oops Hi rudi,
Even allowing for 230v AC at max., the peak-to-peak voltage here would not exceed 325-ish volts.
Accordingly, Farnell part # 752-344 or 956-5655 (Transient Voltage Suppressors # 1.5KE 400CA, with a reverse stand-off voltage of 342v) should be fine for you.
They are less Than UK £2.00 each (including Tax) here.
Regards,
EDIT: P.S. Don't tell anyone , but they will most probably improve the sound of your system too, if they are used permanently, as they get rid of a lot of 'hash' on the mains (like caused by lightning strikes etc).
Even allowing for 230v AC at max., the peak-to-peak voltage here would not exceed 325-ish volts.
Accordingly, Farnell part # 752-344 or 956-5655 (Transient Voltage Suppressors # 1.5KE 400CA, with a reverse stand-off voltage of 342v) should be fine for you.
They are less Than UK £2.00 each (including Tax) here.
Regards,
EDIT: P.S. Don't tell anyone , but they will most probably improve the sound of your system too, if they are used permanently, as they get rid of a lot of 'hash' on the mains (like caused by lightning strikes etc).
Hi Cal, fair enough, saves me a reply or two.
However, I feel the link which I took the trouble of posting is very relevant to the discussion in this thread, for better or worse, it offers information which hasn't been thus far freely given to those who would simply try values at random, as the OP did in this case.
It further explains why doing so is not the best idea, leading to unpredictable results. Rather it leads them towards the enlightenment of proper design, and the often predictable benefits that come with it.
Electronics is an art based on physics and science, not plug and play. I feel we should share that level of information in order to further the art, over all else. Please return the link.
However, I feel the link which I took the trouble of posting is very relevant to the discussion in this thread, for better or worse, it offers information which hasn't been thus far freely given to those who would simply try values at random, as the OP did in this case.
It further explains why doing so is not the best idea, leading to unpredictable results. Rather it leads them towards the enlightenment of proper design, and the often predictable benefits that come with it.
Electronics is an art based on physics and science, not plug and play. I feel we should share that level of information in order to further the art, over all else. Please return the link.
Thanks, don't blame you at all either, it was needed.
Here's the link on snubber design, there's alot of good tips in it, I'm not saying it's the entire story though, but what app note ever is right?
http://www.calex.com/pdf/3power_impedance.pdf
Here's the link on snubber design, there's alot of good tips in it, I'm not saying it's the entire story though, but what app note ever is right?
http://www.calex.com/pdf/3power_impedance.pdf
When they first came out, I was very interested in the idea, though it took a little while to understand exactly what it was that Carlos was trying to achieve. As I learn more about electronics, I find myself more and more in that situation, I want to understand exactly what is happening to produce the sound differences that we hear.
So, I tried the snubbers exactly as Carlos suggested, on two of my OPA GCs, one inverting, one non-inverting, and to be honest, I didn't notice much difference. I also tried them on my Krellclone prototype, again with little effect. From my experience, I suspect that if this doing something, and from the opinions stated here it seems to, then it is addressing a weakness of the LM chips. I suppose I should dismantle my LM4780 and try it on that one day, but I enjoy listening to it too much at the moment, and it's currently part of my main system. When I get all these other amps in the queue properly finished I might well give it a go.
Pinkmouse, you tried snubbers on S.L.O.W devices, exactly where you will detect less impact of a good, low impedance PSU.
Your OPA GCs, with some gain (around 10x and up), won't have a bandwidth of much more than ~50khz.
Even so, the snubber values could be optimized.
There we go... with no idea of what you're doing, it's hard to have any conclusion. Even so, some can be very sure of their conclusions!
There are no problems with LM chips, they are just much faster and higher bandwidth. And that demands a fast, low impedance PSU. And as these are power amps we're talking about, capacitance is also welcome, driving capability is not only the output stages, the PSU also helps a lot.
All this is something that is hard to understand for some...
Your OPA GCs, with some gain (around 10x and up), won't have a bandwidth of much more than ~50khz.
Even so, the snubber values could be optimized.
There we go... with no idea of what you're doing, it's hard to have any conclusion. Even so, some can be very sure of their conclusions!
There are no problems with LM chips, they are just much faster and higher bandwidth. And that demands a fast, low impedance PSU. And as these are power amps we're talking about, capacitance is also welcome, driving capability is not only the output stages, the PSU also helps a lot.
All this is something that is hard to understand for some...
Carlos, I do have an open mind, and I will try it on my LMs one day. But I have to say, and no offense meant, your posts didn't exactly give me much to go on to optimise the snubber for my OPAs. If you can tell me exactly how you do optimise them for a circuit I will give them another try, or if you don't want to reveal the theory, then I'll send you a couple of OPA549s for you to experiment with and you can tell me the components I should use.
I do appreciate the importance of a low impedance supply for audio amps, but being honest, I don't really see the need to optimise performance at bandwidths way above the audible frequency.
Oh, and before you write me off as an incurable objectivist, you might want to go back and read some of the discussions that Peter and I amongst others had discussing component and circuit choice long before we even had a separate chip amp forum here at diyAudio.
I do appreciate the importance of a low impedance supply for audio amps, but being honest, I don't really see the need to optimise performance at bandwidths way above the audible frequency.
Oh, and before you write me off as an incurable objectivist, you might want to go back and read some of the discussions that Peter and I amongst others had discussing component and circuit choice long before we even had a separate chip amp forum here at diyAudio.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.