Gentlemen,
The "bucking transformer" you suggest is formed by combining the primary with an auxiliary winding such that no galvanic isolation becomes the result. Such a transformer may be used as a pre-transformer before the
main transformer which will provide the galvanic isolation.
But, is that a simple solution when we may be about 1 Volt too high on the secondary side?
The "bucking transformer" you suggest is formed by combining the primary with an auxiliary winding such that no galvanic isolation becomes the result. Such a transformer may be used as a pre-transformer before the
main transformer which will provide the galvanic isolation.
But, is that a simple solution when we may be about 1 Volt too high on the secondary side?
Of course, the bucking transformer doesn't provide galvanic isolation but the big transformer will. It's really not a complicated solution and a small transformer might very well be much cheaper than any regulation scheme and much more reliable than any trick to gain a few volts.
Furthermore, it's unclear how much voltage we need to drop at this point. More than 1V that's for sure. 30vac will give 42.42Vdc unloaded, minus 0.7V for the diodes. But that doesn't account for the transformer regulation nor the mains voltage variations. To be on the safe side, I'd multiply that 30vac by 1.05*1.1... 35vac worst case. That means close to 50vdc.
On top of that, thermal consideration are getting touchy when the lm3886 is close to its voltage limits. See Tom's articles on taming the lm3886 for that. A 25vac transformer is about as high as I'd personally go with the lm3886.
Furthermore, it's unclear how much voltage we need to drop at this point. More than 1V that's for sure. 30vac will give 42.42Vdc unloaded, minus 0.7V for the diodes. But that doesn't account for the transformer regulation nor the mains voltage variations. To be on the safe side, I'd multiply that 30vac by 1.05*1.1... 35vac worst case. That means close to 50vdc.
On top of that, thermal consideration are getting touchy when the lm3886 is close to its voltage limits. See Tom's articles on taming the lm3886 for that. A 25vac transformer is about as high as I'd personally go with the lm3886.
"00940", I recognize and appreciate you have relevant considerations to support your proposal. We do not need to agree on what is the easiest to do in this situation where gimpchop has a transformer and would appreciate not buying another.
If we assume the DC voltage may get considerably above 42V and the LM3886 is touchy near it maximum supply voltage, I will support your second proposal: the zener diode and power-Darlington.
I often start out with the zener and Darlington, then I realize the zener is neither very precise nor able to handle the power loss, then I replace the zener with a TL431 (and a few supporting components), then I realize also the TL431 will get quite warm and I introduce a buffer transistor, then I realize the current feeding of the TL431 from the positive supply has a huge variation if I just use a resistor for which reason I replace it with a current source.
All-in-all, I have come to the conclusion that for those who better be recommended a simple solution, an LM317 (or LM337) with a buffer transistor is by far the most easy to handle. The buffer transistor need not be a Darlington.
A question to you: Is such a bucking transformer a standard product or does it need to be made for the exact purpose?
If we assume the DC voltage may get considerably above 42V and the LM3886 is touchy near it maximum supply voltage, I will support your second proposal: the zener diode and power-Darlington.
I often start out with the zener and Darlington, then I realize the zener is neither very precise nor able to handle the power loss, then I replace the zener with a TL431 (and a few supporting components), then I realize also the TL431 will get quite warm and I introduce a buffer transistor, then I realize the current feeding of the TL431 from the positive supply has a huge variation if I just use a resistor for which reason I replace it with a current source.
All-in-all, I have come to the conclusion that for those who better be recommended a simple solution, an LM317 (or LM337) with a buffer transistor is by far the most easy to handle. The buffer transistor need not be a Darlington.
A question to you: Is such a bucking transformer a standard product or does it need to be made for the exact purpose?
A good and logical solution.
However, I have more such transformers from Japanese amplifiers (also Pioneer) and they tend to have an outer (wide) copper-band, soldered together, that prevents access to the windings. It probably serves to reduce the leakage field. In order to get to the windings you need to open that long soldering and then start removing secondary windings. I would be afraid of causing damage to the windings that may render the transformer useless. This is why I would be hesitant to use this else logically correct solution.
Last edited:
It can be a standard product.
I don't know if gimpchop is on 230V or 110V. Let's say 230V. We want to drop about 30V from 230V. The total power needed can't be that much since it's a pair of lm3886 on 8ohm speakers. A 50VA/30vac transformer will provide 1.6A. Good enough to power the big transformer. Such a transformer is less than 20€ new.
On 110V, we don't need to drop that much voltage but we need more current. A 50VA with a 12Vac or 15Vac secondary would be perfect.
I get your point: a second transformer working in parallel with the first one and with the secondary windings of the secondary transformer subtracting (counter-phase) voltage from the secondary windings of the first transformer. Then the galvanic isolation is maintained and it can be a standard transformer.
From what I read from the article you referred to, he elaborates on the principles of the auto-transformer. Then, I combined that with Tom's comment about only handling primary current.
A possibility.
Am I wrong in concluding that if the first transformer has a dual winding with center-tap, the second bucking transformer need to have two separate windings (of the same size) so we can tie each of these correction windings to the ends of the two center-tapped windings?
From what I read from the article you referred to, he elaborates on the principles of the auto-transformer. Then, I combined that with Tom's comment about only handling primary current.
A possibility.
Am I wrong in concluding that if the first transformer has a dual winding with center-tap, the second bucking transformer need to have two separate windings (of the same size) so we can tie each of these correction windings to the ends of the two center-tapped windings?
Not really. The new transformer (the 50VA one) comes first. It is transformed into an autotransformer. 230VAC comes in, 200VAC comes out (without galvanic isolation). Then comes the big transformer which is simply fed 200VAC rather than 230VAC. The secondaries will fall by the same ratio. The second transformer provides the safety isolation.
I vote for LM317/337+ power transistors, lm338 or even some extra diodes. Or try 2 of this:
DC-DC 7-50V to 3-45V Max 6A Adjustable Buck Regulator Block Power Supply Module 699901899761 | eBay
DC-DC 7-50V to 3-45V Max 6A Adjustable Buck Regulator Block Power Supply Module 699901899761 | eBay
Dutiable rectifier board
Hi can anyone tell me if this rectifier board is suitable for running 2xlm3886 boards.pls thanks.
Rectifier Filter Power Supply Board Speaker Protection Omron DIY Kit f Amplifier | eBay
Hi can anyone tell me if this rectifier board is suitable for running 2xlm3886 boards.pls thanks.
Rectifier Filter Power Supply Board Speaker Protection Omron DIY Kit f Amplifier | eBay
@gimpchop - it's against the Forum Rules to start multiple threads on the same/similar topic, as a result, some of your threads have been merged.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.