-
Welcome to ASR. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!
AIYIMA T20 Tube Pre-amplifier Review
- Thread starter amirm
- Start date
- Joined
- Dec 12, 2019
- Messages
- 16,460
- Likes
- 37,372
why?
Especially if the loudspeaker is a fairly sensitive one, such as the Klipsch product in the photo posted above.
I'm just used to having a giant black brick called an AVR.
- Joined
- Dec 12, 2019
- Messages
- 16,460
- Likes
- 37,372
And my daily driver amplifier is capable of about 3 watts per channel at 1% THD.
Used with loudspeakers are roughly 104 dB SPL at 1 meter with one watt input, it's perfectly adequate for my needs.
Admittedly, the amplifier's not terribly small, but it's not AVR-sized (or massed), either.
Much of the size and mass (and heat, paradoxically) of a modern AVR has little to do with the amplifier part.
- Joined
- Dec 12, 2019
- Messages
- 16,460
- Likes
- 37,372
That set up used to be even more streamlined. The Fosi BT20a Pro was my “gateway drug” into playing around with small footprint gear, and audio gear again in general. Damn thing
vilgerfortz
New Member
- Joined
- Feb 17, 2026
- Messages
- 1
- Likes
- 1
Hi everyone,
I’m planning some upgrades to my Aiyima T20 tube preamp and Aiyima A20 stereo power amp and want to make sure I understand the power requirements properly before I pull the trigger on a better supply.
Here’s my planned setup:
T20 modifications:
Right now the plan is to keep the stock 32V/5A PSU on the A20 unless there’s reason to upgrade for the op-amp change.
Questions:
Thanks!
I’m planning some upgrades to my Aiyima T20 tube preamp and Aiyima A20 stereo power amp and want to make sure I understand the power requirements properly before I pull the trigger on a better supply.
Here’s my planned setup:
T20 modifications:
- Stock tubes → PSVANE Horizon 12AX7-AT
- Stock op-amps → Burson V7 Classic Dual discrete op-amps
- Stock op-amps → Burson V7 Classic Dual discrete op-amps
- T20 original PSU requirement is DC 12V / 2A (the stock PSU that comes with the unit).
- A20 original PSU requirement is DC 32V / 5A (the stock PSU that comes with the unit).
Right now the plan is to keep the stock 32V/5A PSU on the A20 unless there’s reason to upgrade for the op-amp change.
Questions:
- Does anyone know (or can measure) the actual operating current draw of a stock T20 (with stock op-amps)?
- When using Burson V7 Classic Dual on the T20, what kind of current draw increase should be expected?
- Is 12V @ 3A (e.g., Sbooster BOTW P&P ECO MKII) likely sufficient for the T20 with these upgrades?
- For the A20, does replacing the op-amps with Burson V7 Classic Dual typically drive current requirements above the original 32V @ 5A stock supply?
- Any real-world experience with this exact combo?
Thanks!
RandomEar
Addicted to Fun and Learning
- Joined
- Feb 14, 2022
- Messages
- 757
- Likes
- 1,914
Changing op amps or the PSU will not alter the sound of your devices at all, excluding a couple of edge cases like a severely underpowered PSU or an op amp which starts to oscillate in the circuit and risks blowing your speakers. Op amp rolling and power supply swaps or mods are a total waste of your hard earned money.
Switching tubes could potentially make a difference in some cases. If you would like to experiment with anything, this is the mod to try.
I have v7 on mine with ifi elite and its totally fine....
Despite being discrete the total power draw is around 10-15w max, so you will have no issue at all.
Sbooster maybe a bit overkill as costs more the the t20 itself (2x?)...
I also have one and is wonderful product (not hearable improvement) powering rme dac
extraclassic
Member
- Joined
- Sep 14, 2020
- Messages
- 5
- Likes
- 4
Hi everyone, I am not very new to ASR but have not posted, however, I have learned a lot through Amir's posts and the community feedback - this forum is a great resource, but you already know that. For those who have helped me along the way without knowing, thank you. I am presenting another view of the popular Aiyima T20, this time with mods. I posted this mod overview on DIY Audio two days ago and am sharing here due to the healthy interest in this preamp. I felt like this would be a good starter post for the forum and am happy to contribute.
TL;DR: GOTO 10th paragraph. BASICally, skip to the end conclusion.
For the record, I am not an electrical engineer, nor do I have the equipment to make noise measurements – I am just a somewhat experienced hobbyist that likes to assemble predesigned modular or repurposed audio boards and guts to make Frankenstein-esque audio components at glacial pace. Too many hobbies. Hobbies within hobbies. The perspective of this review is that the subjective testing of the T20 is performed by replacing a more expensive preamp, and the rest of the component chain is up in a different league, performance/cost wise. This is intended so I could reveal as much as I could from the mod changes.
In the following passages, there is one rule: I will express many subjective opinions. I do recognize the belief that the original equipment is the best suited for a given piece of equipment, particularly when it comes to opamp rolling due to circuit design and controversy, but many opamps alternates work well in given circuit, sometimes better that the originally selected opamp. This is not always the case though, as you'll see in my results that are subjective, but of course one can measure the distortion and see what works "best", which could be the original opamp. Tube rolling is another animal, but this circuit does not optimize the 12AX7 power need. More on that later. The point here is that based on many years of experience, some types of distortion is pleasing to the ear, which is why folks buy tube gear to begin with, however, in a revealing system, small changes in the circuit via opamp rolling are indeed audible. At the end of any experiment like this, the important thing to me is that I get a result that sounds natural and realistic. I bought the Aiyima T20 to experiment with, so I am definitely not relying on or expecting the T20 to take me to nirvana, however, I think that with these mods, the T20 makes for an incredible bargain and definitely can be the basis of a great system.
I have had and have my fair share of tube gear so when I bought the Aiyima T20 I expected that the tubes will take 100+ hours to sound like the given tube should sound, and typically this occurs in jumps after the first couple of listening sessions. The T20 did not work like that, really. I was a little disappointed – I bought the T20 when it first hit the shelves so it’s definitely broken in. I was expecting some of that classic Aiyima (e.g. A70) direct, clear, and nearly clinical sound, but based on price (not expecting miracles, just bang for buck), it sounded to me like something was up. I read around the interweb to see what some folks have done with their T20s and the feedback has been all over the map.
After enough forum reads, I studied the board from tube signal output to opamps to caps to back panel connectors. Using an iPhone with Magnifier app running, I confirmed that the traces (tiny traces) from opamp to output go through the Panasonic electrolytic caps, seemingly working as coupling caps for the opamp outputs. This design is old school, but can give great results, however, an electrolytic cap here is no bueno, no matter the quality level. Electrolytics tend to compress the soundstage and veil the sound, but in a bad way – smearing, as some would say (subjective to user experience and taste). In this case, it appears intentional* since WIMA caps were leveraged at the tubes. Before the 1980’s, aluminum can style electrolytic caps were often standard types in the audio chain (not unusual in PC motherboards today), along with ceramic caps (gasp!) and others. Today, if the baseline is clean and lively sound with a touch of tube harmonics, then the electrolytics in the signal path must go. Note: I have no idea how the electrolytics measure compared to other cap types – could be identical as far as SNR goes.
I replaced 10uf50V Electrolytic caps with AD MPT caps** seen below. I was worried the leads on the caps were going to be too low of gauge to pass through the original through-hole, but it just fit, after removing 98.5% of the original solder in the through-hole with a solder vac iron.
Just for clarity, since I recently read confusing feedback in a forum about MPT caps, a MPT is a MKP but abbreviated differently to describe the shape, in this case a cylinder/tube shape, aka Metalized Polypropylene Tubular. Typically, a 10uf polypropylene cap is going to result in a size too large for this location on the board. I searched by polypropylene type and as low of voltage I could find above 50V (the Panasonic’s voltage rating) since lower voltage usually means less size/mass. I found these AD MPT caps which just squeeze in at 24mmX11mm. The needed a Z-bend in the bottom lead so it could get close to the top lead that is dropped down to the base so the leads could align with the available hole spacing.
New Old Stock for the AD brand is typically not very Old, so I did not stop to think and hit Buy It Now on 10 NOS AD MPT capacitors. The AD brand of polypropylene caps are underrated, IMO. For the price, they win. They nuke the WIMAs infamous red rectanglular boxes (again, Opinion, and AD makes rectangles too), which are typically polypropylene as well, not bad, but economized, however, in the T20, the WIMAs are the polystyrene variety, which also are not bad, but economized, which typically means “to make more efficient” but they are not my jam – they just don’t seem the best option for the $$. It’s tough to find a variety of sizes for newer manufacture caps, especially something small enough to fit in the original spot.
I could not wait to hear the results in action, so I did the first two “channels” from the outside, which are for the XLR outs (living room running balanced except for the turntable to phono preamp). Each opamp takes care of the two soundwave channels for each balanced channel. The improvement with the AD capacitors changes the game completely on this unit. It brings on the imaging, soundstage, and clarity. I then went on an opamp rolling bonanza. Here is what I could test in three hours, in order of best to last place:
Top three:
Most interesting: OPS200 discrete 8-pin amp from AliExpress. See image below. This is a very good discrete unit and is either first place or here at number four, depending on what you prefer. The leader in this test, the OPA2107P, is not organic like this discrete opamp replacement, and the two soundstages are also different. The OPS200 narrows the soundstage a bit and imaging is pretty good, enough to put the presentation of sound that is more in front of the listener as if on a stage the size between speakers and then expanding outward towards the listener instead of all around the listener where sound can appear to be beyond the walls. Both exhibited good depth. I like this discrete unit more than many others because of the organic naturalness of the sound. Sparkos could be a step up here, as could Sonic Imagery or the like, but the cost to reward ratio goes way down (for those excellent discrete units) in this application / circuit / product.
Very good subjective performance:
I read good things about the Burson V7C discrete 8-pin amp, probably more so for the Vivids, but using these in the T20 surprised me. The sound clarity of the V7C was fine but the soundstage quite weak and definitely not as good as the options above. Bottom line – I do not recommend the Burson V7C in this unit. I am not going to try inserting V7 Vivid units.
After all the tests, I soldered in the rest of the AD poly caps and added one more OPA2107P to feed the RCAs with the same good stuff as the balanced output of the XLRs.
And so concludes my adventure with the Aiyma T20 mods. The AD MPT 250V caps replaced the Panasonic electrolytic caps, OPA2107P opamps are in the three 8-pin swap slots, and I am done. I am satisfied. The sound is optimized rather than economized, yet the modification was a material cost of about $30. The subjective performance leap is remarkable and welcome. I may get bored and try a OPA1612 or another discrete opamp style 8-pin replacement. For a lower budget, just leave the original opamps in. Now, what about the tubes? I tried removing the PSVANE stock tubes and put in a pair of used but in great condition ECC81 Gold Aero Tungsram short plate tubes - thumbs up! I don't know what these tubes go for on the market, but this sounds about as good as it can get with excellent imaging, soundstange, and clarity - more clarity than with the PSVANE pair. I will also try a linear power supply which may not have much of an impact, but that also depends on the utility power source and most of my equipment goes through some filtering, except higher power amps. For now, the modded Aiyima T20+ is back at my office / gaming / studio table and sounds great as part of a budget system. I am using a pair of Fosi V3 TPA3255 mono amps with it and that’s going into EPOS ES-7N monitors (not too budget), while the Aiyima T20 is fed with the Denefrips Ares II via USB from my PC, also through XLR connections like the test system, end to end balanced. From the exterior of the T20 preamp, there is no visual indication of modification.
After the testing, I found where the limits are for this preamp. The investment I made into this project leads me to believe that a different pair of tubes will probably help, but based on the base cost of the unit and that it now sounds as good as a $1.5K preamp (no joke), adding expensive tubes will may be a bad return on investment, however, if you want to, please change those coupling caps - they are the one thing holding this unit back from being a serious giant killer. I'd be happy to let a reputable community member / Amir test the unit for noise and distortion to see what may have changed, but the cap change is the foundation of the sound change in this preamp, so I expect little change from the original, except for quality of sound which is subjective.
The End.
*If the electrolytics were intentionally used here, I guess it was probably to soften the sound and make it relatively appealing with most recordings. This circuit could sound a bit harsh if DC coupled, subtracting the bypass cap and adjusting the circuit – just an unproven possibility.
**No change to opamps or tubes will result in a better sound than putting a decent poly cap in place of an electrolytic in the analog sound path. As some have discovered, tube rolling benefits are limited, especially when you consider tube price to performance to the T20 price and performance. In addition, distinctive differences in opamp rolling, whether due to the underlying circuit design or compatibility, are less varied than they could be with the electrolytic output caps in place.
TL;DR: GOTO 10th paragraph. BASICally, skip to the end conclusion.
For the record, I am not an electrical engineer, nor do I have the equipment to make noise measurements – I am just a somewhat experienced hobbyist that likes to assemble predesigned modular or repurposed audio boards and guts to make Frankenstein-esque audio components at glacial pace. Too many hobbies. Hobbies within hobbies. The perspective of this review is that the subjective testing of the T20 is performed by replacing a more expensive preamp, and the rest of the component chain is up in a different league, performance/cost wise. This is intended so I could reveal as much as I could from the mod changes.
In the following passages, there is one rule: I will express many subjective opinions. I do recognize the belief that the original equipment is the best suited for a given piece of equipment, particularly when it comes to opamp rolling due to circuit design and controversy, but many opamps alternates work well in given circuit, sometimes better that the originally selected opamp. This is not always the case though, as you'll see in my results that are subjective, but of course one can measure the distortion and see what works "best", which could be the original opamp. Tube rolling is another animal, but this circuit does not optimize the 12AX7 power need. More on that later. The point here is that based on many years of experience, some types of distortion is pleasing to the ear, which is why folks buy tube gear to begin with, however, in a revealing system, small changes in the circuit via opamp rolling are indeed audible. At the end of any experiment like this, the important thing to me is that I get a result that sounds natural and realistic. I bought the Aiyima T20 to experiment with, so I am definitely not relying on or expecting the T20 to take me to nirvana, however, I think that with these mods, the T20 makes for an incredible bargain and definitely can be the basis of a great system.
I have had and have my fair share of tube gear so when I bought the Aiyima T20 I expected that the tubes will take 100+ hours to sound like the given tube should sound, and typically this occurs in jumps after the first couple of listening sessions. The T20 did not work like that, really. I was a little disappointed – I bought the T20 when it first hit the shelves so it’s definitely broken in. I was expecting some of that classic Aiyima (e.g. A70) direct, clear, and nearly clinical sound, but based on price (not expecting miracles, just bang for buck), it sounded to me like something was up. I read around the interweb to see what some folks have done with their T20s and the feedback has been all over the map.
After enough forum reads, I studied the board from tube signal output to opamps to caps to back panel connectors. Using an iPhone with Magnifier app running, I confirmed that the traces (tiny traces) from opamp to output go through the Panasonic electrolytic caps, seemingly working as coupling caps for the opamp outputs. This design is old school, but can give great results, however, an electrolytic cap here is no bueno, no matter the quality level. Electrolytics tend to compress the soundstage and veil the sound, but in a bad way – smearing, as some would say (subjective to user experience and taste). In this case, it appears intentional* since WIMA caps were leveraged at the tubes. Before the 1980’s, aluminum can style electrolytic caps were often standard types in the audio chain (not unusual in PC motherboards today), along with ceramic caps (gasp!) and others. Today, if the baseline is clean and lively sound with a touch of tube harmonics, then the electrolytics in the signal path must go. Note: I have no idea how the electrolytics measure compared to other cap types – could be identical as far as SNR goes.
I replaced 10uf50V Electrolytic caps with AD MPT caps** seen below. I was worried the leads on the caps were going to be too low of gauge to pass through the original through-hole, but it just fit, after removing 98.5% of the original solder in the through-hole with a solder vac iron.
Just for clarity, since I recently read confusing feedback in a forum about MPT caps, a MPT is a MKP but abbreviated differently to describe the shape, in this case a cylinder/tube shape, aka Metalized Polypropylene Tubular. Typically, a 10uf polypropylene cap is going to result in a size too large for this location on the board. I searched by polypropylene type and as low of voltage I could find above 50V (the Panasonic’s voltage rating) since lower voltage usually means less size/mass. I found these AD MPT caps which just squeeze in at 24mmX11mm. The needed a Z-bend in the bottom lead so it could get close to the top lead that is dropped down to the base so the leads could align with the available hole spacing.
New Old Stock for the AD brand is typically not very Old, so I did not stop to think and hit Buy It Now on 10 NOS AD MPT capacitors. The AD brand of polypropylene caps are underrated, IMO. For the price, they win. They nuke the WIMAs infamous red rectanglular boxes (again, Opinion, and AD makes rectangles too), which are typically polypropylene as well, not bad, but economized, however, in the T20, the WIMAs are the polystyrene variety, which also are not bad, but economized, which typically means “to make more efficient” but they are not my jam – they just don’t seem the best option for the $$. It’s tough to find a variety of sizes for newer manufacture caps, especially something small enough to fit in the original spot.
I could not wait to hear the results in action, so I did the first two “channels” from the outside, which are for the XLR outs (living room running balanced except for the turntable to phono preamp). Each opamp takes care of the two soundwave channels for each balanced channel. The improvement with the AD capacitors changes the game completely on this unit. It brings on the imaging, soundstage, and clarity. I then went on an opamp rolling bonanza. Here is what I could test in three hours, in order of best to last place:
Top three:
- OPA2107P. These were made for this circuit. Maximum symbiosis. Image below taken with flash to show the Burr-Brown ships clearly.
- OPA2227P. Almost as good or the same as the OPA2107A.
- MUSES01. This exceedingly rare opamp sounds great in this unit.
Most interesting: OPS200 discrete 8-pin amp from AliExpress. See image below. This is a very good discrete unit and is either first place or here at number four, depending on what you prefer. The leader in this test, the OPA2107P, is not organic like this discrete opamp replacement, and the two soundstages are also different. The OPS200 narrows the soundstage a bit and imaging is pretty good, enough to put the presentation of sound that is more in front of the listener as if on a stage the size between speakers and then expanding outward towards the listener instead of all around the listener where sound can appear to be beyond the walls. Both exhibited good depth. I like this discrete unit more than many others because of the organic naturalness of the sound. Sparkos could be a step up here, as could Sonic Imagery or the like, but the cost to reward ratio goes way down (for those excellent discrete units) in this application / circuit / product.
Very good subjective performance:
- OPA2604
- LME4562
- LME49720
- OracleII-02
I read good things about the Burson V7C discrete 8-pin amp, probably more so for the Vivids, but using these in the T20 surprised me. The sound clarity of the V7C was fine but the soundstage quite weak and definitely not as good as the options above. Bottom line – I do not recommend the Burson V7C in this unit. I am not going to try inserting V7 Vivid units.
After all the tests, I soldered in the rest of the AD poly caps and added one more OPA2107P to feed the RCAs with the same good stuff as the balanced output of the XLRs.
And so concludes my adventure with the Aiyma T20 mods. The AD MPT 250V caps replaced the Panasonic electrolytic caps, OPA2107P opamps are in the three 8-pin swap slots, and I am done. I am satisfied. The sound is optimized rather than economized, yet the modification was a material cost of about $30. The subjective performance leap is remarkable and welcome. I may get bored and try a OPA1612 or another discrete opamp style 8-pin replacement. For a lower budget, just leave the original opamps in. Now, what about the tubes? I tried removing the PSVANE stock tubes and put in a pair of used but in great condition ECC81 Gold Aero Tungsram short plate tubes - thumbs up! I don't know what these tubes go for on the market, but this sounds about as good as it can get with excellent imaging, soundstange, and clarity - more clarity than with the PSVANE pair. I will also try a linear power supply which may not have much of an impact, but that also depends on the utility power source and most of my equipment goes through some filtering, except higher power amps. For now, the modded Aiyima T20+ is back at my office / gaming / studio table and sounds great as part of a budget system. I am using a pair of Fosi V3 TPA3255 mono amps with it and that’s going into EPOS ES-7N monitors (not too budget), while the Aiyima T20 is fed with the Denefrips Ares II via USB from my PC, also through XLR connections like the test system, end to end balanced. From the exterior of the T20 preamp, there is no visual indication of modification.
After the testing, I found where the limits are for this preamp. The investment I made into this project leads me to believe that a different pair of tubes will probably help, but based on the base cost of the unit and that it now sounds as good as a $1.5K preamp (no joke), adding expensive tubes will may be a bad return on investment, however, if you want to, please change those coupling caps - they are the one thing holding this unit back from being a serious giant killer. I'd be happy to let a reputable community member / Amir test the unit for noise and distortion to see what may have changed, but the cap change is the foundation of the sound change in this preamp, so I expect little change from the original, except for quality of sound which is subjective.
The End.
*If the electrolytics were intentionally used here, I guess it was probably to soften the sound and make it relatively appealing with most recordings. This circuit could sound a bit harsh if DC coupled, subtracting the bypass cap and adjusting the circuit – just an unproven possibility.
**No change to opamps or tubes will result in a better sound than putting a decent poly cap in place of an electrolytic in the analog sound path. As some have discovered, tube rolling benefits are limited, especially when you consider tube price to performance to the T20 price and performance. In addition, distinctive differences in opamp rolling, whether due to the underlying circuit design or compatibility, are less varied than they could be with the electrolytic output caps in place.
Go for it! Though I'd send it with a 3A linear power supply just for good measure (pun intended).
That would be a fascinating measurement if you sent it in.
RandomEar
Addicted to Fun and Learning
- Joined
- Feb 14, 2022
- Messages
- 757
- Likes
- 1,914
Welcome to ASR @extraclassic . As I'm sure you are aware if you've been reading ASR for a long time, good and reliable testing of audio devices cannot be done sighted. There's a multitude of problems with the way you tried to establish the assumed differences in sound you reported:
- Sighted testing is heavily skewed by bias. Being aware of potential biases does not compensate for them.
- Long switching times like you had to use for swapping/soldering parts makes it impossible to hear small potential differences because our echoic memory is just not long enough for that. It's a couple of seconds at best.
- Precise level matching to 0.2 dB or better, ideally done by measuring the output voltage, is required for any comparison of audio devices. This is related to the equal-loudness contour of our hearing and the resulting fact that "louder almost always sounds better".
Here if you want to put together a good pre amp, if you think it's fun to solder and so on, there's this pre amp board. If you want something to tinker with, that is.
Michael Beeny has designed it and it measures really well, super low distortion and so on.
Speaking of quality of the products; if, when, how and what affects the sound. By affects I mean audibly. Michael Beeny discusses that in his videos. He has a sensible engineering approach to it. However, he recommends choosing a volume potentiometer (Alps) of good quality.
Who knows, someone in the USA might put this together and send it to Amir for testing:
Last edited:
Yes indeed.
If you take Amir's chart for amplifiers (there isn't one for the Aiyima T20 or possibly pre-amplifiers for some reason), there are measurement tiers:
- Excellent
- Very Good/Good
- Fair
- Poor
And at what point does an audible difference become apparent, as opposed to bias or imagined etc. Perhaps there is no audible difference between fair and excellent?
This would be interesting to explore.
This came in as I was typingHere if you want to put together a good pre amp, if you think it's fun to solder and so on, there's this pre amp board. If you want something to tinker with, that is.
Michael Beeny has designed it and it measures really well, super low distortion and so on.
Speaking of quality of the products; if, when, how and what affects the sound. By affects I mean audibly. Michael Beeny discusses that in his videos. He has a sensible engineering approach to it. However, he recommends choosing a volume potentiometer (Alps) of good quality.
Who knows, someone in the USA might put this together and send it to Amir for testing:
spot on! I'll stick my head out enough to say I think volume controls make a difference.
Last edited:
Speaking of great design and super quality, check this out:Yes indeed.
If you take Amir's chart for amplifiers (there isn't one for the Aiyima T20 or possibly pre-amplifiers for some reason), there are measurement tiers:
So the issue here is what component or design choice by a the designer/manufacturer might move a product from fair to good to excellent.
- Excellent
- Very Good/Good
- Fair
- Poor
And at what point does an audible difference become apparent, as opposed to bias or imagined etc. Perhaps there is no audible difference between fair and excellent?
This would be interesting to explore.
This came in as I was typing
Accuphase C-280 Review (Preamplifier)
Hello Everyone, This a review and detailed measurements of the Accuphase C-280 peamplifier. Accuphase C-280 - Presentation This unit is from a friend who wanted to know how it performs and is ok to let me report the results, thanks to him! Released in 1982, the C-280 was the top of the...
www.audiosciencereview.com
extraclassic
Member
- Joined
- Sep 14, 2020
- Messages
- 5
- Likes
- 4
Hi RandomEar, thank you for the feedback. I enjoy the forum back and forth and know that this forum focuses heavily on measurements.
I totally get it. Sighted bias is real. Time delays between switching components is real - audible memory is the worst. While I am certain the caps made an audible difference, measuring will likely give a zero sum game or loss, or at least close to zero. In 30 years of modding audio, the pitfalls of common subjective estimations are obvious to me, but not so much when I first started in this fun hobby. I certainly will not claim that any measurements are different, improved, worse, etc.
Regarding the opamps, the improvement that can be made from the original units is fairly negligible, or not an improvement at all due to circuit design. The tolerances between opamps and tubes are definitely In a different ballpark. But like anything that changes in a circuit, tubes or otherwise, there is some impact.
At any rate, my post is really intended to share a personal experience and should be taken for what it is worth to the reader. As the endless debate goes, there is more to audio than measurements, though less distortion and noise is always better to remove unwanted sounds which travels source material better than without noise/distortion. I am addicted to the imaging and sound field + soundstage size, so I build equipment or mod equipment to that end. Unfortunately, there is no good standard to measure sound field and soundstage, which cannot be represented in a stereo oscilloscope measurement, yet can be easily perceived by the human ear for most people.
A variable that gets lost in these type of discussions is individual perception. That’s why we measure - it’s the best way to obtain comparisons and electronic performance data. In other words, everyone’s ears (and eyes, and other senses) are different. That is why subjective analysis is important, and yet is also inadequate to describe accuracy. A -120dB vs a -110dB THD or SINAD measurement is almost impossible to hear. That means we can pump out products that have incredibly excellent technical performance and yet these products can subjectively sound worse than another product with less technical performance. Regardless of measurements, perception is always reality for the individual. Yet, to level the playing field, measurements are critical. It’s a dichotomy that makes describing subjective sound very difficult since room and room treatment, equipment, and the individual’s hearing ability can have wild variables.
That all said, I’m likely preaching to the choir. I appreciate technical measurements and this forum. I will always caveat my opinions as opinions, including everything I just wrote above.
extraclassic
Member
- Joined
- Sep 14, 2020
- Messages
- 5
- Likes
- 4
Good call, DanielT! I love Michael’s work - he does experimenting to help users avoid pitfalls and does a great job explaining how he puts things together. I hope Mr. Beeny is around for a long time to come. A wonderful resource for beginners and experts alike.Here if you want to put together a good pre amp, if you think it's fun to solder and so on, there's this pre amp board. If you want something to tinker with, that is.
Michael Beeny has designed it and it measures really well, super low distortion and so on.
Speaking of quality of the products; if, when, how and what affects the sound. By affects I mean audibly. Michael Beeny discusses that in his videos. He has a sensible engineering approach to it. However, he recommends choosing a volume potentiometer (Alps) of good quality.
Who knows, someone in the USA might put this together and send it to Amir for testing:
RandomEar
Addicted to Fun and Learning
- Joined
- Feb 14, 2022
- Messages
- 757
- Likes
- 1,914
@extraclassic
Sound stage is a property of the recording and influenced to some degree by speakers and room.
If you can hear it, you were able to measure it orders of magnitude earlier. An audio signal is not complex: It has time and amplitude information, nothing else.
If it measures identical, it will sound identical. The fact that people subjectively perceive differences between devices which measure the same is irrelevant, because these differences vanish in blind testing: They are not real, not reproducible, they are 100% a trick of our brain.
The world will not end if you invest 20$ into "better" caps and your brain tells you that the sound has improved. You're happier and that is worth something. But there are companies and whole industries out there making a business out of essentially scamming people by selling them "improved imaging" for 5000$ per cable and lying to them about how this stuff works. It's important to realize what matters for good audio reproduction and what doesn't.
Sound stage is a property of the recording and influenced to some degree by speakers and room.
If you can hear it, you were able to measure it orders of magnitude earlier. An audio signal is not complex: It has time and amplitude information, nothing else.
If it measures identical, it will sound identical. The fact that people subjectively perceive differences between devices which measure the same is irrelevant, because these differences vanish in blind testing: They are not real, not reproducible, they are 100% a trick of our brain.
The world will not end if you invest 20$ into "better" caps and your brain tells you that the sound has improved. You're happier and that is worth something. But there are companies and whole industries out there making a business out of essentially scamming people by selling them "improved imaging" for 5000$ per cable and lying to them about how this stuff works. It's important to realize what matters for good audio reproduction and what doesn't.
extraclassic
Member
- Joined
- Sep 14, 2020
- Messages
- 5
- Likes
- 4
SMen, this is the magic question. All products are produced to a price point, even those one-offs I make for folks. That means some components in mass produced audio equipment have cost saving decisions that can limit potential. Measurement wise, changes to the T20 may not be able to be improved. Subjective results, however, can change. In the case of the mods I performed for the T20, the sound to me is markedly improved. Can it match my best audio equipment. No, however the cost to sound ratio of the T20 is pretty hard to beat, especially with the cap change. Again, that’s my subjective viewpoint.Yes indeed.
If you take Amir's chart for amplifiers (there isn't one for the Aiyima T20 or possibly pre-amplifiers for some reason), there are measurement tiers:
So the issue here is what component or design choice by a the designer/manufacturer might move a product from fair to good to excellent.
- Excellent
- Very Good/Good
- Fair
- Poor
And at what point does an audible difference become apparent, as opposed to bias or imagined etc. Perhaps there is no audible difference between fair and excellent?
This would be interesting to explore.
This came in as I was typing
extraclassic
Member
- Joined
- Sep 14, 2020
- Messages
- 5
- Likes
- 4
I agree with much of what you stated, RandomEar. No doubt that measurements are irrefutable evidence of technical performance if done correctly.
While I don’t always agree with measurement to audible sound equivalency, that’s okay because that’s my opinion based on historical critical blind testing (which I did not perform on the T20). Those blind or double blind audible tests are skewed as well by individual perception, but much better than one point of view without testing. What is subjectively better to me isn’t what is better to others. And that is okay too.
The industry scams are real, no doubt. I have zero interest in something that is high cost low value. A basic example for the T20 changes I made is that I really wanted the Burson V7 opamps to work in the T20. They were the most expensive opamp replacements I tried, but they were not useful at all. I love value and since I also love modding, I don’t make changes that are terrible ROIs, like buying $800 tubes for a $170 preamp. I don’t even want to discuss cables. There is way too much voodoo in that subject.
Room acoustics are the largest impact on sound, then speakers, then the remaining components / audio equipment. Speakers typically have the highest distortion heard in a given system, distortion that can make equipment comparisons irrelevant. Measurement taken in a “lab” and in a room with a mic are always different, unless of course we choose to hang out in an anechoic chamber, but then our human bodies would impact sound like they do in a room. This assumes all other things being equal. Room acoustics can really hamper or help performance of a system. When asked about the number one improvement to make, i suggest two things. First, speaker placement is critical. This is in effect a method to optimize a speaker design with a room. Then I suggest room treatment.
Thank you for the feedback - I appreciate your perspective and understand it.
Similar threads
- Poll
