Topping PA5 Review (Amplifier)

[nagster]

I read that, but don’t understand amplifier topology enough to understand the issue. I have been feeding my PA5 an SE signal for months now and have no issues with it. Given that, is there a reason to upgrade the DAC/Preamp?

I used the D90SE to compare the sounds of the RCA~TS connection and the XLR~TRS connection, but I couldn't tell the difference.

However, if you get a high performance balanced output DAC / balanced Preamp, you will get a maximum output increase of about 15% and a reduction of 0.0001% in THD+N ratio.

Topping PA5 Review (Amplifier)

Useless as you do not know the input impedance of the amp and what blocking cap is in these. Amps run from 10 to 50K typically. You can get a simple crossover from folks like Rolls. Not as bad as it looks. You can DIY a really good crossover like the X-Kitx. I agree with your want. A really...

www.audiosciencereview.com

 

[Deleted member 46664]

If the PA5 is fully differential, as it appears to be (I'm not aware of anyone tracing the circuit, including the potted module), it will perform more like the TPA3251 spec sheet states for an SE connection, as opposed to BTL (the PA5) or PBTL (the 4 amplifiers bridged into a mono configuration). The results are not "terrible" distortion, but it is poorer (about .006 thd at 5w per the spec sheet) compared to (about .002 thd at 5w per the spec sheet). The power is also less, as others have explained.

Okay there are two kinds of "balanced audio" ...

The right way ...

And the wrong way...

In the first case, using a proper differentiator on the input, as in the schematic segment I posted earlier, you can operate it either single ended or balanced with only a minor loss in level when single ended. In the second case, "fully balanced", grounding the cold signal shuts down half of the amplifier and might even do some damage.

Before I'm going to spend $459 (Cdn) I want to know it's safe to use single ended until I update the rest of my system.
Of course there is the parallel concern about the power supply also running in these threads...

Maybe I should just take a pass on this one...

 

[HansHolland]

Okay there are two kinds of "balanced audio" ...

The right way ...
View attachment 200508

And the wrong way...
View attachment 200509

In the first case, using a proper differentiator on the input, as in the schematic segment I posted earlier, you can operate it either single ended or balanced with only a minor loss in level when single ended. In the second case, "fully balanced", grounding the cold signal shuts down half of the amplifier and might even do some damage.

Before I'm going to spend $459 (Cdn) I want to know it's safe to use single ended until I update the rest of my system.
Of course there is the parallel concern about the power supply also running in these threads...

Maybe I should just take a pass on this one...

Pages back someone showed osciloscoop pictures when driven single ended. From that I concluded that this amp is balanced in "the wrong way" (as you named it).

If driving this amp single ended, means shorting the negative input. The result will be 0 (zero) Volt at the output. The feedback will keep the output at 0 Volt, also when current is going through it.
Or you can see it as that the negative path tries to be a wire. Then, why not use a wire (= a single ended amp).

If driven balanced, both outputs have voltage swing (opposite from each other). So output voltage is double.

I don't expect problems when driving this amp single ended, but why should you buy it when using it single ended (the measurements will be less good then the ones presented here which were done in balanced mode).

edit: the original poster of the pictures (nagster) explained them on page 214, I was wrong on the interpretation

 

[Deleted member 46664]

Pages back someone showed osciloscoop pictures when driven single ended. From that I concluded that this amp is balanced in "the wrong way" (as you named it).

Went back nearly a hundred pages and didn't find them ... I'll look more tomorrow.

The reason the fully duplicated channels is "the wrong way" is that first, you don't get the noise cancellation feature of the discriminator on the input and second because, as I pointed out shutting down half the electronics for SE use is just bound to cause problems.

Most of the pro stuff I've worked on is the "right way" and plugging in a TS connector -or- using a modified XLR cable cases no problem whatsoever.

So, until I have a concise answer, I'm not spending the money. Good thing I caught this before hitting the Check Out button....

 

[KSTR]

Pages back someone showed osciloscoop pictures when driven single ended. From that I concluded that this amp is balanced in "the wrong way" (as you named it).

If driving this amp single ended, means shorting the negative input. The result will be 0 (zero) Volt at the output. The feedback will keep the output at 0 Volt, also when current is going through it.
Or you can see it as that the negative path tries to be a wire. Then, why not use a wire (= a single ended amp).

If driven balanced, both outputs have voltage swing (opposite from each other). So output voltage is double.

I don't expect problems when driving this amp single ended, but why should you buy it when using it single ended (the measurements will be less good then the ones presented here which were done in balanced mode).

I'd be glad, too, if you could point us to the post with the scope shot as I have missed as well.

I personally would find very hard to believe that this amp does not have a subtractor circuit at the input because that's just what balanced connections is all about: make no assumption about the signal other than that the difference between (+) and (-) pins is our payload signal -- hence, do not amplify anything else than this difference signal. And for a power amp with bridged output it is clear that you must make the drive signal for each bridge half the same except for polarity.

If the amp really is using two seperate paths without re-referencing (splitting) the power-stage input signals back to the local ground then this is a major design failure.

 

[Deleted member 46664]

If the amp really is using two seperate paths without re-referencing (splitting) the power-stage input signals back to the local ground then this is a major design failure.

I've seen it in other stuff ... "fully balanced" DACs for example. And yes it is a seriously bad idea.

 

[Deleted member 46664]

For those of a DIY bent, I'm attaching a PDF of a very simple Single Ended (RCA) to Balanced (XLR/TRS) converter that you can build and use with any balanced input amplifier. Generally it's best to install the converter at the RCA end, since that will give you the benefit of balanced audio's noise cancellation features for the longer run of your cables.

I've tossed in it's complimentry Balanced to Single converter as well.

 

[antcollinet]

Went back nearly a hundred pages and didn't find them ... I'll look more tomorrow.

I've just checked the first 100 pages - and couldn't find it...

 

[Deleted member 46664]

I've just checked the first 100 pages - and couldn't find it...

So it must be somewhere between 101 and 190...

 

[KSTR]

I've seen it in other stuff ... "fully balanced" DACs for example. And yes it is a seriously bad idea.

Yes, there's a ton of (mostly) boutique stuff that violates all common engineering practice. Plus some cheap stuff as well (like the Topping D10 Balanced where they "forgot" to remove the common-mode signal -- for a number of understandable reasons like keeping the 4Vrms@0dBFS target with only +-5V rails)

Single Ended (RCA) to Balanced (XLR/TRS) converter

Even simple, we don't need signal-balancing for a balanced input (unless it's broken like allegedly in the PA5). Just make the interface 3-wire (at the sending end) so that the shield current's voltage drop is not part of the signal anymore... and add the impedance balancing resistor on the cold end wich might improve SNR with really large runs of cable.

 

[nsfgp]

Post #4181 just above you by Nagster linked to his post with the scope pictures; may be the one you all are looking for.

 

[KSTR]

So it must be somewhere between 101 and 190...

Pity that the forum search function is pretty useless... it could have been so easy with a button "search posts with attachments only" or similar. Meh...

 

[Deleted member 46664]

Even simple, we don't need signal-balancing for a balanced input (unless it's broken like allegedly in the PA5). Just make the interface 3-wire (at the sending end) so that the shield current's voltage drop is not part of the signal anymore... and add the impedance balancing resistor on the cold end wich might improve SNR with really large runs of cable.

That's why I thought to upload this ... if the PA5 is broken... there's an easy fix.

 

[antcollinet]

Post #4181 just above you by Nagster linked to his post with the scope pictures; may be the one you all are looking for.

Damnit - that would have saved me about half an hour this morning if I'd spotted it - so on page 158.

Now bookmarked

@Douglas Blake

 

[KSTR]

Post #4181 just above you by Nagster linked to his post with the scope pictures; may be the one you all are looking for.

That settles it, thanks.
==> broken. Sad, I had not expected that Topping would do such sloppy engineering even though the 4-gang volume pot already pointed to the outcome, in hindsight.

I mean, what's the cost and effort of one dual opamp per channel to re-balance the drive to the local ground? I don't get it... perhaps a move to justify better and more expensive follow-up products?

 

[antcollinet]

For those of a DIY bent, I'm attaching a PDF of a very simple Single Ended (RCA) to Balanced (XLR/TRS) converter that you can build and use with any balanced input amplifier. Generally it's best to install the converter at the RCA end, since that will give you the benefit of balanced audio's noise cancellation features for the longer run of your cables.

I've tossed in it's complimentry Balanced to Single converter as well.

There are also low cost pre made modules that just need a PSU adding. Such as:

SparkFun THAT 1646 OutSmarts Breakout - BOB-14003 - SparkFun Electronics

The THAT 1646 OutSmarts Breakout Board offers an easy solution to adding a balanced audio output to your circuits. The THAT OutSmarts technology has been d

www.sparkfun.com

 

[Deleted member 46664]

Post #4181 just above you by Nagster linked to his post with the scope pictures; may be the one you all are looking for.

Thank you!

 

[Deleted member 46664]

There are also low cost pre made modules that just need a PSU adding. Such as:

SparkFun THAT 1646 OutSmarts Breakout - BOB-14003 - SparkFun Electronics

The THAT 1646 OutSmarts Breakout Board offers an easy solution to adding a balanced audio output to your circuits. The THAT OutSmarts technology has been d

www.sparkfun.com

Cool ... Except that I didn't design that one (LOL)

 

[Deleted member 46664]

That settles it, thanks.
==> broken. Sad, I had not expected that Topping would do such sloppy engineering even though the 4-gang volume pot already pointed to the outcome, in hindsight.

Sad indeed ... it shure did measure nicely with balanced inputs... but it just saved me $450 (Cdn).

 

[iLoveCats]

You guys could probably talk each other into it being the soups fault that you are trying to eat it with a fork.