Err ...Suspicious is what JC said, curious us when you pull it apart and can't put it back ..
🙂
🙂
More importantly, who decides when you're a "full audio design engineer"? Can someone be a partial audio design engineer? Or do the rest of us have to continue staring at the ground and drooling?
You're an Full Audio Design engineer when stereophile reviews and test your product. Half bake Audio engineers go to that other magazine...
🙂
🙂
The advice I hoped to give was: IF you do not have a whole lot of experience and education concerning what you are concerned about, in this case, audio amps, perhaps you should believe the schematic given to you, UNLESS you can show otherwise. For the record, Wavebourn, I was 9 years old when I was in the 5th grade, and I was still a 'kid'. I did not take algebra until I as in the 9th grade, and I was 13. Do we have a difference in understanding what 5th grade is?
Yes. We did not have Elementary, Middle, and High schools. We did not choose what to take. We had 10 year program, from Monday to Saturday, all lessons compulsory to get so called "Middle" education. We had no "Multiple Choice" exams, all tests and exams required to think and show understanding of material learned. If somebody failed, he/she must stay one more year in the same grade. Pretty tough, huh?
Then, there were available Profesional-Technical schools and technicums that included last 2 grades of the school, plus 3 grades of professional training. Graduates of them were not called Engineers.
For Engineers there were Institutes, each of them had own specific, for example Medical Institute (for MDs), Institute of Electronics (for MS EEs), and so on. Institutes were connected with scientific institutes and industry enterprises, like Institute of Semiconductor Devices, electronics manufacturing plants, and so on.
Also, there were Universities that prepared scientists and professors.
Quite different education.
Hi,
Class B/AB crossover distortion is not monotonic. This was the angle I commented on. And this kind of distortion (which D. Self comming from Class B calls GM Doubling distortion and I, coming from a Class A angle would call GM halving distortion) is endemic in almost all modern amplifiers.
Ciao T
Class B/AB crossover distortion is not monotonic. This was the angle I commented on. And this kind of distortion (which D. Self comming from Class B calls GM Doubling distortion and I, coming from a Class A angle would call GM halving distortion) is endemic in almost all modern amplifiers.
Ciao T
A. Wayne, I and my associates, some of whom I have known and worked with for about 40 years, got first place in Amplifier, preamplifier, and phonostage, in the latest edition of the TAS. Maybe, just maybe, we know what we are doing?
Hi,
No, with AMR and Diyhifisupply (and I used to modify Shanling Gear for the UK Importer).
Coming out in March 2012. I doubt JA would be happy for me to say much more.
Ciao T
No, with AMR and Diyhifisupply (and I used to modify Shanling Gear for the UK Importer).
Coming out in March 2012. I doubt JA would be happy for me to say much more.
Ciao T
Hi,
The Schematic I have shows them also in the linestage...
Do try it on her.
I had an amp like that (Audio Innovation Alto), except it had already expired. I ended up putting a "gainclone" inside, with caerfully chosen passive parts (Elna Silmic PSU Cap's, Allen Bradly Carbon Comp resistors and an Alps Blue Pot). It does quite well...
The only things I can see to recommend the topology of the Gainclone chip (which is really a generic Self blameless - see attached) is that thermal sensing of the output junctions is almost instant (much better than the "thermaltrack" transistors for example), that all BJT Bases are buffered by followers and that the output stage forms part of the miller loop and is fitted with high value emitter resistors, which means the handover between the halves will be better managed, IF the biasing is right, which is assured by chip design.
Otherwise there is really not much unusual there. I can also see easily why this chip reacts so strongly to power supply design (hint - it's the miller compensation loop), my one is CRC filtered, but well designed regulated PSU's do better...
Ciao T
The Schematic I have shows them also in the linestage...
Do try it on her.
I had an amp like that (Audio Innovation Alto), except it had already expired. I ended up putting a "gainclone" inside, with caerfully chosen passive parts (Elna Silmic PSU Cap's, Allen Bradly Carbon Comp resistors and an Alps Blue Pot). It does quite well...
The only things I can see to recommend the topology of the Gainclone chip (which is really a generic Self blameless - see attached) is that thermal sensing of the output junctions is almost instant (much better than the "thermaltrack" transistors for example), that all BJT Bases are buffered by followers and that the output stage forms part of the miller loop and is fitted with high value emitter resistors, which means the handover between the halves will be better managed, IF the biasing is right, which is assured by chip design.
Otherwise there is really not much unusual there. I can also see easily why this chip reacts so strongly to power supply design (hint - it's the miller compensation loop), my one is CRC filtered, but well designed regulated PSU's do better...
Ciao T
Attachments
Hi,
Not really.
The 33K is there so the global feedback does not "open up" when the fuse blows. 33K seems excessive, a 1K value should be better, but let's not blow that fuse in the first place, shall we?
The schematic I have lacks these, but it has a 120 ohm resistor (mislabeled 120K) to the headphone sockets. I guess proofing service manual schematics was no Rotel's strong suit in these days.
Ciao T
Not really.
The 33K is there so the global feedback does not "open up" when the fuse blows. 33K seems excessive, a 1K value should be better, but let's not blow that fuse in the first place, shall we?
The schematic I have lacks these, but it has a 120 ohm resistor (mislabeled 120K) to the headphone sockets. I guess proofing service manual schematics was no Rotel's strong suit in these days.
Ciao T
Hi,
I guess you could call it that. It is hard to describe, but if this quality is present I am willing to overhear many other flaws, if this quality is not present I immediately become hyper critical of certain technical aspects of the sound quality, the Bass's, the Trebles, this little bit of grain and so on...
And yes, the TDA1541 has it, some Delta-Sigma DAC's have it, much tube gear has it, some transistor gear has it, even some VERY HIGH feedback transistor gear has it, but loop feedback less tube gear seems to have it most reliably, as do Multibit DAC's.
Ciao T
PS, for disclosure, my current personal system is:
PC Source (1)
TDA1541 Non-Oversampling DAC (2)
Passive Pre (3)
EL34 Push Pull Amp without looped feedback (4)
Average 3-Way speakers (5)
Silver and Silver/Copper Signal Cables (6)
AMR Prototype Mains Cables (7)
Ikea Lack-Rack (8)
appx. 100m^3 Room (9)
1) PC-Source -
dedicated build, fanless, Atom 330 with touchscreen,
SPDIF from motherboard audio chipset (VIA Vinyl Codec, Bitperfect up to 96KHz),
Fidelized Windows 7 with Mediaportal and Pureaudio ASIO Music Player, all music as FLAC, also acts as AV Source
2) TDA1541 DAC -
active transformer coupled CS8414 frontend, Wildmonkeysects PLL Filter, SMD decoupling using 0603 Cap's directly on pins, 1206 on the SMD to DIP Adaptor
Pins
Floating, common mode suppressed supplies with extra choke filtering and CCS / local shunt regulator arrangements with "crazy" regulator bypassing including Sanyo ultra Low Z 2,200uF/6.3V caps, Sanyo Os-Con and 10uF 1206 SMD Caps
Tube Analog Stage without feedback (GEC five star 6072A), silver capacitors including output coupling cap bypass plus AMR Tinfoil output coupling cap, SINC Rolloff correction with silver cap and copper choke, CLC filtered PSU with added RCRCRC filtering and Film PSU Caps in the final RC filter stage
silver wired with air-dielectric, CMC RCA's
3) Passive Pre -
Grayhill selector switch, 10K Noble Potentiometer,
silver wired with air-dielectric,
uses a collection of RCA's including Cardas, AN-Silver CMC Gold & Silver (what was lying around basically, after testing all these I setteled on the goldplated pure copper CMC as "good enough" and affordable).
4) Push Pull Amp -
output stage with local feedback uses cathode & screengrid feedback only, Antique Sound Labs output Transformers, Shugang EL34B (basically 6CA7 clones, hence not pentodes but beam tetrodes), automatic fixed bias
concertina phase-splitter running a russian black plate octal 5687 analog with 10K loads,
E83CC (Mullard Box Plate CV4004) parallel voltage gain stage with 80uF/35V Film Cap Cathode Bypass (so no local feedback),
silver coupling caps and silver wired with air-dielectric, goldplated CMC RCA's & binding posts,
around 30W output and a dumping factor of a little over 2
5) Speaker -
10" ceramic coated Alu Cone Woofer, 5" ceramic coated Alu Cone Wideband as Midrange, HiVi circular magnetostat as Tweeter
20mm stone fillet lined MDF Cabinet, Avalon Style "stealth" front panel, polyfilling applied internally
first order series crossover at 300Hz/3KHz, no baffle step correction (as it would lead to non-flat power response), Copper Foil inductors, AMR Film Cap's, single resistor (HF attenuator/EQ) is multiple parallel carbon composite ones, silver wired with air-dielectric, CMC Binding posts
mostly flat 6 Ohm impedance above 120Hz, +/- < 0.5 Ohm 250Hz - 10KHz, dropping above 10KHz to 4 Ohm (this is how it came out, it was not actually a design goal, but flat response also required flat impedance, funny that), modest LF impedance peaks from a damped MLTL bass system - so friendly to work with low dumping factor amps,
around 87dB/2.83V/1m and LF cutoff in the upper 20Hz region in room
6) Cables -
SPDIF - Belden Plenum Video with WBT Nextgen Plugs
LINE - Silver Special cables - that is 5 * 0.6mm bare silver in air cells for ground, 1 * 0.2mm enameled Audio Note Silver for Signal, Metalised mylar weave for screen (one side connected only), Neutrick Pro RCA Plugs
SPEAKER - 80 Conductor SCSI 3 Ribbon, single core silverplated copper conductor in FEP Teflon, interleaved connection so approximately 2.5 Ohm CI, single piece pure copper gold plated spade connectors
7) Mains Cables -
use a variety of different cables and plugs by Oyade, Yarbo and others, power distribution strip is re-wired with solid core wire and has a number of RC circuits and capacitors fitted across life & neutral
8) Ikea-LackRack -
Three Layer Rack build from Ikea Lack Tables, 2-component epoxy used during assembly to create a rigid structure, on adjustable spikes with brass cups on the floor
9) Room - we just moved here a few weeks back
28m^2 open plan living space, varying hight ceiling between 3.2m and 4.5m high,
several well stuffed Sofas, a carpet in the seating area, no other room treatment, but the various paintings (the whole place has a faint art galery feeling to it the way we designed it) will soon get Basotect backing
The single plane widows in the somewhat loose doors forming one wall are a great plate bass absorber, around 6m^2 worth
Speakers are placed around halve-way along the long axis of the room, generally not my preferred way of doing things and hence strong toe-in. Sofa's on the sidewalls within 1m of the speakers front help absorb early reflections somewhat.
Doing a handclap test reveals no slap echo or reverb tail, bass is tight and controlled, still did not get around to do an RT60 measurement, but from hearing the room it should be okay
I guess you could call it that. It is hard to describe, but if this quality is present I am willing to overhear many other flaws, if this quality is not present I immediately become hyper critical of certain technical aspects of the sound quality, the Bass's, the Trebles, this little bit of grain and so on...
And yes, the TDA1541 has it, some Delta-Sigma DAC's have it, much tube gear has it, some transistor gear has it, even some VERY HIGH feedback transistor gear has it, but loop feedback less tube gear seems to have it most reliably, as do Multibit DAC's.
Ciao T
PS, for disclosure, my current personal system is:
PC Source (1)
TDA1541 Non-Oversampling DAC (2)
Passive Pre (3)
EL34 Push Pull Amp without looped feedback (4)
Average 3-Way speakers (5)
Silver and Silver/Copper Signal Cables (6)
AMR Prototype Mains Cables (7)
Ikea Lack-Rack (8)
appx. 100m^3 Room (9)
1) PC-Source -
dedicated build, fanless, Atom 330 with touchscreen,
SPDIF from motherboard audio chipset (VIA Vinyl Codec, Bitperfect up to 96KHz),
Fidelized Windows 7 with Mediaportal and Pureaudio ASIO Music Player, all music as FLAC, also acts as AV Source
2) TDA1541 DAC -
active transformer coupled CS8414 frontend, Wildmonkeysects PLL Filter, SMD decoupling using 0603 Cap's directly on pins, 1206 on the SMD to DIP Adaptor
Pins
Floating, common mode suppressed supplies with extra choke filtering and CCS / local shunt regulator arrangements with "crazy" regulator bypassing including Sanyo ultra Low Z 2,200uF/6.3V caps, Sanyo Os-Con and 10uF 1206 SMD Caps
Tube Analog Stage without feedback (GEC five star 6072A), silver capacitors including output coupling cap bypass plus AMR Tinfoil output coupling cap, SINC Rolloff correction with silver cap and copper choke, CLC filtered PSU with added RCRCRC filtering and Film PSU Caps in the final RC filter stage
silver wired with air-dielectric, CMC RCA's
3) Passive Pre -
Grayhill selector switch, 10K Noble Potentiometer,
silver wired with air-dielectric,
uses a collection of RCA's including Cardas, AN-Silver CMC Gold & Silver (what was lying around basically, after testing all these I setteled on the goldplated pure copper CMC as "good enough" and affordable).
4) Push Pull Amp -
output stage with local feedback uses cathode & screengrid feedback only, Antique Sound Labs output Transformers, Shugang EL34B (basically 6CA7 clones, hence not pentodes but beam tetrodes), automatic fixed bias
concertina phase-splitter running a russian black plate octal 5687 analog with 10K loads,
E83CC (Mullard Box Plate CV4004) parallel voltage gain stage with 80uF/35V Film Cap Cathode Bypass (so no local feedback),
silver coupling caps and silver wired with air-dielectric, goldplated CMC RCA's & binding posts,
around 30W output and a dumping factor of a little over 2
5) Speaker -
10" ceramic coated Alu Cone Woofer, 5" ceramic coated Alu Cone Wideband as Midrange, HiVi circular magnetostat as Tweeter
20mm stone fillet lined MDF Cabinet, Avalon Style "stealth" front panel, polyfilling applied internally
first order series crossover at 300Hz/3KHz, no baffle step correction (as it would lead to non-flat power response), Copper Foil inductors, AMR Film Cap's, single resistor (HF attenuator/EQ) is multiple parallel carbon composite ones, silver wired with air-dielectric, CMC Binding posts
mostly flat 6 Ohm impedance above 120Hz, +/- < 0.5 Ohm 250Hz - 10KHz, dropping above 10KHz to 4 Ohm (this is how it came out, it was not actually a design goal, but flat response also required flat impedance, funny that), modest LF impedance peaks from a damped MLTL bass system - so friendly to work with low dumping factor amps,
around 87dB/2.83V/1m and LF cutoff in the upper 20Hz region in room
6) Cables -
SPDIF - Belden Plenum Video with WBT Nextgen Plugs
LINE - Silver Special cables - that is 5 * 0.6mm bare silver in air cells for ground, 1 * 0.2mm enameled Audio Note Silver for Signal, Metalised mylar weave for screen (one side connected only), Neutrick Pro RCA Plugs
SPEAKER - 80 Conductor SCSI 3 Ribbon, single core silverplated copper conductor in FEP Teflon, interleaved connection so approximately 2.5 Ohm CI, single piece pure copper gold plated spade connectors
7) Mains Cables -
use a variety of different cables and plugs by Oyade, Yarbo and others, power distribution strip is re-wired with solid core wire and has a number of RC circuits and capacitors fitted across life & neutral
8) Ikea-LackRack -
Three Layer Rack build from Ikea Lack Tables, 2-component epoxy used during assembly to create a rigid structure, on adjustable spikes with brass cups on the floor
9) Room - we just moved here a few weeks back
28m^2 open plan living space, varying hight ceiling between 3.2m and 4.5m high,
several well stuffed Sofas, a carpet in the seating area, no other room treatment, but the various paintings (the whole place has a faint art galery feeling to it the way we designed it) will soon get Basotect backing
The single plane widows in the somewhat loose doors forming one wall are a great plate bass absorber, around 6m^2 worth
Speakers are placed around halve-way along the long axis of the room, generally not my preferred way of doing things and hence strong toe-in. Sofa's on the sidewalls within 1m of the speakers front help absorb early reflections somewhat.
Doing a handclap test reveals no slap echo or reverb tail, bass is tight and controlled, still did not get around to do an RT60 measurement, but from hearing the room it should be okay
Hi,
Pioneer PD2026, PD2028, PD2029
Cirrus CS4328
Nippon Precision SM5872
These I know for sure. Standard implementations of these (except Cirrus) tend to be grotty.
Give a Pioneer PD202X DAC a low jitter data source, a decent 5V regulator and a decent analogue stage and it can perform quite outstanding, same for the SM5872, it is very difficult to apply well.
The CS4328 is perhaps the easiest, it is direct line out, also needs a very low jitter source.
Ciao T
Pioneer PD2026, PD2028, PD2029
Cirrus CS4328
Nippon Precision SM5872
These I know for sure. Standard implementations of these (except Cirrus) tend to be grotty.
Give a Pioneer PD202X DAC a low jitter data source, a decent 5V regulator and a decent analogue stage and it can perform quite outstanding, same for the SM5872, it is very difficult to apply well.
The CS4328 is perhaps the easiest, it is direct line out, also needs a very low jitter source.
Ciao T
- Status
- Not open for further replies.