Just upgraded my Musical Fidelity A1 to a pair of MA50 monoblocks. Talk about improvement! But the caps are original and need replacing. Currently 4 x 6800uF 63V per monoblock, each cap 31mm diameter x 46mm long, radial. Can't actually find any this size. The supply rails measure 33.6V so 63V seems a bit pointless when 35V caps might do, which would make it much easier to find one that fits, possibly even permitting a larger value. Of course that would leave little safety margin for overvoltage, so I wonder if it is possible to reduce the supply rails slightly by taking a few windings off the transformer secondaries. Anyone see any problems with that? It would also stop the amps running so worryingly hot.
My worry is that one secondary winding might be completely hidden under the other, making it impossible to remove turns off both of them. Without actually taking the transformer to bits, I can't see a way to check this. Any ideas?
My worry is that one secondary winding might be completely hidden under the other, making it impossible to remove turns off both of them. Without actually taking the transformer to bits, I can't see a way to check this. Any ideas?
35V is a bit too close, but 40V version is OK.
Just some of them:
6800uF Aluminum Electrolytic Capacitors - Snap In | Mouser
Sajti
Thanks, but I actually want to reduce the rail voltage for another reason as well. The Iq of the MA50 is only 0.72A in total (2 x 0.36A @ Vcc=34V). This is exactly the same per channel as it is for the A1 in total (2 x 0.7A @ Vcc=20V), yet the rail voltage is double. I think the Iq should be higher in relation to Vcc, and it clearly has the potential to be doubled, as in the A1. The only limitation is the transistor temperature, which operates near its practical limit. So the obvious thing to do is reduce Vcc.
Completely. Might as well build my own amplifier as go down that route.
My point is that excessive rail voltage is wasteful. You can't derive any benefit from it without appropriate current capacity, and I want to maintain class A operation under all listening conditions. Typical minimum speaker impedances are around 4-6 ohms (though can be much lower), which means assuming a peak drive voltage of say 24V would deliver a peak current of perhaps 6A. This would require quiescent current of 3A for class A operation - slightly more than the 0.7A on offer!
The other reason the supply rails are excessive is that input voltages from my DAC are within strict limits, the gain is fixed, and the max voltage levels required are therefore determined solely by how loud I want the volume. I know from earlier experiments that +/-20V output swing is more than sufficient for my needs.
If I could find another MF A1 going cheap, I would swap the transformers to my MA50s.
Not convinced about that. Just looked at one data sheet and it says "35 Vdc (44 Vdc Surge)". But if the mains voltage goes up a bit it will be constantly over 35V at around 100 degrees C. Wouldn't it reduce the already short life (a few thousand hours, i.e. a few hours per day for a couple of years) of the capacitor?
John.
I want to keep costs to a minimum, which is why it makes so much sense to take some turns off the transformer secondaries, to allow non-detrimental use of 35V capacitors of possibly higher capacitance, resulting in slightly cooler operation and therefore longer life / reliability.
John.
In what way? eg. do you have difficult to drive or low efficiency speakers?
Don't mess around with the transformer.
Either use a variac or get a tapped autoformer which is designed to do step the mains down (or up) a little bit. You occassionally see tapped isolation transformers too.
Toroidal transfomer secondary windings are usually wound in parallel to reduce the number of winding operations. It should be straightforward to do as you wish but I don't see it as a wise move when class A amp's biggest drawback is low headroom. We can cope with MF's only partial class A operation but to squish headroom further, rather wrecks the beauty of their advantages - excess heat notwithstanding.
Modern electrolytics are considerably smaller than those only 10-15 years ago, so you should have little trouble getting small enough electrolytics to fit your case whether you use 35,40,50 or 63V types. There is no technical reason not to use higher voltage caps and the choice is down to cost and size. If you can buy higher voltage types cheaply or have them in stock, that's what you fit, unless you are a very large scale manufacturer. All will work at your rail voltages on typical domestic supplies and 105C types are widely available and should be preferred.
If you believe your amplifier reaches 100C internally, a lot worse things would have happened by now including the failure of film caps, PVC sleeving, plastic mouldings and PCBs would be quite browned off. It would also stink as the plastics and paint finish decrepitated, rubber feet melted, fires started due to shorts and nearby furniture virtually blistered or curled up. Be realistic.
Others here have given good advice. Leave the MA50 supply voltages be, just replacing caps and taking advantage of monobloc construction to give more ventilation, not less if you want to make useful improvements. They have lasted well enough until now and that is what professional design is really about.
I've never left them on long enough to see whether the furniture catches fire. But they do start to smell, after a lengthy evening's enjoyment. There are clearly burn marks to the PCB around certain Zener diodes that are used as voltage references. I will change some of their feeder resistors because their values seem to have been based on the presence of a pre-amp section (same PCB as A1) which is not fitted.
You could be right about the voltage headroom being such an asset. And with Iq of only 0.7A you are right to consider it a class AB amplifier. I might do as you suggest and just replace the caps, and the resistors mentioned, and use the best quality heatsink compound I can find when I put the lid back on.
Thanks for your input.
You could be right about the voltage headroom being such an asset. And with Iq of only 0.7A you are right to consider it a class AB amplifier. I might do as you suggest and just replace the caps, and the resistors mentioned, and use the best quality heatsink compound I can find when I put the lid back on.
Thanks for your input.
I'm curious as the A1 seems to be a love it/hate it sort of amplifier (not having heard one myself) - in what ways were the MA-50s better?
As I say, the clarity and transparency. Voices are closer to sounding real. The handling of loud bits is more controlled, especially in voices and piano. And the pace seems faster, more "driven", more engaging, though not necessarily more rhythmical. In terms of timing, they make the A1 sound a bit jaunty and laid-back. The MA50s seem to have even purer tone than the already pure A1, making them even more "musical". I can hear very subtle changes of pitch of which I was unaware before. I'm intrigued to find out whether they will actually improve with the new capacitors!
Any low noise axial fan can be suitable. The attachment is a good question, because I'm not familiar with this MF amplifier.
Sajti
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.