SOLD: MODDED Topping DX7s - No ESS "Hump" & low headphone output impedance

[axismundi]

For sale Topping DX7s dual ESS Sabre 9038Q2M DAC and headphone amplifier, well taken care of and with little use, in excellent condition (almost as new). The DAC comes with original box and accessories.

The DAC section has been modified in order to address the ESS IMD "hump" Results after modification were measured with dScope audio analyzer and show that the "hump" has been eliminated (see attached). The amplifier section has also been improved, now output impedance for single-ended output is below 0.4ohm (instead of 10ohm) / below 0.8ohm for balanced output (instead of 20ohm).

Included:
- DAC
- Power cable
- USB cable
- 6.25mm to 3.5mm headphone adapter

Asking 450 EUR+shipping. Shipping to EU should cost around 25 EUR, for other destinations feel free to drop me a note. Bank transfer preferred, Paypal accepted (add 3.5% to cover Paypal fees).

The item will be carefully packed and tracking number will be provided. Shipping out of France.

 

[Pillars]

Do you have a PCB shot on this?

 

[axismundi]

Sorry for the late reply, did not get any notification...

I did not take any shots of the PCB. But work is immaculate, I have decades of experience on SMD rework (for much smaller footprints than what is used in DX7s) and I am rather on the OCD side when it comes to how I treat the boards I work on. You'd not be able to spot any difference vs. how the board looked like originally.

 

[tecnogadget]

Then you can help us to understand what causes that ESS “hump” and in the meantime lets hope manufacturers catch up !

 

[axismundi]

Topping did catch-up, at least with some of the newer models (D50s vs. D50) and they are not the only ones. There is a dedicated thread here covering the causes of the IMD issue (https://www.audiosciencereview.com/forum/index.php?threads/ess-thd-‘hump’-investigation.5752/).

 

[mateoJazz]

Can you describe the modifications with the photos of the changes and put in the diy section?

 

[axismundi]

As said, no pictures were taken and the causes of the IMD issue have been already described elsewhere.

Let's keep this a classifieds entry please, feel free to PM me for matters not related to this sale.

 

[Jimmy]

I think that what other users are implying is that you are asking a higher price than what a new unit costs, in Audiophonics 399€ + shipping with a 2 year warranty, so for trust reasons and to show other users your work it would be advisable.

 

[axismundi]

I think that what other users are implying is that you are asking a higher price than what a new unit costs, in Audiophonics 399€ + shipping with a 2 year warranty, so for trust reasons and to show other users your work it would be advisable.

Whether that's what (all) users that posted before had in mind is open to debate, however it is a fair point to raise.

To that, my response is:
1. This is not snake oil, ESS IMD issue is a well documented one and solutions to alleviate this do exist. I have directed the interested users to one of the threads dealing with this topic.
2. Measurements with professional audio analyzer have been included in the listing and I think they speak for themselves. Should that not be trustworthy enough, I am not sure anything else could change that.
3. For what is worth, Audiophonics does not have DX7s in stock. Based on a quick search, all the retailers that seem to have this model in stock sell it at a price higher than what I am asking for. And that is, of course, for an inferior product.

 

[Jimmy]

Audiophonics has the silver version in stock for 399€+shipping, and what you are asking for is blind faith, without giving any details of your mods (components used, methodology, etc.), for example lowering output impedance for TPA6120 can be achieved just by lowering output resistors value, but that's not what TI recommends and can cause instability.

You may be lucky and sell it, but the only way to clear the air and create trust is by showing PCB pictures and the concept behind.

 

[06S2k07Si]

I'm going to wholeheartedly agree. This website's main focus is objectivity and science and or or proof...the whole point of this site. The difficult thing is that you haven't shown PROOF of what you did other than one graph. We are not saying that you are lying but an explanation of everything such as components used or modifications made (as said above) would be beneficial to your sale.

Regarding your point #1. Most people who are interested in the DX7s are aware of this especially if they are on this site.
Regarding your point #2. Anyone can copy a graph like that from anywhere and photoshop something like it. ...and yes we are sure something CAN change this and it was already said above...pictures would help greatly.

 

[axismundi]

Jimmy/06S2k07Si, I get your points and your are entitled not to take things at face value. That's entirely fine with me.

That being said, if I was to go that far and forge IMD graphs I do not see how PCB pics would change things - those could be forged as well. Or for that matter, providing technical details regarding the mods - I could simply rephrase what one informed reader can already find online (including in the thread I linked above). Maybe I miss something here but I doubt either would provide irrefutable proof. As a matter of fact, if I was trying to misrepresent facts here I would actually be better off as of this stage playing along.

The simple fact is that the mod was done for my listening (and tinkering) pleasure. As such, while plenty of measurements were performed in the process, no PCB pictures were taken. Happy to share more measurements should that be of interest, however as misguided and as daring as you may find this, I am afraid I do not intend to re-open the DAC. Should this be considered a punishable sin by some, I can only provide my deepest apologies.

@ Jimmy, default configuration for TPA6120 includes an output resistor in order to isolate capacitive loads. Plenty of solutions to circumvent that in an elegant way without causing instability - again, info available in the public domain. I will not pretend I came up with the solution myself (in the same way I did not pretend I found the IMD fix either), other people smarter than me did it already.

Also @ Jimmy, my bad, had missed that the silver one is currently in stock, was not intentional. However, that's not even the point and I do not have the slightest intention to dispute one's options and purchasing choices. I am absolutely sure that not many people would choose this DAC over the standard version and that's absolutely fine and that's how it should be.

 

[06S2k07Si]

Thank you for your explanation. To be clear I hold no I'll will towards you. I do think it's important to understand that I believe people here tend to desire details and facts etc to satisfy them...one main reason I like this site. When some users see something that perhaps does not align with this it is in turn questioned.

 

[JohnYang1997]

As you didn't come up with the solution. You may at least reference to the original exact posts. Asking for premium from others work is another issue on its own.

 

[AudioJester]

As you didn't come up with the solution. You may at least reference to the original exact posts. Asking for premium from others work is another issue on its own.

If it was shared in public, then the OP still "did the work" and can charge what he wants for that? Unless you are suggesting IP has been used inappropriately??

 

[JohnYang1997]

If it was shared in public, then the OP still "did the work" and can charge what he wants for that? Unless you are suggesting IP has been used inappropriately??

1, He should tell what exactly is modified, especially the 6120 part.
2, Isn't this falling into commercial use? One should at least ask for permission for that.

 

[AudioJester]

1, He should tell what exactly is modified, especially the 6120 part.
2, Isn't this falling into commercial use? One should at least ask for permission for that.

Yes, I understand what you are saying.

 

[solderdude]

One should realize that for a LOT of headphones the 10 Ohm output R is no problem at all.
Another thing is that I haven't seen any proof by blind listening tests that the IMD hump is audible.

Solving the IMD hump is just replacing a few resistor and capacitors which can be done quite neatly.
And even if there were pictures chances are the SMD caps used will not have any markings on them anyway.

So ... it is an individuals choice to mod it or not and to sell it at whatever price they want and buyers don't need to buy it.

 

[A800]

Another thing is that I haven't seen any proof by blind listening tests that the IMD hump is audible.

Maybe one will actually miss it.

 

[axismundi]

One should realize that for a LOT of headphones the 10 Ohm output R is no problem at all.
.

Agree, with some qualifications:
High impedance and linear with frequency - should be fine. Many headphones fall in this category.
Low impedance and linear with frequency - volume attenuation due to resistor divider (output impedance/earphones impedance) means quite a bit of power is wasted. Not a problem with DX7s, there is plenty of power available.
Non-linear impedance - pretty much an issue in my book as frequency response is affected. BA IEMs and in particular multiple BA IEMs/hybrids would typically suffer from this. As it happens, I mostly use IEMs (and most of them are hybrids and some have absolutely wild impedance swings), so for my usage scenarios high output impedance is a no-no.
So I guess it really boils down to one's collection of headphones. As a side note, in this day and age I'd be hard-pressed to find a justification of not targeting a low output impedance in a headphone amp since this - all things being equal - gives you the freedom to plug in whatever headphones you may have around without having to worry.

Lastly and purely from an academic point of view, in all the cases above damping of the membrane movements would be less optimal with high Z out but in practice I am not sure it makes a perceptible difference.

Another thing is that I haven't seen any proof by blind listening tests that the IMD hump is audible.

Agree, haven't seen either. Now the frequency range where this manifests itself is pretty unfortunate, but equally people are swearing by vinyls and tubes so not sure that many would be able to spot the difference in a blind test. As for my own experience, to me DX7s sounded just fine before and sounds just fine now - I cannot confidently say I perceived a difference after the IMD fix but in the same time I did not attempt a blind test (which in practice would require two devices anyway, so that's that).