Topping Pre90 Review (preamplifier)

[phoenixsong]

What is ADC ?

Analogue to digital converter. It's for when you're trying to capture raw sound with microphones into your laptop, for example. So it does the opposite of what DACs do

 

[MaxwellsEq]

Well seeing as all the audio information has to pass through those chips the final sound quality can not be better than what's maximally achievable using them

That's not how electronic circuits work.

 

[Arci66]

Thanks for the great review and for the huge discussion.
I am evaluating to give a try to this preamp.
My chain is a Eversolo A8->Marantz MM7025-> SF Sonetto I
The preamp of the eversolo A8 doesen't look so energic; to have a sound level of about 80-85 dB the volume has to reach -20 dB at -50dB sound level is very weak.
The XLR output level of the eversolo A8 is 4.2 V
Marantz MM7025 XLR Input sensitivity/impedence: 2.4V/30KOhm

Do you think putting the Topping Pre90 in the chain can improve the performances of my system?
Thanks in advance for your time

 

[MaxwellsEq]

The preamp of the eversolo A8 doesen't look so energic; to have a sound level of about 80-85 dB the volume has to reach -20 dB at -50dB sound level is very weak

Those numbers sound fine to me. The numbers such as "-20dB" only means something if you have a calibrated end-to-end chain. If you turn the volume up to full I assume it's too loud...

 

[Arci66]

Those numbers sound fine to me. The numbers such as "-20dB" only means something if you have a calibrated end-to-end chain. If you turn the volume up to full I assume it's too loud...

Yes, my wife come shouting
So you mean there is no point to add a Topping Pre90?
Somebody measured the output impedence of eversolo A8 : The output impedance was usefully low across the audioband, at 101 ohms, balanced, and 51 ohms, unbalanced.
So on XLR 101 ohms the maratz has a input impedence of 30 kohm

 

[Veri]

The circuit is not "discrete" either its all based on the Texas Instruments (tpa6120a2) chip.
Do you people even own a screwdriver?

It is designed as the equivalent of a composite amplifier with nested feedback loops done entirely in a discrete way. You should read read mocenigo's post again, really..
Do you even have some electronic design credentials? Or just an enthusiastic DIY'er?

 

[ZestClub]

It is designed as the equivalent of a composite amplifier with nested feedback loops done entirely in a discrete way. You should read read mocenigo's post again, really..
Do you even have some electronic design credentials? Or just an enthusiastic DIY'er?

We were talking about the TI tpa6120a2 headphone amplifier chips function in the Pre90 at the time - and its not very "discrete" is it!

 

[ZestClub]

But all that "nested feedback loops done entirely in a discrete way" technology doesn't come cheap like you might think/
Those 20 year old headphone amp chips are way pricey you know.
Way way up there at the $2 mark - I mean, I am just amazed that Topping would put something so expensive in one of their preamps,

 

[DSJR]

We all hear differently - that may depend on your genetic ancestry how your brain processes sound - some people can't hear the difference between a 1970's NE5532 and others can clearly hear the muffled top end. I only listen to very recent recordings so no 5532's in my chain thanks!

MUFFLED TOP END???

Sorry to shout above, but my mental processes always regarded the 5532 as being 'stripped out' and 'bare' sounding That is, until I forgot what I was listening to and 'thought' it was something else. More modern chips in older circuits can apparently oscillate, I gather.

Anyway, I hear sometimes significant differences in musical productions, a delightful soundfield between and behind the speakers if there in the recordings and no longer ever feel the need to 'listen' to the gear and all this, from a preamp with a 5532 per channel with a volume pot in front of it. Recent times have changed my hearing in any case, so have to admit I'm so glad I'm no longer an 'audiophile/audiophool' as such.

 

[misterdog]

But all that "nested feedback loops done entirely in a discrete way" technology doesn't come cheap like you might think/
Those 20 year old headphone amp chips are way pricey you know.
Way way up there at the $2 mark.

Proven test data shows it to be more than adequate, and a bargain at it's price.

Maybe once you've built your preamp you can send it to Amir and show us how superior it is ?
Until such times all this 'I think' is just noise on the internet.

 

[ZestClub]

Proven test data shows it to be more than adequate, and a bargain at it's price.

Maybe once you've built your preamp you can send it to Amir and show us how superior it is ?
Until such times all this 'I think' is just noise on the internet.

So with 4 x OPA1612's and 2 x tpa6120a2 we are now at $16 of parts - not sure where the other hundreds of dollars go though...
Perhaps the volume knob was machined out of solid Palladium or something?

 

[staticV3]

@ZestClub I think it's about time you form a coherent and factual point of criticism if you want to actually contribute.

You've just been jumping from one easily debunked misconception to the next.

Complaining that some of the chips inside the Pre90 cost way less than the final MSRP does not help your case. It just comes across as you trolling for trolling's sake.

 

[LTig]

So with 4 x OPA1612's and 2 x tpa6120a2 we are now at $16 of parts - not sure where the other hundreds of dollars go though...
Perhaps the volume knob was machined out of solid Palladium or something?

You just demonstrated a total lack of understanding about industrial production and the market. There is about a factor of 6 to 8 (or much more) between material costs and consumer pricing.

 

[mocenigo]

But its all accurate though - the Topping Pre90 does use a 20 year old headphone chip amp (tpa6120a2) and OPA1612's which are Meh.
I am changing mine at the moment to a header so I can swap different Opamps in and out, so I will let you all know how it goes.
The extra (redundant) TSOP footprint under the OPA1612 (once its removed) can be seen in this photo as well.

May be old components, but if they are good what's the problem? I would actually say that anyone that claims "OPA1612's [] are Meh" has a problem. These are among the finest integrated components around.

 

[mocenigo]

Well seeing as all the audio information has to pass through those chips the final sound quality can not be better than what's maximally achievable using them.

Maybe you never heard of negative feedback, and what it can do. Or maybe you believe that it is always bad?

 

[Mikig]

I find it hard to believe that an op-amp, or rather its reputation, can permanently ruin your listening experience…

If you think about your entire system, the listening room, the curtains or cushions or the floor of the listening room, your ears, your perception and your taste, the electrical system in your home, the recording, how it was obtained, with what instruments, how it was mixed, how it is reproduced, the temperature of the recording and listening environment, the right position of the speakers in the room, the rack or furniture between the speakers, etc. etc.,

 

[mocenigo]

It is designed as the equivalent of a composite amplifier with nested feedback loops done entirely in a discrete way. You should read read mocenigo's post again, really..
Do you even have some electronic design credentials? Or just an enthusiastic DIY'er?

Actually, I have to correct myself. There are at least 3 different version of the Topping NFCA circuit and I am not 100% sure which product uses the discrete version. But not the Pre90.

This one should be a composite amplifier with the TPA6120 inside the feedback loop of the OPA1612, with the TPA6120 providing the power and the OPA1612 the precision. This is something well understood, and the results speak for themselves. However, the passive components are there to deal with little quirks of the used ICs and depend a lot on stuff like phase rotations to guarantee stability. Replacing the ICs is most likely a recipe for instability and thus disaster.

Even Stereophile put the Pre90 in their Class "A" preamps, in other words in the same league with products costing 35K-50K USD and more.

 

[ZestClub]

Actually, I have to correct myself. There are at least 3 different version of the Topping NFCA circuit and I am not 100% sure which product uses the discrete version. But not the Pre90.

This one should be a composite amplifier with the TPA6120 inside the feedback loop of the OPA1612, with the TPA6120 providing the power and the OPA1612 the precision. This is something well understood, and the results speak for themselves. However, the passive components are there to deal with little quirks of the used ICs and depend a lot on stuff like phase rotations to guarantee stability. Replacing the ICs is most likely a recipe for instability and thus disaster.

Even Stereophile put the Pre90 in their Class "A" preamps, in other words in the same league with products costing 35K-50K USD and more.

You just demonstrated a total lack of understanding about industrial production and the market. There is about a factor of 6 to 8 (or much more) between material costs and consumer pricing.

I have been in electronics for 40+ years so thats my job (not audio though) - I know electronics production very well and the ratio here of part cost to product cost is obscene. Essentially this is a $35 Aliexpress board with a nice case, a display and some relays/ resistors instead of a Pot.

And while we are on the subject why have two obviously different batches of OPA1612 been used on the boards (2 x OPA1612 batch 1) next to the tpa, and (2 x OPA1612 batch 2) at the input ? The second batch has very faint printing on it not like a standard TI part at all (unlike part 1).
Why would the assembly subcontractors load 2 different batches of the exact same part into the placement machine, and place them in set positions ?
Could that indicate that there is some "issue" with batch 2 ?

 

[ZestClub]

May be old components, but if they are good what's the problem? I would actually say that anyone that claims "OPA1612's [] are Meh" has a problem. These are among the finest integrated components around.

Do you think TI doesn't know the market value of its products?

When OPA627's cost $30 for one single opamp meaning to replace a dual opamp the cost is $60 - so $240 to replace the 4 x OPA1612's on the Pre90.
Yet they wholesale the OPA1612 out at $3.00 and the tpa6120a2 out at $2 - and believe me if buying on a call-off from a tier 1 distributor those costs will be even lower - do you really believe that those parts are so great and so in demand that they would let them go at those bargain bin prices?

 

[mocenigo]

I have been in electronics for 40+ years so thats my job (not audio though) - I know electronics production very well and the ratio here of part cost to product cost is obscene. Essentially this is a $35 Aliexpress board with a nice case, a display and some relays/ resistors instead of a Pot.

Well, casework also has a cost, all the other components, connectors, etc. they have to maintain inventory.
They have to pay the person that designs it, and all the other employees.
But apparently you have been working in electronics for 40+ years for free?

And while we are on the subject why have two obviously different batches of OPA1612 been used on the boards (2 x OPA1612 batch 1) next to the tpa, and (2 x OPA1612 batch 2) at the input ? The second batch has very faint printing on it not like a standard TI part at all (unlike part 1).
Why would the assembly subcontractors load 2 different batches of the exact same part into the placement machine, and place them in set positions ?
Could that indicate that there is some "issue" with batch 2 ?

If some of the opamps had issues then this would show in measurements? But, again, this is one of the most transparent preamplifiers ever designed.
You are really scraping the bottom of the barrel…