• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. 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!

Topping LA90 Review (Integrated Amplifier)

Rate this amplifier:

  • 1. Poor (headless panther)

    Votes: 35 4.4%
  • 2. Not terrible (postman panther)

    Votes: 50 6.2%
  • 3. Fine (happy panther)

    Votes: 193 24.1%
  • 4. Great (golfing panther)

    Votes: 524 65.3%

  • Total voters
    802

Billy Budapest

Major Contributor
Forum Donor
Joined
Oct 11, 2019
Messages
1,810
Likes
2,674
Is "solder blobs in the center of the main PCB" two circled parts? Even if you magnify the photo, you don't really understand what it is.
I'm also curious about the four soldering points on each of the L / R channels.
If it is a speaker output, it should be good in two places, but ...
What you circled on the far right and far left look like Molex connectors.

There appear to be four unused connectors inside the LA90, probably to be used as test points during manufacture or for diagnostics for repairs.

Or maybe the unused flat cable connector in the middle is used to connect to a front panel PCB not pictured in Audiophonics’ photos. That still leaves an internal USB-C port and the two Molex-looking connectors.
 

solderdude

Grand Contributor
Joined
Jul 21, 2018
Messages
15,891
Likes
35,912
Location
The Neitherlands
Has the idea that the output is capacitor coupled been ruled out? Is it possible that they use output capacitors and take feedback post capacitor.

In the latter case the max. output level at lower frequencies would be compromised and phase shifts (depending on the load) would occur.
One would have to see frequency response at full power. I don't think this would be something JohnY would do this way though. There are 337 and 317which also hint at symmetrical power supplies.
I would either need to have one on my desk or JohnY should be willing to spill some beans. Haven't seen him online lately though.
 

pjug

Major Contributor
Forum Donor
Joined
Feb 2, 2019
Messages
1,775
Likes
1,561
In the latter case the max. output level at lower frequencies would be compromised. One would have to see frequency response at full power. I don't think this would be something JohnY would do this way though. There are 337 and 317which also hint at symmetrical power supplies.
I would either need to have one on my desk or JohnY should be willing to spill some beans. Haven't seen him online lately though.
What I don't get: if you split the rails to give a virtual ground at 32V, then won't the negative speaker terminal be 32V? And that is not the case here.
 

solderdude

Grand Contributor
Joined
Jul 21, 2018
Messages
15,891
Likes
35,912
Location
The Neitherlands
The virtual ground becomes the ground. There will be +/- 32V in that case.

What's weird is that for balanced it seems important to connect both - outputs. This would not be needed when an amp has a regular dual power supply.
It may be that both channels have their own virtual ground and they need to be connected for balanced.

As said. I either need one to scrutinize or JohnY needs to spill some beans. Not going to buy an LA90 as I don't need it (have plenty of power amps doing nothing already :))
 

Billy Budapest

Major Contributor
Forum Donor
Joined
Oct 11, 2019
Messages
1,810
Likes
2,674
What you circled on the far right and far left look like Molex connectors.

There appear to be four unused connectors inside the LA90, probably to be used as test points during manufacture or for diagnostics for repairs.

Or maybe the unused flat cable connector in the middle is used to connect to a front panel PCB not pictured in Audiophonics’ photos. That still leaves an internal USB-C port and the two Molex-looking connectors.
@Toku Looking at the top photo, I can now see that those are not Molex connectors but are solder points, as you stated.
 

pjug

Major Contributor
Forum Donor
Joined
Feb 2, 2019
Messages
1,775
Likes
1,561
The virtual ground becomes the ground. There will be +/- 32V in that case.
It will be +/-32 relative to the virtual ground, but is it +/-32V relative to the external supply ground? I get the feeling I'm missing something trying to understand this.
 

solderdude

Grand Contributor
Joined
Jul 21, 2018
Messages
15,891
Likes
35,912
Location
The Neitherlands
The external supply is not connected to ground in this case.
The - of the PS = -32V and the + of the PS will be the +32V. The ground is generated in that case and signal ground (in and out and internal reference) is connected to that generated ground.

I don't know if this is done though. Just stabbing in the dark. Maybe John uses another trick. The internal opamps for the NF will probably not run on 64V unless these are discrete ones (as in the A90LE ?).
 

pjug

Major Contributor
Forum Donor
Joined
Feb 2, 2019
Messages
1,775
Likes
1,561
The external supply is not connected to ground in this case.
The - of the PS = -32V and the + of the PS will be the +32V. The ground is generated in that case and signal ground (in and out and internal reference) is connected to that generated ground.
Ah. I thought it was shown that the - of the power supply showed continuity with the case. I probably misread. Has it been verified that it is fact as you say above?
 

antcollinet

Master Contributor
Joined
Sep 4, 2021
Messages
7,409
Likes
12,294
Location
UK/Cheshire
Anyone have more thoughts on how Topping is handling powering this with +64V, after seeing the internal photos? Has the idea that the output is capacitor coupled been ruled out? Is it possible that they use output capacitors and take feedback post capacitor. I was looking at the article linked below and wondering if this is essentially what they are doing.

edit: Oh I didn't read the above article very carefully. Cap is not in the feedback loop there. But it seems some people at least attempt this. Can anyone who has opened the amp tell us if the output is capacitively coupled? It would be nice to know if the idea can be ruled out.
I believe the 4 regulators per side are being used to split the 64V to +/-32. There has been some disagreement here though. So until someone reverse engineers the schematic, or measures it we won't know for sure.


What I don't get: if you split the rails to give a virtual ground at 32V, then won't the negative speaker terminal be 32V? And that is not the case here.
No, the power supply output is not referenced to ground, so the center of the split rails can be grounded.
 

Toku

Major Contributor
Joined
Apr 4, 2018
Messages
2,391
Likes
2,776
Location
Japan
The two 4700uF/35V electrolytic capacitors on the L-ch and R-ch on the board in the photo are for creating virtual grounding for R-ch and L-ch, right? It doesn't seem like an output coupling capacitor for the LM3886.
 

Attachments

  • la90-PCB2.jpg
    la90-PCB2.jpg
    157.5 KB · Views: 312

Rottmannash

Major Contributor
Forum Donor
Joined
Nov 11, 2020
Messages
2,969
Likes
2,606
Location
Nashville
What is that device with "MORNSUN" printed on it?
 

solderdude

Grand Contributor
Joined
Jul 21, 2018
Messages
15,891
Likes
35,912
Location
The Neitherlands
Yes, it is a single 12V DC power supply.
I am quite certain this is used for powering the (12V) relays and maybe also feeds the logic circuits via extra regulators.

The 'device' between the electrolytics directly next to the DC input cable may have something to do with the virtual ground rail ?
It is the most logical place to do this. It is unlikely to be a 5V regulator or something like that. Dropping 57V is not logical.
 
Last edited:

pjug

Major Contributor
Forum Donor
Joined
Feb 2, 2019
Messages
1,775
Likes
1,561
I believe the 4 regulators per side are being used to split the 64V to +/-32. There has been some disagreement here though. So until someone reverse engineers the schematic, or measures it we won't know for sure.



No, the power supply output is not referenced to ground, so the center of the split rails can be grounded.
So do I understand correctly that this amplifier is completely floating when no inputs are connected? So it relies on the XLR give the ground reference? I don't know that there is anything wrong with that but it seems a little odd to me.

Do you think the reason Topping uses a rail splitter instead of a +/- 32V adapter is for performance, squeezing out a little better SINAD? I find it hard to believe that using a single ended supply and rail splitter is cheaper than using just using a +/- 32V external supply.
 

solderdude

Grand Contributor
Joined
Jul 21, 2018
Messages
15,891
Likes
35,912
Location
The Neitherlands
So do I understand correctly that this amplifier is completely floating when no inputs are connected? So it relies on the XLR give the ground reference? I don't know that there is anything wrong with that but it seems a little odd to me.

No it is referenced by the rail splitter.

Do you think the reason Topping uses a rail splitter instead of a +/- 32V adapter is for performance, squeezing out a little better SINAD? I find it hard to believe that using a single ended supply and rail splitter is cheaper than using just using a +/- 32V external supply.

Perhaps it has do with availability of a 'standard' single 240W/64V adapter they bought from 'JLY Power'.
+/- 32V would be a custom one and much more expensive.

As mentioned a few times... we don't know. Just guessing. Only a teardown or JohnYang can tell us.
 
Last edited:

Manojrc

Member
Joined
Oct 4, 2020
Messages
47
Likes
13
Given that noise and distortion isn't an issue on any of the top 15 amps tested I'm wondering what the point of it is? How big is the market for modestly powered ultra-fi?

I guess we're about to find out.
Still an overkill for my speakers which has a sensitivity of 100 dB !!!
 
Top Bottom