Nominal mains voltage in Australia is 230V, it changed from 240V many years ago.
That said, I regularly see 245V in the ACT.
From Wikipedia:
The nominal voltage in most areas of Australia had been set at 240v in 1926. In 2000, Standards Australia issued a system Standard AS60038, with 230v as the nominal voltage with a +10% to –6% variation at the point of supply, i.e., 253v to 216.2v. A new power quality standard, AS61000.3.100, was released in 2011 that details additional requirements. The new standard stipulates a nominal 230v, and the allowable voltage to the customer's point of supply is, as mentioned, +10% to –6%. However, the preferred operating range is +6% to –2%. (244v-225v).
Last edited:
That might be the case in theory, but as you know the mains is distributed at a nominal 415VAC three phase supply and this has not changed and is usually higher than that, hence the actual single phase voltage is derived from that. The supply authorities have not raced around to change any infrastructure to drop the single phase voltage from 240 to 230VAC. That preferred operating range you mention is rarely achieved especially in rural and regional areas.
@tombo56,
Are you using a 3U 300mm chassis for Bliss? I didn't find any specific reference about the chassis. The case I was planning to use is overkill for an amp that only needs to dissipate 35W per channel. I have a smaller chassis option, but might require the transformer to be in a separate enclosure.
Are you using a 3U 300mm chassis for Bliss? I didn't find any specific reference about the chassis. The case I was planning to use is overkill for an amp that only needs to dissipate 35W per channel. I have a smaller chassis option, but might require the transformer to be in a separate enclosure.
My case is 3U 300mm. It is cramped for build with dual supplies. 400 mm deep case would be better.
There is no such thing as overkill in DIY audio.
Hot heatsinks with smaller case are not a problem but bias current will vary from cold to hot state more than -0.1 A. Again, not a problem.
3U 300mm case has, at 1A bias and 17.5 V supply, some 15 – 16° C rise over ambient temperature.
There is no such thing as overkill in DIY audio.
Hot heatsinks with smaller case are not a problem but bias current will vary from cold to hot state more than -0.1 A. Again, not a problem.
3U 300mm case has, at 1A bias and 17.5 V supply, some 15 – 16° C rise over ambient temperature.
I've been scratching my head on this.
I've been simming an Exicon output stage, and apart from the usual variables like quiescient current (and heat and design bulk, etc) vs distortion, I see a few more subtle questions.
For instance, it looks like the gate-drain capacitances may partly cancel out with the push-pull layout: if you charge the N gate, the gate-drain voltage decreases while for the P channel the gate-drain voltage symmetrically increases, so it looks like the actual drive sees 2 capacitances in series, a much easier load than with just one MOSFET. (Correct me if I'm wrong on this... I plan to do a simulation later.)
As an example, notice how in M2x land, the output stage can be driven by JFETs with
Also, I think the IMD from Miller type effects (non-linear gate-drain capacitance, modulated by large output voltage swings) has to go somewhere. It doesn't just disappear. An op-amp will have very strong drive, but will probably suffer from the dynamically changing load. Its negative feedback may look great with simple HD tests, but the noise floor will probably go up. That's where my thinking is at. Maybe I'm wrong, I don't know.
Last edited:
Take a note of my later distortion measurements after discovering REW bug (it’s fixed now). Distortion at 1W is from 0.00007% to 0.00009%, depending on measured channel sample.
At full power and sine wave signal, actual MOSFET gate currents are 4 uA rms. OS is follower so GS voltage has little changes.
Square wave and achieving slew rate limits requires 10 mA peak currents to gates.
Opamp actually never leaves class A operation as required output for full power is 18 uA rms.
Intermodulation distortion is amazingly low. Check post #54 at this thread and this link for more IMD measurements:
https://www.diyaudio.com/community/threads/opa828-opa2828-vs-opa627.400109/post-7484535
Here is a sample measurement where IMD is mostly below -130 dB at 1 W combined output. Measurement contains used DAC IMD as well.
Hello, my name is Gregory, I'm from Poland. It's my first post ever on this forum. My English is very, very bad, however my son helps me a lot. I have a question to @tombo56 about your amplifier, what would have to be done to modify it to become an headphone amp (because that's the only way I can listen to music), or to create an entire new amplifier project (as good as your's current amp) for headphones. Greetings, Gregory.
Hi Gregory, welcome to diyaudio forum.
This forum contains some fantastic headphone designs. Are you sure none of them suits your requirements? In example:
https://www.diyaudio.com/community/...t-headphone-amp-with-140db-distortion.387033/
Every power amplifier can be used for driving headphones but it is not first or best choice and is certainly much more expensive. In this case, it would require only one resistor change to lower bias to some 100 mA and amplifier boards could be powered from small +/- 15 V supply (nothing fancy). Almost no heatsinks would be required for amplifier.
I would recommend to first check headphone amplifier designs available here.
This forum contains some fantastic headphone designs. Are you sure none of them suits your requirements? In example:
https://www.diyaudio.com/community/...t-headphone-amp-with-140db-distortion.387033/
Every power amplifier can be used for driving headphones but it is not first or best choice and is certainly much more expensive. In this case, it would require only one resistor change to lower bias to some 100 mA and amplifier boards could be powered from small +/- 15 V supply (nothing fancy). Almost no heatsinks would be required for amplifier.
I would recommend to first check headphone amplifier designs available here.
Planning to use this toroidal - https://www.antekinc.com/as-3212-300va-12v-transformer/ with active rectification of LT4320 regulators.
Thanks
Thanks
I was just wondering about @stocktrader200 's tube pre-amp stage that gave it 'life' despite adding 1% HD.
In my case, the latest iteration of an amplifier project has Exicons paired with small signal MOSFETs as compound pairs. Barring any disastrous oscillation that might make it unusable, it reduces HD of the source followers by about 20dB, so that seems promising. What I might do is implement soft clipping for the top 6dB. So for ±10V 'net' that leaves about ±5V for a super clean signal.
The top traces show individual Exicon currents through 0.1 ohm sense resistors, with about 200mA bias. VM1 is 4V out into an 8 ohm simulated speaker load. THD about 2.3E-5%, which is quite impressive considering the strongly visible compression.
That was with a fake battery bias and an NE5534 op-amp driving the lower gate. What's interesting is that if you put something like 1M in series with the battery and a 47n shunt, trimming the voltage somewhat, the distortion doesn't change.
I couldn’t wait any longer, had to fire up the Bliss boards. All went smooth and drama free!😁
Using my DC bench supply set at +/-17V and adjusted the trimpot for 250mV across R16/R17. The current draw is 1.15A per rail with heatsink cold. As it warms up current is slowly decreasing and should settle a bit above 1A. DC offset is nonexistent <1mV.
Getting closer to first sound! Good Times🥳🎄
Using my DC bench supply set at +/-17V and adjusted the trimpot for 250mV across R16/R17. The current draw is 1.15A per rail with heatsink cold. As it warms up current is slowly decreasing and should settle a bit above 1A. DC offset is nonexistent <1mV.
Getting closer to first sound! Good Times🥳🎄
Attachments
We used to use that Semikron heatsink section for the 3 phase full wave AC to DC converters in large DC drives using large hockey puck thyristors - we were talking anything from 300kW to 1.5 mW drives. The stacks were also massively fan cooled. I remember we bought that section in the hundreds at a time. Those were the days - playing around with serious power!! Now I am playing around with minuscule 10w amplifiers.
Hi, thanks for response. I have one more question - is there any general rule that influences the choice of power of the transformer used (VA) to the power provided by the amplifier (W)?Hi Gregory, welcome to diyaudio forum.
This forum contains some fantastic headphone designs. Are you sure none of them suits your requirements? In example:
https://www.diyaudio.com/community/...t-headphone-amp-with-140db-distortion.387033/
Every power amplifier can be used for driving headphones but it is not first or best choice and is certainly much more expensive. In this case, it would require only one resistor change to lower bias to some 100 mA and amplifier boards could be powered from small +/- 15 V supply (nothing fancy). Almost no heatsinks would be required for amplifier.
I would recommend to first check headphone amplifier designs available here.
It depends on amplifier operating class (A, AB are most common). Class A amplifiers consume all the time 2-3 times more power that they can produce at their output. If your intention is still using this as a headphone amplifier, class A bias can be reduced to 0.15 A. So, two amplifier channels would consume about 10 W at +/- 15V supply. Usual rule is that transformer should have 3 times more declared VA rating, so 30 VA transformer would be required.