No. That are TO3 parts and they don’t fit. Attaching them with even short wires is not something I would recommend for an amplifier with > 1 MHz power bandwidth. Another issue is that they require higher Vds voltage and would not yield the same output power and distortion.
Of course you could add, say a 16-18V Zener to 0v after R7 and R8 (increased to say 100ohms - the zener at a 5-10mA mA will have a dynamic z of about the same, so raw DC impedance doesn't increase much if at all) to ensure safety in case of mains voltage variation on the raw supply. Or any manner of elaboration at that point.Yes, +/- 17 V is target supply voltage. It can't be higher than some +/- 18.5 V due the opamp in front.
Not that it will alter the performance any, the selected opamp has fabulous PSRR and DC gain to compensate chnages in raw supply impedance near DC...
Nice design -and thanks for sharing; I've much sympathy for this kind of approach, I use something not too dissimilar as my headphone amp - good for about 8w into 8ohms, it always surprises people when set to use with sensitive speakers : D
And lowest distortion and best subjective sound of all. If you haven't visited this thread already:Not that it will alter the performance any, the selected opamp has fabulous PSRR and DC gain
https://www.diyaudio.com/community/threads/opa828-opa2828-vs-opa627.400109/
Hope you report it? 🙂I’ve found a bug in REW software distortion measurement
//
For now, only here in the REW thread.Hope you report it? 🙂
//
https://www.diyaudio.com/community/...ion-measurements-with-rew.338511/post-7475209
Yeah, I understand that well, hence – my way.Some would argue but not me that the ultimate amp has no op amps. It that true.
Word of caution.🙂
This amplifier is not suitable for those that look for a ‘musical’ amplifier that makes mediocre recordings sound pleasant by masking details and adding pleasant distortion.
I’ve attached Excel BOM with Mouser part numbers to the first post. Where 10 pcs. price is lower than price of just several required parts, that quantity is used. All components are of very good quality. Resistors are Vishay Beyschlag professional series. My final build was assembled with Dale resistors, just in case. 🙂
But, evaluation build assembled with Beyschlag measures and sounds the same.
If someone likes brownies like me, I will add alternate BOM with exact Dale parts as used in my build. Take a look please, and report any errors if found.
But, evaluation build assembled with Beyschlag measures and sounds the same.
If someone likes brownies like me, I will add alternate BOM with exact Dale parts as used in my build. Take a look please, and report any errors if found.
Max. output is 9 Vrms so even direct output from DAC with integrated volume control will work OK.
Anyway, with most DACs, using digital DAC volume control below -10 dB levels dramatically ruins their performance. I do use preamplifier.
R19 & R20 are Do Not Populate positions. They are here if someone decides later to reduce bias below minimal 0.7 A simply by adding resistors in parallel. They can be inserted in sockets so one can have winter/summer bias setting. 🙂
Anyway, with most DACs, using digital DAC volume control below -10 dB levels dramatically ruins their performance. I do use preamplifier.
R19 & R20 are Do Not Populate positions. They are here if someone decides later to reduce bias below minimal 0.7 A simply by adding resistors in parallel. They can be inserted in sockets so one can have winter/summer bias setting. 🙂
Good idea on the winter/summer time DNP resistors - one could even wire them up with a switch on the amp front panel to change the bias to suit the ambient temp for the seasons.
Thanks for the BOM as well.
Thanks for the BOM as well.
ECW laterals, Q8 and Q9, are not available at Mouser and can be purchased from:
https://gb.profusion.uk/uk/audio/transist/lateral-mosfets
https://www.audiophonics.fr/en/composants-electronique-transistors-puces-c-512.html
https://gb.profusion.uk/uk/audio/transist/lateral-mosfets
https://www.audiophonics.fr/en/composants-electronique-transistors-puces-c-512.html
As it seems that some members will build this amplifier, probably it is not too soon to discuss power supply.
In general, you’d want an oversized PS with this amplifier, like it was recommended for Hiraga’s Le Monstre with 100.000 uF capacitance minimum. If you aim for several W only, any PS will be OK. But, for 20 W/4Ω with low distortion, a very good PS is required. Two 120-150 VA or single 250-300VA transformers will be adequate.
Next point is rectification and capacitors. Main question is what will be actual output voltage with various transformers load regulation, depending on their size, quality and load. My experience is that good quality toroidal transformers, under constant 1/3 load of their nominal rating, will produce 1.25 – 1.27 x AC voltage at first reservoir capacitor, with standard diode bridge rectifier. So, 2x14V transformer is required and it will produce some 17.8 V at first capacitor. Add some drop on CRC chain and we are there.
Active rectifiers will provide some 1.5 V more, compared to diode bridge.
But what if you have 2x15V transformers already? No worries. We can use even 20V supply by just changing two resistor values (R7, R8) in opamp supply. Opamp draws some 5.9 mA and by increasing those resistors we can have required opamp supply voltage. Output stage wouldn’t care even for +/- 100 V supply. So, if you have 18V, resistors should be 150Ω, for 19 V - 330Ω, for 20 V - 470-520Ω.
Before you ask, can we use standard First Watt PS, I wouldn’t go that far without additional Zener protection diodes at opamp supply, and as there is no designated place for them on the PCB, they could be soldered across C4, C5 pins.
Capacitors for PS
No special considerations, any good quality 25 V types are suitable. For first reservoir capacitor, 33-47 mF is enough. PS output capacitors should be largest capacity you can manage, 68 – 100 mF or more. Any ripple at the PS output is not an issue as this amplifier has about 100 dB PSRR in the audio bandwidth. We care for current reserves and load regulation.
In general, you’d want an oversized PS with this amplifier, like it was recommended for Hiraga’s Le Monstre with 100.000 uF capacitance minimum. If you aim for several W only, any PS will be OK. But, for 20 W/4Ω with low distortion, a very good PS is required. Two 120-150 VA or single 250-300VA transformers will be adequate.
Next point is rectification and capacitors. Main question is what will be actual output voltage with various transformers load regulation, depending on their size, quality and load. My experience is that good quality toroidal transformers, under constant 1/3 load of their nominal rating, will produce 1.25 – 1.27 x AC voltage at first reservoir capacitor, with standard diode bridge rectifier. So, 2x14V transformer is required and it will produce some 17.8 V at first capacitor. Add some drop on CRC chain and we are there.
Active rectifiers will provide some 1.5 V more, compared to diode bridge.
But what if you have 2x15V transformers already? No worries. We can use even 20V supply by just changing two resistor values (R7, R8) in opamp supply. Opamp draws some 5.9 mA and by increasing those resistors we can have required opamp supply voltage. Output stage wouldn’t care even for +/- 100 V supply. So, if you have 18V, resistors should be 150Ω, for 19 V - 330Ω, for 20 V - 470-520Ω.
Before you ask, can we use standard First Watt PS, I wouldn’t go that far without additional Zener protection diodes at opamp supply, and as there is no designated place for them on the PCB, they could be soldered across C4, C5 pins.
Capacitors for PS
No special considerations, any good quality 25 V types are suitable. For first reservoir capacitor, 33-47 mF is enough. PS output capacitors should be largest capacity you can manage, 68 – 100 mF or more. Any ripple at the PS output is not an issue as this amplifier has about 100 dB PSRR in the audio bandwidth. We care for current reserves and load regulation.
Wide supply voltage range was not among design goals and it doesn’t make sense with this design. I’m only trying to help with otherer than specified supply voltages.
As you certainly have realized I’ve never used word ‘overengineered’. 🙂
As you certainly have realized I’ve never used word ‘overengineered’. 🙂
I realise now that I mixed up with an other thread - please excuse me.Where did I say that?
//
Hi Tombo56,
Did you fabricate a bracket to support the R21 PS modules on the power supply board? The pics do not show a close up of the details. The two solder points (input/output) are not enough support for this module with the heatsink making it top heavy.
Did you fabricate a bracket to support the R21 PS modules on the power supply board? The pics do not show a close up of the details. The two solder points (input/output) are not enough support for this module with the heatsink making it top heavy.
- Home
- Amplifiers
- Solid State
- Building an ultimate low power class A amplifier – my way