Calculate the Vdrop across the 317 for these two worst case operational conditions: 1.) Highest mains supply voltage combined with lowest output current and 2.) Lowest mains supply voltage combined with highest output current.
There is a further worst case operation: 3.) Highest mains supply voltage combined with highest output current. This one determines the heatsink size.
There is a further worst case operation: 3.) Highest mains supply voltage combined with highest output current. This one determines the heatsink size.
An update:
post1437 links to David Eather's paper on temperature derated SOAR into reactive loads still works.
post1437 links to David Eather's paper on temperature derated SOAR into reactive loads still works.
is the amp marginally better than the 5200 1943 lme49810? as I see alot of people appreciate the lateral mosfet amplifiers and exicon being very good in the case
I don't have golden ears. I have a LME49810 with 5 pairs of 5200 and 1943 and Owens boards. The version of 49810 that I bought did not test well by some here at Diyaudio, pansonhk I think. Owens version has presented his data and it is great. There may be some suttle sound differences between the boards, but are the differences attributed to the fets vs bipolars, the power supplies, regulated front end vs unregulated front end, wiring, layout, choice of capactors, bias levels, types of power cords etc etc etc.
Hey guys,
please could anybody tell me, how to get the mosfets?
Are those ACD101NDD/ACD103PDD replaced by other mosfets? These guys are not findable although I've played Google to the end!
By the way... did anybody try other lateral mosfets with success??
Thanks,
Stammheim
please could anybody tell me, how to get the mosfets?
Are those ACD101NDD/ACD103PDD replaced by other mosfets? These guys are not findable although I've played Google to the end!
By the way... did anybody try other lateral mosfets with success??
Thanks,
Stammheim
OK, I've found the information, that ALF16N20W and ALF16P20W are to be used...
Hello guys,
Owen recommends to use a little higher voltage for the IC then for the output stage. As I have a +-65V SMPS which I want to use, I'd need about 75V for the LME. As I would have to take a trafo with 55V and down-regulate the voltage or take a 18V trafo with voltage multiplier I'm asking myself what the differences between using one voltage and using different voltages are.
If somebody has a got hint for a 75V supply... you're welcome 😉
Best regards,
Stammheim
Owen recommends to use a little higher voltage for the IC then for the output stage. As I have a +-65V SMPS which I want to use, I'd need about 75V for the LME. As I would have to take a trafo with 55V and down-regulate the voltage or take a 18V trafo with voltage multiplier I'm asking myself what the differences between using one voltage and using different voltages are.
If somebody has a got hint for a 75V supply... you're welcome 😉
Best regards,
Stammheim
headroom,dynamic range, output voltage/power
Small transformer,diode bridge,filter,regulators (LM317/337) floating regulator is simple and effective.
Need a transformer with a secondary around 120VAC center tapped.
Or you can boost the o/p from the SMPS by using a dual secondary transformer arrangement. Only need a ~10VA transformer
something like this 595-1025-ND
have fun
Rick
Hello Rick,
there's no problem to realize a linear supply, but the amount of difference in reality was not clear to me.
How to use the transformer to use the SMPS's voltage sounds interesting, but is not clear to me. Could you explain me, how it works??
Another possibility:
I have the Hypex SMPS3K with 85V for my bass-amps. Could I use it parallel in combination with the LM337/317 to power the LME? Theoretically of course, but are there any reasons, why I shouldn't?
Would be much cheaper and nearly without effort.
Thanks,
Stammheim
there's no problem to realize a linear supply, but the amount of difference in reality was not clear to me.
How to use the transformer to use the SMPS's voltage sounds interesting, but is not clear to me. Could you explain me, how it works??
Another possibility:
I have the Hypex SMPS3K with 85V for my bass-amps. Could I use it parallel in combination with the LM337/317 to power the LME? Theoretically of course, but are there any reasons, why I shouldn't?
Would be much cheaper and nearly without effort.
Thanks,
Stammheim
Hi Stammheim,
The higher supply for the LME front end really only buys you more headroom, so you need to ask yourself if you really need the additional power output, or if you can live without it.
If you don't need the additional power, then running everything off the same supply isn't a problem. You'll lose about 4-5V of output swing capability, but if you don't need it, you won't miss it 🙂
The only other possible advantage is that you can run lower noise supplies for the input stage, which does provide a tangible benefit in terms of output noise.
Regards,
Owen
The higher supply for the LME front end really only buys you more headroom, so you need to ask yourself if you really need the additional power output, or if you can live without it.
If you don't need the additional power, then running everything off the same supply isn't a problem. You'll lose about 4-5V of output swing capability, but if you don't need it, you won't miss it 🙂
The only other possible advantage is that you can run lower noise supplies for the input stage, which does provide a tangible benefit in terms of output noise.
Regards,
Owen
Hello Owen,
the question is, how noisy the voltage is, when I get it from the SMPS and decrease it to 75V by the LM's...
So I wouldn't do it because I need more power, but to have a better overall-result.
Thanks,
Stammheim
the question is, how noisy the voltage is, when I get it from the SMPS and decrease it to 75V by the LM's...
So I wouldn't do it because I need more power, but to have a better overall-result.
Thanks,
Stammheim
If you take one of the transformer dual secondary windings and connect one end (dot or non-dot) to the + or - o/p of the SMPS, it will float the other end of the transformer secondary winding (dot or non-dot) above the SMPS output by the voltage imposed by the secondary winding. Then you can either just rectify/filter the ac riding on the dc and have a unregulated supply or add a regulator to make it a floating regulated boosted o/p.
You need two of them since you need both + and - boosted supplies for the LME49830. Assumption is all these are floating supplies.
Sounds interesting, but I don't understand the principle. So do I connect the whole winding to + and GND? Do I get out the power fron the other secondary winding? How do I calculate which transformer to use for which voltage?
Generally:
What about using the SMPS3k with the regulators? Would that be OK? Unnoisy enough?
Best regards,
Stammheim
Generally:
What about using the SMPS3k with the regulators? Would that be OK? Unnoisy enough?
Best regards,
Stammheim
The motivation for higher rails on the LME chip vs the output stage is to allow the output stage the full voltage swing ... stops voltage sag from the output stage on large transients impacting the input stage.
Like Owen said ... how close to the full output voltage swing are you planning on driving these? With voltage rails that high I wouldn't worry about it. For more moderately powered amps then a separate supply to the LME chip makes sense.
PSRR for the LME chip is pretty solid so for high power amp I wouldn't worry too much about the noise on its supply.
Cheers,
Chris
Like Owen said ... how close to the full output voltage swing are you planning on driving these? With voltage rails that high I wouldn't worry about it. For more moderately powered amps then a separate supply to the LME chip makes sense.
PSRR for the LME chip is pretty solid so for high power amp I wouldn't worry too much about the noise on its supply.
Cheers,
Chris
Fine, so I'll downregulate the +-85V to +-75V and try using an inductor. Any hints what I could use? Current compensated inductor?
Thanks for help! Although reading and learning it's sometimes hard to find a solution for an electronical problem 🙂
Sometimes to better isolate the early stage on the cheap, manufacturers just insert a diode and a resistor into each rail feeding the early stage and use for each rail near the early stage a 100uF (maybe also with // 0.1uF) cap. The Hafler 200 amplifiers had that. If I recall correctly the early stage on that still dropped a volt of more though with each large kick drum beat, etc, how bad would it have been without the diode and cap!!
I think there might have been an issue in translation ... I'm suggesting to just connect the input stage to the same power supply as the output stage.
1.) connect the output stage and the driver chip to the same supply.
overdrive the input and the output may not go into saturation and should take a short time to return to music making.
2.) separate the output stage and the driver chip using two same voltage supplies.
overdrive the input and the output supply sags a little/lot. The output stage goes in mild saturation and thus takes a bit longer to return to music making.
3.) supply the driver chip with a high voltage and the output stage with a separate lower voltage.
Drive the input such that the driver does not overload and the output stages severely overloads and you have the longest time delay before the output returns to music making.
Of the three options I think that the third will sound the worst in loud music reproduction.
There are techniques to reduce the saturation of overdriven stages, but none are implemented in this PCB.
During quiet and moderate music and "sound effects" episodes, it makes no difference what voltages are used for the supplies.
overdrive the input and the output may not go into saturation and should take a short time to return to music making.
2.) separate the output stage and the driver chip using two same voltage supplies.
overdrive the input and the output supply sags a little/lot. The output stage goes in mild saturation and thus takes a bit longer to return to music making.
3.) supply the driver chip with a high voltage and the output stage with a separate lower voltage.
Drive the input such that the driver does not overload and the output stages severely overloads and you have the longest time delay before the output returns to music making.
Of the three options I think that the third will sound the worst in loud music reproduction.
There are techniques to reduce the saturation of overdriven stages, but none are implemented in this PCB.
During quiet and moderate music and "sound effects" episodes, it makes no difference what voltages are used for the supplies.
So as the target is not to have the loudest but a truly clear soundind system there is no reason why not to take one voltage for both, do I got it right?
What I've always asked myself: Why do most other LME49830-amps have a V-suggestion for 40-50V for the IC and always more for the output stage?
@IanAS
That's what I also have been thinking about to reduce influences, when two or more amps are connected to one power supply. Did anybody find out, if isolating every amp by a diode and a / some caps have any advantages?
Wish you guys a nice weekend!
Stammheim
What I've always asked myself: Why do most other LME49830-amps have a V-suggestion for 40-50V for the IC and always more for the output stage?
@IanAS
That's what I also have been thinking about to reduce influences, when two or more amps are connected to one power supply. Did anybody find out, if isolating every amp by a diode and a / some caps have any advantages?
Wish you guys a nice weekend!
Stammheim