WolfX700 Measurements of Topping Pre90/Ext90 Preamplifier/ Extender

[JohnYang1997]

Sorry on this one, you're mistaken. 2K is ridiculous and you know it, John. You've gone after one metric at the expense of source compatibility and performance. It's a mistake to pass the buck to the source components. A preamplifier is designed to take varying source impedances and levels and buffer them without messing up the performance of the attached components.

47-50K is the defacto input impedance for line sources into a typical preamplifier/integrated line input. It's also the defacto input impedance of an MM RIAA stage and has been this way for many, many decades. Go peruse a few hundred service manuals for amplifiers and preamplifiers and tell me what the "standard" is. I can tell you, it's closer to 50K than 2K.

20-25K was favoured by pro power amps input impedances in the 1980s.

There's no standard. And it works fine with majority of devices. Also it's only the balanced input that has 2kohm. Single ended is 10k which is what majority of modern devices use.
Also there's simply no reason for me to design for every use case possible. It's meaningless. And it's not beneficial.

 

[YSC]

One question is mostly ppl use the pre amp as volume control, interesting to know what the A90 pre SINAD at it's lowest 10-15% setting will be

 

[restorer-john]

There's no standard. And it works fine with majority of devices.

As I said, and you need to understand, 47K-50K has been the defacto standard for many decades. Defacto means it comes into being through use, adherence and general agreement. One guy standing on the sidelines saying 2K is fine isn't going to change a thing.

Literally hundreds of millions of CD players all specified 2.0V out with load impedances required of over 10K minimum to ensure perfectly flat responses and lowest distortion. Tuners require 20-50K. Phono stages need 10K or greater. DVD players need the same.

There is no debate. 2K ohms is ridiculously low and when you've made a few more preamplifiers, you'll understand.

And I like the look of the preamplifier and the input extender. The display looks a bit blocky in the pictures but likely it looks better in the flesh.

 

[Max]

From my point of view, none doubt @JohnYang1997 has made any mistake. He's a skilled electronic engineer and has surely fullfilled the specifications of the product. I know well the possible limitations of a (here relatively) low input impedance but the main question is the market target of the company. Who is the typical buyer of Topping's product ? The "classical" Hifist with bunch of (often less performing) old gears ? I don't think so. Maybe the current typical Hifist isn't anymore the same decades ago, because Topping (and other manufacturers) has opened the market to new customers. Is the 10k RCA input impedance so low it becomes ridiculous ? I don't think so. Is the 2k XLR input relatively low to speak about ? Yes, but so, don't forget it is a tradeoff made for a current changing market. Topping, as company, has certainly made a good choice. Just my opinion

 

[JohnYang1997]

As I said, and you need to understand, 47K-50K has been the defacto standard for many decades. Defacto means it comes into being through use, adherence and general agreement. One guy standing on the sidelines saying 2K is fine isn't going to change a thing.

Literally hundreds of millions of CD players all specified 2.0V out with load impedances required of over 10K minimum to ensure perfectly flat responses and lowest distortion. Tuners require 20-50K. Phono stages need 10K or greater. DVD players need the same.

There is no debate. 2K ohms is ridiculously low and when you've made a few more preamplifiers, you'll understand.

And I like the look of the preamplifier and the input extender. The display looks a bit blocky in the pictures but likely it looks better in the flesh.

RCA in is 10K.

 

[JohnYang1997]

One question is mostly ppl use the pre amp as volume control, interesting to know what the A90 pre SINAD at it's lowest 10-15% setting will be

At those level there sure isn't distortion contribution. All is noise. So 50mV SNR measured by Amir mostly described what you are asking for.

 

[YSC]

At those level there sure isn't distortion contribution. All is noise. So 50mV SNR measured by Amir mostly described what you are asking for.

IC, seems the SNR still got like 108dB of SNR, may I ask if the relay is located before or after the gain stage to attain this level? I think the noise floor in the dashboard is like -150dB, so 50mV and SNR still retaining 108 dB is impressive

 

[Michal]

I found a nice DIY project based on Baxandall's topology. Input impedance 100k. Measurements on pictures.
How does this relate to the discussed Pre.

 

[Joaquinín]

Tuners require 20-50K.

RCA in is 10K.

My Accuphase T-101 tuner has 200 ohm stated output impedance (RCA out). Isn´t that low enough to work perfectly with this new preamp?

 

[MusicNBeer]

My Accuphase T-101 tuner has 200 ohm stated output impedance (RCA out). Isn´t that low enough to work perfectly with this new preamp?

That will wotk fine.

 

[MusicNBeer]

I'd add a user option to balanced in to select between 2K and 50K input impedance. 2K is ultra performance mode for robust sources. 50K is legacy setting. Relay switch the circuits. RCA I'd probably just set to 50K.

 

[youduoliang]

Where is the square wave response? It won't sound good if it doesn't have perfect square wave response

 

[gvl]

Who is the typical buyer of Topping's product ? The "classical" Hifist with bunch of (often less performing) old gears ? I don't think so.

The thought that the typical buyer needs this to switch between different Topping DACs is flawed IMO. What other use cases are there? Topping should have done more marketing research instead of trusting engineers to make these decisions. This stupid low input impedance will spook a lot of potential customers and the fact that Schiit Freya S is no longer indicates that the preamp market is already thin.

 

[AnalogSteph]

The choice of a 2 kOhm input impedance for a line-level XLR input is weird indeed. That would be a typical value for a mic input (which are often 2-3k). At line level, 20k would seem to be a more typical value in my experience. I could still see 10k or 6.6k working OK when driven by halfway modern opamps, but 2k would require parts with quite a bit of oomph, particularly at high levels.

It's not exactly good for practical input CMRR either, which is going to be an order of magnitude more sensitive to source impedance imbalance. I presume the input stage is a standard 1-opamp bal / SE converter topology? Consider upgrading to the 3-opamp buffered version then, with 1 Meg input bias resistors for the buffers (given the unity gain application, I'd select something with superb common-mode linearity for those, maybe some of the CMOS input OPA16xx).

 

[VeerK]

I know about Topping's power amplifier development plan-a design that is approximately twice the height of the Pre90, the same width and slightly deeper.

In fact, I also took the time to give a suggested 3D design of the appearance (hope that Topping can finally adopt this design)

I am also looking forward to this product...

I hope they consider potential buyers that would be matching the D90, A90/Pre90 with the power amp, at this price point aesthetics matter as well as functionality.

 

[JohnYang1997]

The choice of a 2 kOhm input impedance for a line-level XLR input is weird indeed. That would be a typical value for a mic input (which are often 2-3k). At line level, 20k would seem to be a more typical value in my experience. I could still see 10k or 6.6k working OK when driven by halfway modern opamps, but 2k would require parts with quite a bit of oomph, particularly at high levels.

It's not exactly good for practical input CMRR either, which is going to be an order of magnitude more sensitive to source impedance imbalance. I presume the input stage is a standard 1-opamp bal / SE converter topology? Consider upgrading to the 3-opamp buffered version then, with 1 Meg input bias resistors for the buffers (given the unity gain application, I'd select something with superb common-mode linearity for those, maybe some of the CMOS input OPA16xx).

5532 4562 1612 all having no issue driving 600Ohm load
Then that means adding more relays when input disconnected, changing different opamps and add more opamps for worse performance. While there's zero issue pairing with majority of new designs with current configuration....

 

[JohnYang1997]

The only devices that may have issues are the ones with small DC blocking caps at balanced output.
The ones that don't have caps will have zero issue no matter what output impedance is.
The ones that have 100uF or higher electrolytic caps will have zero issue.
Output opamp is not a concern, jellybean 5532 will drive 600ohm with zero issue. Let alone current lm4562, opa1612 etc.

 

[restorer-john]

5532 4562 1612 all having no issue driving 600Ohm load

True, and there's plenty of 600 ohm transformers used for balanced XLRs in TOTL digital gear from the past.

 

[Cortes]

There are these passive Stepped attenuators in US
https://goldpt.com/sa2x.html

which look imposossible to beat. No channel imbalance. Better volume control than expensive professional DACs.
https://www.superbestaudiofriends.o...ssive-attenuator-technical-measurements.7303/

 

[ririt]

More and more details about this preamp...however still nothing about a launching date