Topping E50 impedance + new way to use?

[mike7877]

Consider the E70 V, L70, and A70 Pro -all have line level SE+BAL outputs on the rear, just like the E50:
E70 Velvet - SE output impedance is 50 ohms, balanced: 100.
L70 (preamp)'s SE/BAL (RCA/XLR) 20 and 40 ohms
A70 Pro (pre)'s SE/BAL (RCA/XLR): 50 and 100 ohms

When the output impedance of a balanced output is twice its single ended, what can be safely assumed is: internally, the output stage used for SE was also been used for BAL.

When there is also a 6dB improvement in dynamic range on the balanced output, you can be almost entirely sure about your assumption that the output stage is identical.

There are other things which can be looked at (slight but not insignificant improvement to THD+n for one..), but with +6dB dynamic range and exactly double output impedance in place, the chance that all stages aren't identical is slim to none.


Earlier this evening, when, for convenience, I went to use my DT990s with my D50 by way of a really weird RCA (male) to 3.5mm (female) y-shaped adapter. And wow, was I surprised - very surprised! Pleasantly surprised... By the sound!
It was unexpectedly loud and clear, with no immediately apparent bump in the bass response. Due to a brain fart (where I forgot DT990s are 250 ohms), I thought I was plugging in 35 ohm headphones into a 50 ohm output... I immediately remembered the DT990s were 250 ohms, but that wasn't quite enough to explain the decent bass performance! So I searched for the L50 manual, opened it, and found this:

In the last row is "Output Impedance", and it's 20 ohms for SE (RCA).

With a 250 ohm headphone attached, the damping factor calculates to 12.5 (250/20). 12.5 isn't perfect, but it's better than a lot of situations which sound perfectly fine. 20 is usually good enough for high fidelity, and 40 is flawless. Some headphones actually start to sound thin much over 15-20, so higher isn't always better!


At first I thought 20 ohms was a typo, but then I saw that the dynamic range wasn't the required 6dB. If it was 5dB I'd have accepted it as a typo, maybe even 4. But 2? ... no, I can't. Especially with bandwidth being identical! Of course, by itself the 20Hz-40kHz bandwidth means nothing, but with just +2dB DR? What is up with this! What's the point even? Topping saving a bit of money using $9 op amps instead of $12? What's the deal?

Also, I'm going to try some critical listening, comparing using the E50's RCA outs directly to headphones, to amplifying its RCA outs with an L30 II

 

[Vladetz]

You should not connect headphones to dac directly

 

[ZolaIII]

Which headphones? DT990 Pro? Those almost don't show any source impedance dependance. Better say it's so small that it doesn't really matter, brain does know to do that to you. It's not DR it's V and it's V*2=+6 dB and those are 102 dB per 1V so standard 2V unbalanced DAC output is pretty good choice to drive them with performance delta at - 12 dB 500 mV and having enough peek headroom. That's for very loud 90 dB program. Anyway goal is that multitone SINAD stays above possible dB at desired SPL. It gets hard sometimes with very efficient low impedance high SPL headphones and IEM's for standard even state of art DAC amp combos. It's better to use purposely designed output stage with such like mobile phone one or dongle for such usually not giving more than 500 mV max output to begin with and good noise performance of course.
Elementary logic really if something is 102 dB per V and goes great with 2V max and something else is 114+ dB per V it's better to start with something that can give - 12 dB to begin with. Of course you won't have very good crosstalk on low impedance loads.

 

[Dunring]

You should not connect headphones to dac directly

Yes, I've seen burned out DACs on eBay where they were including RCA to 3.5mm cables where they were plugging them direct into a DAC and said it suddenly stopped working. I think it's a really risky thing to do.

 

[formdissolve]

Came across this thread and want to pile on another warning, straight from the horse's mouth:

 

[mike7877]

Which headphones? DT990 Pro? Those almost don't show any source impedance dependance. Better say it's so small that it doesn't really matter, brain does know to do that to you. It's not DR it's V and it's V*2=+6 dB and those are 102 dB per 1V so standard 2V unbalanced DAC output is pretty good choice to drive them with performance delta at - 12 dB 500 mV and having enough peek headroom. That's for very loud 90 dB program. Anyway goal is that multitone SINAD stays above possible dB at desired SPL. It gets hard sometimes with very efficient low impedance high SPL headphones and IEM's for standard even state of art DAC amp combos. It's better to use purposely designed output stage with such like mobile phone one or dongle for such usually not giving more than 500 mV max output to begin with and good noise performance of course.
Elementary logic really if something is 102 dB per V and goes great with 2V max and something else is 114+ dB per V it's better to start with something that can give - 12 dB to begin with. Of course you won't have very good crosstalk on low impedance loads.

Ah yes, the crosstalk - completely slipped my mind. I don't know what happened to this thread, I guess I never got notifications or something

Came across this thread and want to pile on another warning, straight from the horse's mouth:

View attachment 376767

I thought about this a bit more - I think this is good general advice (telling people to not use headphones), but considering this device is advertised to be used as a preamp as well as a source (reaffirmed by the built in setting to switch from DAC to preamp mode), I think higher impedance headphones (especially) should be safe. I have an amplifier with a 1k ohm input impedance, and 600 isn't unheard of. They're not extremely common (especially new amps today), but I'm pretty sure stuff is still designed to be able to drive down to at least 600 ohms. Datasheets of all the common audio op amps are spec'd at 600 and 2k ohms. Also notable is the maximum instantaneous output current before damage to op-amps is usually 2x the continuous rating. Since Topping used a 20 ohm resistor on the output, and one of its purposes is temporary short circuit protection, I think any total impedance down to 40 ohms {20(R)+20(HP)} should be pretty far away from doing any actual damage to the output. Although traditional headphones (what's being discussed) are reactive, they aren't ported like loudspeakers, and they don't have crossovers either, so I think adding another 10 ohms for worst case reactivity and 10 more again for good measure, for a total of 60 (40 headphones) should be a pretty safe bet. Would sound quality suffer? Possibly, but probably not - and if there is an audible problem, I'd guess it would probably be small and only with low impedance cans (less than 80 ohm). Actually, since I tested my 1440s and there wasn't audible distortion I shouldn't say guess.
I could be missing some aspect but I don't think I am

 

[rcstevensonaz]

Came across this thread and want to pile on another warning, straight from the horse's mouth:

View attachment 376767

That photo only shows that headphones should not be connected to the TRS (BAL) output. Does that imply that it is OK to attach headphones directly to the RCA (SE) output?

 

[staticV3]

That photo only shows that headphones should not be connected to the TRS (BAL) output. Does that imply that it is OK to attach headphones directly to the RCA (SE) output?

No headphone support is explicitly stated for the 1/4" TRS Line out because many headphones come with a 1/4" TRS cable.

For the RCA output, lack of headphone support is implied by the non-headphone connector.

Just to be clear: you shouldn't plug any headphone into any Line out.