I'm assuming that both cases (for comparison) are a CRC supply, and what we're looking at is the benefit of a bypass diode for the resistor.
Next, the caps discharge as required to power the amp for moving the speaker. That energy comes from the caps (there would be no sound if there was no discharge). Even at this point, the voltage difference in our comparison, is not significant.
Once the power has been expelled (as is required to move the speaker), then we are left momentarily unprepared at the power supply reservoir.
Clarity returns AFTER the caps regain full charge, so I suggest that we might like to recover charge very quickly. This is the point where the difference has become significant.
Options include:
1). Buy a much larger transformer (more current capacity = charges faster), or
2). Short the "R" of the CRC with a solid copper wire (aka don't use CRC filtering), or
3). Diode bypass the "R" of the CRC to limit loss.
Option #3 above filters similar to a normal CRC and is cheaper than replacing the transformer with a higher current model.
No difference. Even a supply that is both weak and lossy, is well able to get the caps up to full voltage prior to a bass beat.
Next, the caps discharge as required to power the amp for moving the speaker. That energy comes from the caps (there would be no sound if there was no discharge). Even at this point, the voltage difference in our comparison, is not significant.
Once the power has been expelled (as is required to move the speaker), then we are left momentarily unprepared at the power supply reservoir.
Clarity returns AFTER the caps regain full charge, so I suggest that we might like to recover charge very quickly. This is the point where the difference has become significant.
Options include:
1). Buy a much larger transformer (more current capacity = charges faster), or
2). Short the "R" of the CRC with a solid copper wire (aka don't use CRC filtering), or
3). Diode bypass the "R" of the CRC to limit loss.
Option #3 above filters similar to a normal CRC and is cheaper than replacing the transformer with a higher current model.
Last edited:
Individually snubbing fast diodes results in a little bit worse performance than something simpler. So, I actually can't answer your question.
P.S.
Alternatively, the *features* diode like Fairchild Stealth, is good to use when you're fairly sure that the builder will Omit the snubbers, and that is why I used those diodes on the schematic.
Hello all...
Although Im not a master of simulations, I tried to make a simple transformer PSU model with this D//R filter in it.
And it does indeed have some sort of affect to the output, compered to a RC filter alone.
But we certainly are talking about only a millivolts difference here. Still definitely worth investigating some more, if "all out performance" is needed...
Notice, however, that a spice guru can make much better and more accurate simulation than me!
Although Im not a master of simulations, I tried to make a simple transformer PSU model with this D//R filter in it.
And it does indeed have some sort of affect to the output, compered to a RC filter alone.
But we certainly are talking about only a millivolts difference here. Still definitely worth investigating some more, if "all out performance" is needed...
Notice, however, that a spice guru can make much better and more accurate simulation than me!
4) Use choke input.
Blasphemy, I know. (Repeat after me "Iron is evil")
But better regulation and less noise.
Probably one of the better explanations here
The diodes provide a direct current path in case of a big (short) load step. In normal operation, the voltage drop over the resistors is smaller then the forward voltage of the diode, so filtering via CRC is active. When the diode is conducting, CRC becomes CC i guess.
This will reduce sag, as the secondary capacitors will get current much faster.
?
This will reduce sag, as the secondary capacitors will get current much faster.
?
Will they?
The first C is a singleton electrolytic.
The last C is a multiple bank of many electrolytics.
This bank of electrolytics is sized to allow the amplifier to replay whatever bass signal is applied through the input filter.
Adding another C via diode does NOT speed up the capacitor bank, ("the secondary capacitors will get current much faster").
The mains charges up the First C during a duty cycle that is <<20% and often around 5% to 8%.
This first C charges up the main bank when ever there is a voltage across the dropping resistor. If there is no voltage drop then there is no charging.
If the Vdrop exceeds ~700mV, then either he has sized his PSU components incorrectly, or the amp is attempting to output a signal that is not music.
The diode does NOT
During that 5% to 8% period, if the single C's Voltage ever gets to be more than one diode drop higher than the bank of C's after the diode, the transformer will also be directly charging the bank through the diode.
edit: I just saw what you posted about 700mv. I would think that is a matter of opinion. I'm not sure that the CRC necessarily has to be designed in such a way that it is providing full benefit during the highest peaks of output.
Last edited:
I think it is FedEx and UPS that reliably "process" all shipments through UK customs thereby attracting VAT at 20% which is applied on the total cost of the item, including shipping from the US or China for example, and on top they add a £14 "processing fee".
On the other hand items sent via other courier companies or normal post do not suffer like that.
Might try changing the diode in the simulation to MBR1645 or MUR820, for a tradeoff of a little bit less CRC filtering in exchange for a little more power. They can't completely turn off the CRC filtering anyway, since even the most efficient high current diode does have a voltage drop, and therefore the CRC filtering is never shut off (albeit we might could say "momentarily hindered").
To test those "all out performance" conditions, in fact, I did just put some MBR1645 crc bypass diodes in mine and cranked it up loud, with the digital bass boost switched on at the source.
Oh, yes, almost as much power as shorting the CRC's resistor, but more clarity than shorting the CRC's resistor. There an appreciable greater amount of bass slam than when using the 6a05 for bypass diode. In these cases, the bypass diode does exactly what it looks like it will do. With only the resistor, the more power you want, the more heat you get instead, and the bypass diode can be added to limit the waste. So, there's just not a mystery at that part of the circuit.
And, you're right, of course, the bypass diode is probably only applicable for loud replay.
That's like dual mono without the wallet smash effect. The $4 worth of added diodes is a LOT less expensive than the additional transformer and power board it would take for real dual mono.
In reference to this schematic <--link and on the topic of fun with schottky. . .
The output diodes on the left side of the schematic is for the virtual dual mono feature.
With virtual dual mono, the function is in making the left channel amplifier decoupling caps a bit separated from the right channel amplifier decoupling caps. As a bonus, working decoupling sounds really good.
Typical virtual dual mono schemes use separate left and right channel big regulators at the dc output side of the power board (or at the amplifier board). I have merely replaced that idea with something less effective, easier, more efficient, and less expensive (a schottky).
P.S.
Here's a TDA7297 single rail amplifier using virtual dual mono with simple diodes (in addition to an off-board DC power supply).
Actually, I used the MUR820 model. I did not have any 6A diodes in "Tina"(TI).
Anyways... One purpose of the CRC filter is to get rid of possible noise from the mains.
What is your opinion on the added diode in this context?
and even when the import is exempt from VAT and exempt from import tax, they still apply their fee.
That has happened twice recently.
Our Royal Mail do the same.
Why does the postage and handling cost charged by the seller not cover the Courier's costs?
I KNEW I SHOULDN'T HAVE ASKED. YOU'VE SAID WHAT YOU THINK THE DIODES DO, NOW TELL ME HOW EXACTLY THEY GIVE YOU THIS CLAIMED VIRTUAL DUAL MONO EFFECT. BY WHICH FEATURE OF DIODE OPERATION??
I GET THE DISTINCT IMPRESSION YOU'RE JUST REGURGITATING WHAT YOU'VE READ ELSEWHERE WITHOUT ANY UNDERSTANDING OF THE SCIENCE BEHIND THE CLAIMS.
Last edited:
But the lower forward voltage drop alone will have a beneficial effect? For example I am using 4 1N5402 in a bridge configuration to rectify 25VAC. When I draw say 5A DC they heat up so much that they have melted nearby components (5mm away). Now I know the 5402 is only rated at 3A without some cooling, but still, is it not true if I had used the SB3100 instead I would experience less heat?
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.