He guys - this is my contribution to UCD 8K.
8KW EIAJ into 8 Ohms.
Amp is basically working in LTSPICE.
BUT (a big BUT): It takes hours and even days to simulate even the first 100ms in order to reach steady state - even on dual 8 core XEON or i7-8700K.
Amp can run on one single 400V rail and will supply its own housekeeping voltage by the use of coupled L5 winding once it is running. To start up, U17 will charge the housekeeping capacitor C64 in linear mode. Unfortunetely it takes - you guessed it - litterally ages to simulate even the linear charge up of the cap. To simulate a few minutes of performance, it takes about the time that was used to build the Notre Dame cathedral - the guys that started it did not live until it's completion. I have no idea why LTSPICE is THAT slow.
BTW: Input is similar to THAT corporations IC. The input will even have high CMRR when driven with imbalanced source impedance.
See how I managed to use a SINGLE MONO POTENTIOMETER as Attenuator despite the fact that the input is balanced.
L1, L2 - differentially coupled main choke
L3, L4 - small common mode choke
Current limit: U19 will just disable all gates, once a certain threshold is reached and try again later (may not be the "best sounding" current limit)
Open action items:
->Power supply 1 x 400VDC
->Get the amp running IRL
8KW EIAJ into 8 Ohms.
Amp is basically working in LTSPICE.
BUT (a big BUT): It takes hours and even days to simulate even the first 100ms in order to reach steady state - even on dual 8 core XEON or i7-8700K.
Amp can run on one single 400V rail and will supply its own housekeeping voltage by the use of coupled L5 winding once it is running. To start up, U17 will charge the housekeeping capacitor C64 in linear mode. Unfortunetely it takes - you guessed it - litterally ages to simulate even the linear charge up of the cap. To simulate a few minutes of performance, it takes about the time that was used to build the Notre Dame cathedral - the guys that started it did not live until it's completion. I have no idea why LTSPICE is THAT slow.
BTW: Input is similar to THAT corporations IC. The input will even have high CMRR when driven with imbalanced source impedance.
See how I managed to use a SINGLE MONO POTENTIOMETER as Attenuator despite the fact that the input is balanced.
L1, L2 - differentially coupled main choke
L3, L4 - small common mode choke
Current limit: U19 will just disable all gates, once a certain threshold is reached and try again later (may not be the "best sounding" current limit)
Open action items:
->Power supply 1 x 400VDC
->Get the amp running IRL
Attachments
Umm-mm -- looks like your SCT3022's are mis-wired. Pin 2 is the Drain on my data sheet, not the Gate.
Also kind of wondering if your power is bigger and better than mine. I certainly wouldn't be able to scare 35 amps at 400 volts out of any of my native residential wiring. Something like the arc welder wiring would do it.
Looks like fun, Regards,
Rick
Also kind of wondering if your power is bigger and better than mine. I certainly wouldn't be able to scare 35 amps at 400 volts out of any of my native residential wiring. Something like the arc welder wiring would do it.
Looks like fun, Regards,
Rick
Last edited:
About the "power is bigger" (??)
-> A stereo model of this amp could very well be supplied by a three phase PSU with PFC on high line.
e.g.: 3 x 230 / 400VAC / 16A -> 3 individual PFCs, each connected L - N -> 3 individual DC / DC converters, connected in parallel or series at the secondary side. did I say anything about 35 Amps? No.
Supply the 400V DC rail with up to 10 Amps per amp channel plus a little short term overload capability / good energy storage and all is well.
-> A stereo model of this amp could very well be supplied by a three phase PSU with PFC on high line.
e.g.: 3 x 230 / 400VAC / 16A -> 3 individual PFCs, each connected L - N -> 3 individual DC / DC converters, connected in parallel or series at the secondary side. did I say anything about 35 Amps? No.
Supply the 400V DC rail with up to 10 Amps per amp channel plus a little short term overload capability / good energy storage and all is well.
Sorry, just doing the arithmetic -- 270 V rms across an 8 ohm load is 33.75 Amps rms.
In any case, 4 kW of juice is not going to deliver 8 kW of output.
Glad you have access to 3-phase. The closest I've come to it in any residential setting is my current pole out back. All 3-4700V lines arrive at my pole, but 2 of them stop there.
Thanks for the model reference.
Regards,
Rick
In any case, 4 kW of juice is not going to deliver 8 kW of output.
Glad you have access to 3-phase. The closest I've come to it in any residential setting is my current pole out back. All 3-4700V lines arrive at my pole, but 2 of them stop there.
Thanks for the model reference.
Regards,
Rick
4KW is not going to produce 8KW of output? Didn't I clearly state
that 8KW is EIAJ rating, which means burst RMS.
A few half waves of 1kHz at 8KW, after that lower power....
Long term RMS power will be far lower and doesn't need to be that high due to the typical power - time distribution of music.
that 8KW is EIAJ rating, which means burst RMS.
A few half waves of 1kHz at 8KW, after that lower power....
Long term RMS power will be far lower and doesn't need to be that high due to the typical power - time distribution of music.
BTW: Who said 4KW? A proper PFC can boost almost 3500W into the 400V rail, supplied only by ONE 16A-Phase at 230V. Multiply that by the efficiency of the DC / DC converter - that is the power what you get at the secondary side. PER PHASE!
For three phases that power can be multiplied - you guessed it - by 3.
So if the amp - heavily beaten by compressed signal - would consume max 4KW per channel on average (typically FAR less!), I could EASILY put two channels on one 10KW PSU.
For three phases that power can be multiplied - you guessed it - by 3.
So if the amp - heavily beaten by compressed signal - would consume max 4KW per channel on average (typically FAR less!), I could EASILY put two channels on one 10KW PSU.
The 4 kW came from the 400 V, 10 A supply spec.
Apologies -- again. Made a mental note to look up 'EIAJ' rating, but forgot (before posting that). I don't personally trade much in that type of rating, but if it is useful to you, that's perfectly fine.
O.K., so normal residential wiring in your neighborhood is 230V -- again, good on ya'! Standard residential wiring here is 120V, 20A; for 240V you have to resort to wiring that's normally reserved for kitchen ranges (stove/oven), air conditioning, 4 hp and up shop air compressors, stick welders, etc.
Rick
Apologies -- again. Made a mental note to look up 'EIAJ' rating, but forgot (before posting that). I don't personally trade much in that type of rating, but if it is useful to you, that's perfectly fine.
O.K., so normal residential wiring in your neighborhood is 230V -- again, good on ya'! Standard residential wiring here is 120V, 20A; for 240V you have to resort to wiring that's normally reserved for kitchen ranges (stove/oven), air conditioning, 4 hp and up shop air compressors, stick welders, etc.
Rick
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.