That sounds like an 18 volt tranny. Offload and the voltage rises by the "regulation factor" of the particular transformer. The one in my link was 8% so that would give around 20 volts off load. And that was a 100va tranny. Smaller one have worse (higher percentage) regulation.
Its an 18-0-18
Thanks for your help!! The 18 V are all over the place, the 20 V was going to be hard to find, so that great to hear. Again, thanks!!
Your welcome
Well here's a smaller 60va with 11% regulation.
MCTA060/18 - MULTICOMP - 60VA TOROIDAL 2X18V | CPC From This Range
You'll be fine with an 18-0-18 (and that has to be what the original was). Its a case of seeing what is available in your area and that fits.
Well here's a smaller 60va with 11% regulation.
MCTA060/18 - MULTICOMP - 60VA TOROIDAL 2X18V | CPC From This Range
You'll be fine with an 18-0-18 (and that has to be what the original was). Its a case of seeing what is available in your area and that fits.
1) agree and add: check the datasheet for LM3886.
They are capable of driving 8 or 4 ohms but with different acceptable rail voltages.
If you know you have "50W" , now you need to check speaker ohms and you can choose which one is needed.
Of course, knowing expected rail voltage, AC value is 0.707 times that.
Just to play it safe, pick 1 VAC less per winding, if available.
You won't hear any difference and it's better to err "low" than "high".
2)
230V: Europe/Latin America/ 50% of Brazil/Australia/most of Asia/etc.
120V: USA/¿Canada?/50% of Brazil
100V: Japan
By itself, that's not a problem.
OEM supplier quality is the real one
They are capable of driving 8 or 4 ohms but with different acceptable rail voltages.
If you know you have "50W" , now you need to check speaker ohms and you can choose which one is needed.
Of course, knowing expected rail voltage, AC value is 0.707 times that.
Just to play it safe, pick 1 VAC less per winding, if available.
You won't hear any difference and it's better to err "low" than "high".
2)
No, it means it can be used in
230V: Europe/Latin America/ 50% of Brazil/Australia/most of Asia/etc.
120V: USA/¿Canada?/50% of Brazil
100V: Japan
By itself, that's not a problem.
OEM supplier quality is the real one
In my opinion, I'd go with an 18 volt secondary with a higher VA rating.
The 50VA rating probably contributed to the failure. Andrew is right when he states that 50VA will power a 50W amplifier, until being overdriven.
I would assume that it's going to be overdriven to a square wave output to be safe. This might be overkill, but, then again, you've already seen a transformer failure.
A quick calculation shows that 100VA would be close if delivering a square-wave to a 4 ohm load.
Since you shouldn't be able to tolerate that amount of distortion, the 100VA transformer should give a margin of safety and allow you to select an off the shelf that's close to that rating. For instance, 90VA would also be close enough.
I would not try to push my luck with a 20 volt transformer. What about high line conditions? The transformer will live. The LM3886 will too. But, the protection circuit will most likely limit your undistorted power to less than you'd get with an 18 volt transformer.
The 50VA rating probably contributed to the failure. Andrew is right when he states that 50VA will power a 50W amplifier, until being overdriven.
I would assume that it's going to be overdriven to a square wave output to be safe. This might be overkill, but, then again, you've already seen a transformer failure.
A quick calculation shows that 100VA would be close if delivering a square-wave to a 4 ohm load.
Since you shouldn't be able to tolerate that amount of distortion, the 100VA transformer should give a margin of safety and allow you to select an off the shelf that's close to that rating. For instance, 90VA would also be close enough.
I would not try to push my luck with a 20 volt transformer. What about high line conditions? The transformer will live. The LM3886 will too. But, the protection circuit will most likely limit your undistorted power to less than you'd get with an 18 volt transformer.
Sounds very reasonable to me.
But fomoco's suggestion of using a new transformer with somewhat reduced rated voltage sort of amounts to "pre-droop" to me
, and should take care of it.Otherwise it's like tossing the coin between "how will it fail today? ... burnt chipamp or burnt transformer?"
Oh well.
A classic Service Tech principle is: "Repair it, don't reegineer it" .... but I think *this* is one good exception to the rule.
Good point. But... If that's the case, the 3886 will just shutdown. Those things are hard to kill. Transformer lives... All he's out is the price difference between the two transformers.
Good idea, I too have been thinking about upping transformer size, and although the heat sink is rather massive, I am limited here by the size of the bracket that holds the transformer inside the monitor. The spacing is quite tight and I would have to get creative with new placement, something I unfortunately don't have the time for. This is too bad, considering that the 50 VA is driving not one, but two LM3886's.
On the other hand, each chip amp is wired right into a driver with no passive XO components in between. Moreover, a switch on the integrated active XO lets me choose a low end cut off frequency (flat, 60 Hz, 80 Hz) which I always use to limit the bass going to the woofer (and the corresponding LM3886) so my hope is that 50 VA will be sufficient.
Thanks for your help everyone, this makes me want to run across more of these "broken" units.
And a secondaries wiring question - the circuit uses each of the two secondaries per LM3886. What about combining the secondaries in parallel instead, and feeding both LM3886's from the same rails (or is it rail)? Would this not work better, given that the tweeter amp will use less current, and pooling all current together is a better way to load balance?
Thanks for your input!!
Thanks for your input!!
Last edited:
I can't give a definite answer to that... different properties of the core material perhaps, other transformer rated for 50/60hz and replacement 60hz. Dunno
Looking back at the pictures at the start, the windings are series connected to derive the split (-/+) rails. You have to stick with that arrangement.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.