Lepai T-Amp with TA2020

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
For significant improvement to the Lepai, a friend of mine, armed with the right tool, managed to desolder the stock surface-mounted JRC 4558D opamp and replaced with a LM4562 with very good results.
Other than a bigger linear supply, this is the only mod to my Lepai. It's the LP-2020A+ with the blue PCB. I have no before & after measurements, so it's nice to see others having success replacing the op amps. I discussed it some back in post #842 and post #1147.
It's not a difficult mod. It can be accomplished with soldering iron, X-acto utility knife, solder wick, and a bit of patience while working.
 
[snip]
Personally, I don't think changing the heatsink or reapplying the heatsink compound makes any difference. After all, the Tripath TA2020 runs cooler than most Class AB chipsets (some Class AB amps doesn't even utilized a heatsink!). I won't go the extra mile if the tweak doesn't improve sound quality. [/snip]

I agree, but it's 10 minutes work, and you never know - you might end up thrashing the chip to within an inch of it's life for hours at a time (as I might if I use it to drive a passive sub), and it might extend its life a little, so why not?!
 
Last edited:
KopiOkaya said:
For significant improvement to the Lepai, a friend of mine, armed with the right tool, managed to desolder the stock surface-mounted JRC 4558D opamp and replaced with a LM4562 with very good results.
Other than a bigger linear supply, this is the only mod to my Lepai. It's the LP-2020A+ with the blue PCB. I have no before & after measurements, so it's nice to see others having success replacing the op amps. I discussed it some back in post #842 and post #1147.
It's not a difficult mod. It can be accomplished with soldering iron, X-acto utility knife, solder wick, and a bit of patience while working.

Thanks for taking the trouble to reference these earlier Lepai opamp mod posts - it keeps this HUGE thread useful.
 
consider avoiding these inductors

See the blue inductors, those ones are not good. If you do not connect amplifier to speaker, just play some music, you can hear those inductor making sound, which is a bad indicator for extra nonlinearity. I replaced them and the sound improved.



Those of you who are not good with soldering and assembling but prefer a high quality Tripath TA2020 amp can consider the SMSL SA-36A.

T2hQrkXgdXXXXXXXXX_!!57649962.jpg
 
Not to get too off topic but do you know if there is any common consensus about the inductors used in this YuanJing board with a TA2020.

They seem to sound better than the ones that came installed in the LePai TA2020A+ that I have, but I wonder where they fall in the list of inductors from worse to best.

An externally hosted image should be here but it was not working when we last tested it.
 
Lepai tone-defeat reverses output phase!

I've set up the Lepai and Indeed TA2020's together. Input is looped through the Lepai on to the Indeed.

The Indeed is driving Wharfedale DIamond 9.0's, the Lepai a 20 year old JPW passive sub.

The JPW sub is 12 ohms, intended to be connected to 8 ohm speakers such as JPW mini-monitors (obviously), which are 8 ohm. The paralleled loads would give c. 6 ohms, but the Lepai is only seeing the 12 ohm load, hence it can probably manage about 5-6 clean watts per channel into it.

The JPW actually has label on it which declares that its senstivity is 95dB/watt at 1 metre, for what that's worth.

Setting up, I had to determine phase. I tried music and bass test tones - listening at various positions in the room, then shutting down and reversing the speaker connections on the JPW, very laborious and inconclusive. I just left them both connected the same way round (which happens to be with all speaker connections reversed - I gather tripaths invert).

BUT!!! Discovered that the tone circuit reverses the output phase!

Despite they're dropping overall level, when I enabled the tone controls (set at flat), the in-room bass levels jumped substantially.

So there you have it - a very useful feature of the Lepai is that the tone-defeat doubles up as an output phase switch!

So far, so very impressive. I just listened all the way through Air's "Talkie Walkie", and it was like some-one had shaken me awake when it ended.

ETA >> the Lepai reverses channels on its 3.5 input jack, easily fixed by swapping its RCA inputs around.
 
Last edited:
Well done! Great discovery.:cheers:

Cheers to you too!

It was so tedious, shutting down, reversing speaker terminals, firing-up again and trying to ascertain whether the JPW was reinforcing the 9.0's or not - in a furnished room it varies from one part of the room to the next.

But I happened to have "Youth and Young Manhood" (Kings of Leon) on and the difference when I hit the tone button on the Lepai was pretty obvious.

It's probably far from perfect - they don't put variable phase-adjustment on 'proper' sub-woofers for the hell of it - but it's better.
 
A weird issue has developed - this is FYI, although someone might be able to explain it.

Here's the setup;

The amps are connected to my LG monitor's headphone out, which is surprsingly clean - it appears to be a dedicated circuit, switched in electronically when a jack is detected. It's a 'flat' output - no user adjustment for EQ, balance etc'. I've run sweeps and test tones, and response extends from practically DC to beyond by hearing (about 14KHz these days).

My PC is connected to the monitor via the nVidia graphics card's HDMI out - the advantage is that it provides me with a volume control via the LG's remote.

Musically, it actually sounds as good or even better than any of the analogue audio outputs on the PC (motherboard audio, M-Audio graphics card).

BUT ...I'm now getting a lot of noise breaking through from the PC to the TA2020's - crackling, buzzing, farting (!) noises, which correspond to any activity the PC happens to be doing.

Very strangely (but fortunately) it's *completely* independent of volume on the LG (it's the same even if I mute its output), but is still audible on all the LG's inputs (including it's built-in DVB) as I step through them.

However, If I connect headphones (ear-buds), or my Creek headphone amp up and max its volume (which would be excruciatingly loud with a signal) it is completely inaudible, not the faintest trace of it, hence it isn't in the audio band.

So basically it would seem that ultrasonic or (more likely) RF hash from my PC is getting on to the headphone-out of the LG monitor, which ends up modulating the TA2020's in the audio band.

I'm a little concerned that it might be a symptom of a developing fault, either in the LG monitor or PC (or graphics card), since there's no question that it developed months after I first starting using the Lepai (and several weeks after I acquired the Indeed) - there's no question I would have noticed it immediately.
 
Last edited:
......
ETA >> the Lepai reverses channels on its 3.5 input jack, easily fixed by swapping its RCA inputs around.

In case that confused anyone; I have the input connected to the Lepai's RCA's, it's 3.5mm jack then connected to the Indeed TA2020.

Hence the channels are reversed when they get to the Indeed, so I swapped the RCA's on Lepai around (the Lepai is effectively acting as a mono amp, driving the dual-coil driver in the woofer).
 
A weird issue has developed - this is FYI, although someone might be able to explain it.

I'm now getting a lot of noise breaking through from the PC to the TA2020's - crackling, buzzing, farting (!) noises, which correspond to any activity the PC happens to be doing.

Very strangely (but fortunately) it's *completely* independent of volume on the LG (it's the same even if I mute its output), but is still audible on all the LG's inputs (including it's built-in DVB) as I step through them.

However, If I connect headphones (ear-buds), or my Creek headphone amp up and max its volume (which would be excruciatingly loud with a signal) it is completely inaudible, not the faintest trace of it, hence it isn't in the audio band.

So basically it would seem that ultrasonic or (more likely) RF hash from my PC is getting on to the headphone-out of the LG monitor, which ends up modulating the TA2020's in the audio band.

I'm a little concerned that it might be a symptom of a developing fault, either in the LG monitor or PC (or graphics card), since there's no question that it developed months after I first starting using the Lepai (and several weeks after I acquired the Indeed)...

Interesting discovery. It is good to know. :nod:
 
Well, no hash from the motherboard's analogue out (Asus P8Z77).

I intend to try the motherboard's HDMI at some point.

Meanwhile, I connected up my long de-commissioned Cambridge Audio DVD player (a Richer Sounds-owned brand) I bought c. 2003 on the strength of endorsements of its CD replay quality.

Wow. I've just finished listening to a CD-R I made a few years back of Talk Talk's "Colour of Spring" and "Spirit of Eden" (on a TDK 800 MB disc), written with a Plextor writer.

Listened to my favourite tracks from Colour of Spring, and then all the way through Spirit of Eden.

So good - got the volume on the Lepai (driving the JPW sub) just right for the tympani/kettle drums and bass, Hollis's vocal inflections, the brass, the guitar chops are all as clear as I've ever heard.

I just hope this setup is stable and nothing breaks!
 
I'm pretty sure the problem is the Gigabyte nVidia GTX 660 graphics card's HDMI output (which I have to use - yes, I like to play PC games occasionally).

I think it's leaking shed-loads of RF down its HDMI connection to my LG monitor, and that ends up on the LG's (otherwise rather good) headphone-out and modulates the Lepai and Indeed TA2020's.

Sigh. Nothing ever goes to plan.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.