommendations on the coupling capacitor at the input, usually MKP or bipolar 1 to 10 uF. Also some designers recommend bypass for more dynamic or better transient response.
A question about not using coupling capacitor at the input, connecting with a bare copper wire jumper it would be safe with her spike protection(TI) system where any DC signal would be barred in the output of your amplifier to avoid damage to the speakers? Is it possible that TI chips overture&spike protection DC output?
Thank you in advance helps friends
I have a multi-way system, several amplifiers, one for frequency range. A collection of chipamps TI LM3875, LM3886, LM4766 and LM4780.
Multiple systems with electronic crossover, provides more dynamic with quality and low power amplifiers (except bass & sub-bass). And in the search for more perfection in music reproduction have read recommendations on the coupling capacitor at the input, usually MKP or bipolar 1 to 10 uF. Also some designers recommend bypass for more dynamic or better transient response.
A question about not using coupling capacitor at the input, connecting with a bare copper wire jumper it would be safe with her spike protection(TI) system where any DC signal would be barred in the output of your amplifier to avoid damage to the speakers? Is it possible that TI chips overture&spike protection DC output?
Thank you in advance helps friends
I'm no expert but the physics of what a coupling cap does suggests that yes it will affect your transient response especially above 1uF due to the recovery time. I use a 0.1uf (formerly 0.22uf) as a coupling cap in my TubeLab SSE KT88 amp, but this is a high voltage connection. If this is on the input connector I don't think it will hurt to remove it and at least hear how it sounds. Again I'm not an expert so I could be totally wrong about this.
Sent from my iPhone using Tapatalk
Sent from my iPhone using Tapatalk
The physics of what a coupling cap does says no it won't affect your transient response. The physics says that cap value affects low frequency rolloff. The physics says it protects your amp (and possible speakers too) from DC - sources should not emit DC, but either through faults or poor design they sometimes do. I would leave it in place.
I thank everyone and help me with your opinions. If more members to comment, the better. A designer indicated suppress coupling cap (best transientes), which seems to make sense.
I would ask this "designer" for their credentials, there's no validity to that statement. A properly utilized capacitor (size, type, voltage rating, etc.) will behave like a piece of wire for the desired signals.
Mike
Honestly, I really love rigging the coupling caps to facilitate a combination of better transients for more lifelike audio and better speaker support as well.
Conversely, it is less labor and lower cost to do without the couplers, although I do consider the cost to be a severe quality shortfall, and as such, at risk of lacking for relevancy.
Problem:
To enjoy, or to be able to fine tune your couplers to your hearts desire, a prerequisite is a Stable amplifier, and most of the cheap kits just aren't.
Possible solution:
Phil made groundbreaking headway by combining AndrewT's advices and mine as well, for this: http://www.diyaudio.com/forums/chip-amps/265827-lm3886-fullrange.html#post4142277 <--link
It has been reviewed as being more enjoyable. The couplers are working well.
Conversely, it is less labor and lower cost to do without the couplers, although I do consider the cost to be a severe quality shortfall, and as such, at risk of lacking for relevancy.
Problem:
To enjoy, or to be able to fine tune your couplers to your hearts desire, a prerequisite is a Stable amplifier, and most of the cheap kits just aren't.
Possible solution:
Phil made groundbreaking headway by combining AndrewT's advices and mine as well, for this: http://www.diyaudio.com/forums/chip-amps/265827-lm3886-fullrange.html#post4142277 <--link
It has been reviewed as being more enjoyable. The couplers are working well.
If you're after improvement in sound, use a signal trafo to couple your amps rather than capacitors. All the advantages of DC protection remain.
Do you need coupling caps at the input at all ? Does your crossover have a DC componenet on it's output ?
If you really do need them you can get away with smaller ones for the midrange and tweeter amps. Just pick a roll-of that's about 1/10 to 1/5 of your crossover frequency and you will be safe regarding phase shift in the audible range. This way you can use higher quality caps and don't need to bypass anything. If your crossover is DC free just let them out. Best capacitor is no capacitor.
If you really do need them you can get away with smaller ones for the midrange and tweeter amps. Just pick a roll-of that's about 1/10 to 1/5 of your crossover frequency and you will be safe regarding phase shift in the audible range. This way you can use higher quality caps and don't need to bypass anything. If your crossover is DC free just let them out. Best capacitor is no capacitor.
Thanks for clarifications through its opinions. I think it's not our will cite authors of ideas and the absurd they may seem to then make criticisms or even jokes. If I'm not mistaken, the doubt is the fuel to our discussions, exchanging opinions and experiences. Thank you all again.
I tend to remove input capacitors on amplifiers i make, for my self.
after i made sure , verry sure there is no hazard of DC.
there are times when its impossible to do without risking my speakers/other stuff, then i leave them in place. I generaly found that 10 uF Wima caps work quite well in that case.
after i made sure , verry sure there is no hazard of DC.
there are times when its impossible to do without risking my speakers/other stuff, then i leave them in place. I generaly found that 10 uF Wima caps work quite well in that case.
I never use them. Test my circuits with a 10000uf in series with output. If I've got no DC on output before the cap then Im good. Wouldn't do that if I didn't build my own speakers though.
Hi udailey, generally are you able to notice any differences in the sound quality between the direct output and the 10000uf series output ?
i personally did no mesurements.
my ears are better suited.
10 uf is supposedly transparent, but bass is absolutely tighter with no input caps.
for some odd reason stereo imaging is allso noticably far superior.
dunno,. something with phases? i actually did not even try to find out.
my ears are better suited.
10 uf is supposedly transparent, but bass is absolutely tighter with no input caps.
for some odd reason stereo imaging is allso noticably far superior.
dunno,. something with phases? i actually did not even try to find out.
I prefer no cap. Its just there for safety while I test. You have to determine how hard those speakers are to replace. I HAVE toasted one.
Thanks... I asked that cause I also destroyed an expensive FE168Ez. My LM3875 had a 10uf cap at entry but unfortunately it got an internal short circuit so the cap does not helped 🙁
My conclusion is that if you want to DC protect your driver, it must be at the output end, no at the entry 😉
But the problem I see with big output caps - apart from the sound quality - is they are polarised and also not supposed to work at 0VDC tension with AC only. And if one supply rail fail they can explode, then you need to create big NP caps... pfff.
So, why not to use toroidal transformers, only both secondary as a 1/1 ratio output trans. Abraxalito (in post #8 and in his blog) advises to use signal trafo and Shoog use toroidal transformers (with double primary) in push-pull tube amps, from 20hz to 60khz, not too bad…
For a stereo LM3875 amp, two 2x22 VAC @100 VA possibly would made the job… it will need some tests. Note the cost of both trans is about half the price of a FE168EZ.
Some ideas or experiences about that kind of alternative solution ?
My conclusion is that if you want to DC protect your driver, it must be at the output end, no at the entry 😉
But the problem I see with big output caps - apart from the sound quality - is they are polarised and also not supposed to work at 0VDC tension with AC only. And if one supply rail fail they can explode, then you need to create big NP caps... pfff.
So, why not to use toroidal transformers, only both secondary as a 1/1 ratio output trans. Abraxalito (in post #8 and in his blog) advises to use signal trafo and Shoog use toroidal transformers (with double primary) in push-pull tube amps, from 20hz to 60khz, not too bad…
For a stereo LM3875 amp, two 2x22 VAC @100 VA possibly would made the job… it will need some tests. Note the cost of both trans is about half the price of a FE168EZ.
Some ideas or experiences about that kind of alternative solution ?
Big toroidial transformer at the output wll surely protect yor loudspeakers from any DC component, but the primary side has such a low DC resistance that it will almost act like a short circuit at DC for the amplifier. This is a different situation from output transformers in tube amps.
If you look at it that way, yes. Personally I wouldn't be too worried about amp failure or DC offset. I have been building amps for years without this kind of protection and have never destroyed a loudspeaker in my life.
I see your points, Sjef and udailey...
Technically, for AC only, you just have to take care of enough turn/Volt at the trans : if your amp provide 40 Watt on 8R driver => the max output is 18V... so a 2x18 (or more) trans would be ok and at 1/1 ratio the amp will be charged at 8R. Ok...
Now, in the case of DC at the output... but the amp is already destroyed, no ? At least you can heard a strange bad sound at your driver and turn off the toaster... oops I mean the amp ! 😉
Technically, for AC only, you just have to take care of enough turn/Volt at the trans : if your amp provide 40 Watt on 8R driver => the max output is 18V... so a 2x18 (or more) trans would be ok and at 1/1 ratio the amp will be charged at 8R. Ok...
Now, in the case of DC at the output... but the amp is already destroyed, no ? At least you can heard a strange bad sound at your driver and turn off the toaster... oops I mean the amp ! 😉
- Status
- Not open for further replies.