opc:
modushop.biz These numbers are for the spare heatsinks for the chassis linked before. I don't know if these are good measurements or not. I'm thinking of going for the 2U, 400mm depth chassis to get a little extra cooling compared to the 300 mm version.
Could you sign me up for two extra LME chips and 2 extra pairs of transistors?
That would be rally nice because then i only need to find the common components to build the last boards.
Cristi:
Count me in for one supply, maybe two...
PFC: Yes please, unless the cost is to large...
Voltage: 50 V main rail, Maybe 65-70 V aux so an extra linear reg can be used close to the amp.
Power: 500+ W? (i'm flexible but i prefer to be on the safe side since i will run 3 channels on unknown load impedance).
Size: should fit into a 2U case including margin and shield.
Layout: Mentioned this in an email to cristi, I would like to have all devices requiring (large) cooling on one side of the PCB so they can be easily mounted to an external heat sink.
modushop.biz These numbers are for the spare heatsinks for the chassis linked before. I don't know if these are good measurements or not. I'm thinking of going for the 2U, 400mm depth chassis to get a little extra cooling compared to the 300 mm version.
Could you sign me up for two extra LME chips and 2 extra pairs of transistors?
That would be rally nice because then i only need to find the common components to build the last boards.
Cristi:
Count me in for one supply, maybe two...
PFC: Yes please, unless the cost is to large...
Voltage: 50 V main rail, Maybe 65-70 V aux so an extra linear reg can be used close to the amp.
Power: 500+ W? (i'm flexible but i prefer to be on the safe side since i will run 3 channels on unknown load impedance).
Size: should fit into a 2U case including margin and shield.
Layout: Mentioned this in an email to cristi, I would like to have all devices requiring (large) cooling on one side of the PCB so they can be easily mounted to an external heat sink.
I may just do that, but i already have an abundance of such supplies and plenty of spare parts to just knock another one up to suit if need be, but i don't have ANY high quality switchers, so keen to try one out and it should have the option of supplying both voltages
well, my sinks are 75mm high, so anything 65mm or less would be fine by me. i'm only going for a fairly modest output of ~150-200wpc, so maybe adjustable up to 65 for front end and 50 for the main supply with 6 (8 ohms)-9A(4 ohms) output. i'll leave the calks for current in the front end and main supply to someone who knows more what they are doing than I
80% in AB? Won't we need a higher rail voltage than really needed due to the FETs?
i'm not really the best person to ask, like i said above i'll wait for others more knowledgable than me to chime in; this is my first AB amp, only built Class A before and until recently mostly headphone amps and dacs (many dacs) it will really depend on how much class A operation you bias for. but yeah like Andrew I expected better efficiency than that based on other designs
Last edited:
true that, He mentioned the amp will swing right to the main output rails and its assumed you don't want to listen to sine waves =)
Last edited:
consider where the 10A peak (into a resistive load) comes from. NOT the transformer.
Consider also that your speakers will not usually mimic a resistive load.
The generally recommended transformer for a 400W amplifier is 400 to 800VA. Transformers outside that range will also work. Yes even 1.6kVA but that is not a requirement.
Consider also that your speakers will not usually mimic a resistive load.
The generally recommended transformer for a 400W amplifier is 400 to 800VA. Transformers outside that range will also work. Yes even 1.6kVA but that is not a requirement.
Phase differences makes things worse, 400W apparent power still means 400W real power in DC terms which is the PS we're using. It's not really the current that worries me, I want an amplifier that will be able to cope with high dynamic peaks in voltage and current so want to use the highest rail voltages to acheive that power capability.
I may well be wrong here but the way I see it is; the FETs need that higher rail voltage to achieve a 56V voltage swing for the 400W output into 4ohm, 56V for 400W ~7.1Arms or ~10A peak thus resulting in a PS capable of pumping out ~1.1KW for a single channel?
Am I right there? The only way I'm not is the fact a such high rail voltage isn't needed and so effectively lowers the output required of the PS.
The transformer charges up the smoothing capacitors.
It does this 100 times a second in synchronicity with the peak of the mains waveform of 50Hz.
The charging current only passes through the rectifier for very approximately 10% of the time. The other 90% the transformer passes no current through the rectifier and the rectifier has no current to pass to the smoothing capacitors.
I ask again:
It does this 100 times a second in synchronicity with the peak of the mains waveform of 50Hz.
The charging current only passes through the rectifier for very approximately 10% of the time. The other 90% the transformer passes no current through the rectifier and the rectifier has no current to pass to the smoothing capacitors.
I ask again:
particularly during the 90% when the transformer is not charging the smoothing capacitors.
The emf driving the current through the charging circuit must be higher than the emf stored in the capacitor. When the two em's are exactly the same no current flows in either direction.
When the transformer winding emf is higher than the capacitor's then current starts to flow to overcome the diodic forward voltage of the rectifier.
When that excess emf rises to somewhere around 400 to 500mV the current flow rate suddenly increases from a few micro amps to milliamps and beyond.
Due to the shape of the sine wave, that driving emf is only higher enough to get that higher current flow for a short period near the crest of the sine wave.
When the transformer winding emf is higher than the capacitor's then current starts to flow to overcome the diodic forward voltage of the rectifier.
When that excess emf rises to somewhere around 400 to 500mV the current flow rate suddenly increases from a few micro amps to milliamps and beyond.
Due to the shape of the sine wave, that driving emf is only higher enough to get that higher current flow for a short period near the crest of the sine wave.