Your company should be thinkig about mass production and wide market supply of "encapsulated PS", designed with "serious use of industrial methods", this could cause a jump in sound quality of average audio gear.
Hi,
That and mass production of certain custom parts specified by "experienced engineer".
Ciao T
That and mass production of certain custom parts specified by "experienced engineer".
Ciao T
Now to complete my DIY modification of the HCA-3500 into an 'A' quality amplifier.
Let's look at the power supply again. Then remove the non-linear resistors at the input of the amp, replace the power switch and fuse, with a circuit breaker, instead. Substitute aluminum for iron top cover, etc. That is 'all' there is to it.
Let's look at the power supply again. Then remove the non-linear resistors at the input of the amp, replace the power switch and fuse, with a circuit breaker, instead. Substitute aluminum for iron top cover, etc. That is 'all' there is to it.
Attachments
John,
I like to keep real fuses, with good holders they are preferable to circuit breakers. But both are good options.
I would also leave the NTC's in place, but instead bypass them after power-up using a whopping big relay.
There is one more item you do not mention.
IMNSHO the protection circuit should be changed with a muting relay on the input (or possibly a shunt at the VAS output - collapsing both the VAS output to ground and thus also collapsing the output stage Bias) handling overload and a crowbar relay handling output DC protection with the input mute relay being "OR'd" with that.
As the Mute relay will be a small signal one it should engage faster than the output crowbar, so if the DC protection is triggered by bad signals, DC on the input etc. the output short circuit will have nothing to "short" out, so if the Amp is not already fatally damaged it will do nothing.
And if an output transistor (or a bank) has gone south I do not trust the usual output relays to reliably interrupt the current instead of arcing over and welding shut (I tested that in the 80's). So a crowbar with mute is the correct way to protect speakers from blown Amp's.
In my current day Amp's the power to the mains transformer is under software control, so in case of DC protect engaging (which takes one of the two interrupts - remote is the second on the MCU) I also switch off the power to the output stage... And yes, they include mute at the power stage input and a crowbar on the output.
From a sonic viewpoint it gives you a straight line to the binding posts, no relay contacts, which can be between okayish and horrible beyond belief.
Ciao T
I like to keep real fuses, with good holders they are preferable to circuit breakers. But both are good options.
I would also leave the NTC's in place, but instead bypass them after power-up using a whopping big relay.
There is one more item you do not mention.
IMNSHO the protection circuit should be changed with a muting relay on the input (or possibly a shunt at the VAS output - collapsing both the VAS output to ground and thus also collapsing the output stage Bias) handling overload and a crowbar relay handling output DC protection with the input mute relay being "OR'd" with that.
As the Mute relay will be a small signal one it should engage faster than the output crowbar, so if the DC protection is triggered by bad signals, DC on the input etc. the output short circuit will have nothing to "short" out, so if the Amp is not already fatally damaged it will do nothing.
And if an output transistor (or a bank) has gone south I do not trust the usual output relays to reliably interrupt the current instead of arcing over and welding shut (I tested that in the 80's). So a crowbar with mute is the correct way to protect speakers from blown Amp's.
In my current day Amp's the power to the mains transformer is under software control, so in case of DC protect engaging (which takes one of the two interrupts - remote is the second on the MCU) I also switch off the power to the output stage... And yes, they include mute at the power stage input and a crowbar on the output.
From a sonic viewpoint it gives you a straight line to the binding posts, no relay contacts, which can be between okayish and horrible beyond belief.
Ciao T
Hi,
There are SO many areas where thinking things through can create significant improvements, I prefer to combine them all into a product...
Ciao T
There are SO many areas where thinking things through can create significant improvements, I prefer to combine them all into a product...
Ciao T
Well, everyone, although we still had a questionable circuit board layout, and a marginal power supply turn-on solution, we made a pretty good sounding power amp about 12 years ago that we could take to CES with pride. I don't expect everyone to follow our example, in fact, you would be wrong to do so, BUT the SPIRIT of the changes, including simplifying, better parts quality, wiring, and non-magnetic connectors, high speed diodes, etc, etc, should be considered, if not followed.
While I appreciate any constructive criticism, please don't presume that I don't know a potential problem, such as removing the non-linear resistors on the AC input, instead of adding a relay. We got away with it, however, at the time. Later we added the relay.
While I appreciate any constructive criticism, please don't presume that I don't know a potential problem, such as removing the non-linear resistors on the AC input, instead of adding a relay. We got away with it, however, at the time. Later we added the relay.
Ah yes.....That delicate balance between public reputation, failure rates, commercial insurance, and sonic quality (as a given priority).
The multiple fuse (eg, 3x times 3.3A fuse, instead of a single 10A fuse) trick I proposed works quite well, when it comes to linearizing the (as perceived-by ear) damage that fuses do to dynamic draw considerations. The requisite protection still exists, it is just that it is more linear under dynamic draw. Thus making the fusing more sonically 'palatable' to the ear...
The multiple fuse (eg, 3x times 3.3A fuse, instead of a single 10A fuse) trick I proposed works quite well, when it comes to linearizing the (as perceived-by ear) damage that fuses do to dynamic draw considerations. The requisite protection still exists, it is just that it is more linear under dynamic draw. Thus making the fusing more sonically 'palatable' to the ear...
Last edited:
The person with unlimited budget also has to pay for products that customers want, because they can't afford ultimate quality production. Like Soviet government paid for electronics for military made by manufacturers and designers who were paid by the same government.
John,
I agree with all these changes. I keep some extremely strong ND magnet around, to test parts for magnetism, with it you can "feel" if binding posts or RCA's have nickel plate in them (this is a VERY dangerous magnet BTW, get your finger between it and a steel plate and you may loose it).
John, the suggestions where not so much offered so much to correct or help you (we know you know), but to offer solutions to the community, in support of your work and contribution.
Ciao T
I agree with all these changes. I keep some extremely strong ND magnet around, to test parts for magnetism, with it you can "feel" if binding posts or RCA's have nickel plate in them (this is a VERY dangerous magnet BTW, get your finger between it and a steel plate and you may loose it).
John, the suggestions where not so much offered so much to correct or help you (we know you know), but to offer solutions to the community, in support of your work and contribution.
Ciao T
It can be a form of script as well.
the script (cash/value) brought to the table being 'unshared' lore or tweaks/tricks, that have yet to see the light of day.
the industry is so stilted these days that even a simple chassis tweak can catapult a given design to the top of the heap.
Like a horse race run far too close, the differentials are down to inches.
Which is hugely boring, to say the least.
So, we iron down the methodology of correct thought in the genesis and execution of the given gear, to get to the 95% complete level.
Then the payoff that takes it to the last inch of winning the race. the 5% that is new or at the more complete level of thought and integration. The last bit of competitive thumb wrestling that puts one thumb over the tip of the other.
I think we are overdue to have a bit of entirely new innovation take the lead position in this tired horse race.
With cables, I've tried to do that, via utilizing signal and transfer in the proper medium and environment, for the first time in living electrical memory.
The difference is there, even in the basic measurements. for if one takes one of the cables, cuts the signal entirely, the decay is so fast, it is the equivalent of a few picofarads (and I do mean a few), on an approximate 1.1M finished and terminated RCA cable.
Standard application of standard theory will tell you that it is impossible, but the measurement is not wrong. The situation is that we are getting quantum response considerations rearing their heads, as we have entered the ionic plasma area, and not the gross Newtonian analysis that has been the standard for about 150 years. Both are valid but only one is applicable to all, and that is the quantum analysis, in this case. Thus the cable is fundamentally different and is a true change and innovation. Standard electrical theory and application becomes a doorstop, that holds the door open to the quantum analysis. In a real world way, which is the entirely uncommon part.
I will admit to being naughty on purpose. Call it extreme irritation with the public status quo of mental positioning.
I have drawn blood with them, BTW the expensive SS "non-magnetic" tweezers we use in fab stick quite well.
- Status
- Not open for further replies.