Impossible to know, one must compare side by side and there’s always a subjective opinion what’s sounding the best .
4 Wima:s 4,7 uF has replaced the 4 Elna capacitors right before the tpa chip. They were 2mm to thick to fit in a row so two of them is mounted below the board.
The 4 output ceramic caps are replaced by 4 Wima mks2 1uF ( two of them below the bord . This lowers the distortion significantly above 2 kHz with this board ( Audiogues measurements ).
This mod sounded terrible first 10 minutes , I was very dissapointed , however - we watched a movie using the amplifier before night and I left the amplifier on during the whole night .
The sound this morning has transformed completely and is without any doubt an even better sound than before .
Those who don’t believe Wima caps need burn in should do this experience to complete change their mind 🙂
Next to come = lm4562 sounding better than opa1612 or not ?
The 4 output ceramic caps are replaced by 4 Wima mks2 1uF ( two of them below the bord . This lowers the distortion significantly above 2 kHz with this board ( Audiogues measurements ).
This mod sounded terrible first 10 minutes , I was very dissapointed , however - we watched a movie using the amplifier before night and I left the amplifier on during the whole night .
The sound this morning has transformed completely and is without any doubt an even better sound than before .
Those who don’t believe Wima caps need burn in should do this experience to complete change their mind 🙂
Next to come = lm4562 sounding better than opa1612 or not ?
Last edited:
It's so nice to see bold claims backed-up by actual measurements. 😊
Familiar territory. Experienced as a kid, then educated it is nonsense. It just isn’t. Still hear it every time. Therefor scepticism when people try out Wima and decide after a few minutes they are not OK. Give them a day, don’t overthink and simply accept.
The measurements won’t show a difference. That’s OK 😉
The measurements won’t show a difference. That’s OK 😉
Last edited by a moderator:
Thanks ! I guess you want to know how these two amplifiers sound in comparison ?😉
I did a comparison recently , before the last mod.
The hypex ncore 125 and Linn 5125 are very similar in the bass - they are both very good and you get a bass thats easy to follow , soundstage is bigger on hypex ncore , deeper and wider and the 5125 is slightly brighter in the treble, but at the same time not as clean sounding
The tpa3255 with 36V supply and opa1612 sound more punchy, have even better bass thats even easier to follow, very exact soundstage where every instrument can be pinpointed , the treble is better than both the 5125 and ncore, more open than ncore and cleaner than 5125.
Its more fun to listen to the tpa3255. The musicians are there in your listeningroom right infront of you.
The ncore and tpa3255 is dead silent even when having your ear right next to the speaker , with 5125 you can hear a silent noise.
I havent heard many class A/B amplifiers regardless of price thats better in playing bass than ncore, 5125 and this tpa3255. Most amplifiers cant articulate the bass in a convincing way.
I wonder …. In the hypex there are some extra HF filtering done by audiophonics happening before the ne5532 buffer , and maybe this might be the cause of the slightly darker sound ( darker than both 5125 and tpa3255 )
Its interesting that Linn products has swapped the analog chipamp tda7293 that they have used since -93 , and are now using tpa3255 in their own implementation for the Selekt DSM.
Last edited:
Yes, break in is real and sometimes its very audible and sometimes just a small difference. Some manufacturer break in their amplifiers 24 hours for better sound and also checking for faults that might happen within that time-lap.
Update:
I have now tried the lm4562 and compared it to the opa1612 but in this amplifier, I prefer opa1612 with a small margin. It was the opposite with my Aiyima a04 with the tpa3251 chip. Now how can that be ?
My thoughts and speculation about this is that the 3255 and 3251 dont have exactly the same caracter. The 3255 chip sound more punchier and has better bass , the 3251 sound leaner and need a bolder sounding OP, like lm4562.
Another speculation is that the lm4562 might be oscillating with this board and this could be the reason why it sounds slightly worse. I let my ears decide and it will be opa1612 that stays.
Now, can it be even better using a Sparkos lab SS3602 ?
I have now tried the lm4562 and compared it to the opa1612 but in this amplifier, I prefer opa1612 with a small margin. It was the opposite with my Aiyima a04 with the tpa3251 chip. Now how can that be ?
My thoughts and speculation about this is that the 3255 and 3251 dont have exactly the same caracter. The 3255 chip sound more punchier and has better bass , the 3251 sound leaner and need a bolder sounding OP, like lm4562.
Another speculation is that the lm4562 might be oscillating with this board and this could be the reason why it sounds slightly worse. I let my ears decide and it will be opa1612 that stays.
Now, can it be even better using a Sparkos lab SS3602 ?
LM4562 is not a trouble free opamp. It is sensitive to EMI that is all around you and maybe even generated by that very amplifier itself. Of course you want graphs/papers and whatnot. I don't document, I just don't use parts anymore that turn out to have issues. I bought LM4562 years ago only to discover half of the lot had popcorn noise. Then also noticed susceptibility and discarded them.
https://www.audiosciencereview.com/...pecially-of-the-lm4562-lme497x0-family.10687/
Discrete stuff probably is so off that they start to sound nice again. Effect generators. No way that discrete technology can be as precise and well performing as IC opamps and certainly not the tower like structures some produce. Stuff for believers or people that want stuff to be "for audio by XYZ" like marketing for clothing or plastic shoes.
https://www.audiosciencereview.com/...pecially-of-the-lm4562-lme497x0-family.10687/
Discrete stuff probably is so off that they start to sound nice again. Effect generators. No way that discrete technology can be as precise and well performing as IC opamps and certainly not the tower like structures some produce. Stuff for believers or people that want stuff to be "for audio by XYZ" like marketing for clothing or plastic shoes.
Last edited:
Thanks for the link. Very interesting that many bipolar OP:s are sensitive to airborne EMI . Even the dinosaur ne5532. I guess the opa1612 is also, but I hope at lesser extent than lm4562.
Have you tried LME49720? When we build dac they sounded ok as output stage. But 1612 was at last the winner…
Yeah they are good but I sold them all a few weeks ago. There are even better ones around.
Last edited:
Hi, whats the name of those type of ¨banana¨ conectors?... there are flush, and i also saw those on some speakers... thanks for the info
So far in my listening tests 1612 , opa 2134, ne5532 , lm4568 , LF353. The 353 is the best of all.
I do prefer a light airy treble rather then brassy.
I haven't tested my 3255 board yet so the test system is linear mosfet amp with tube preamp.
I do prefer a light airy treble rather then brassy.
I haven't tested my 3255 board yet so the test system is linear mosfet amp with tube preamp.