Tear down of Massdrop THX AAA 789 Headphone Amp

[RenEH]

You mean they aren't in the Amp? Why would you want to add additional things to the Amp?


Nice setup!

It's on the PCB mate.

 

[RenEH]

It could also be because the switch mode power brick used with the final design turned out to be unhappy with capacitive loading. Who knows...

That's cools I've moved on to a mean well anyways. So any harm in throwing in these smoothing caps?

 

[Killingbeans]

Hard to tell.

It might trigger a short circuit protection in the Mean Well and make it refuse to power up, or it might not.

It might cause an instability of some sort, or it might not. That instability could be audible, or it could be measurable only.

I think it's safest to assume that the caps have been left out for a reason, and stop trying to improve a design that's already impeccable in any practical sense

 

[RenEH]

Hard to tell.

It might trigger a short circuit protection in the Mean Well and make it refuse to power up, or it might not.

It might cause an instability of some sort, or it might not. That instability could be audible, or it could be measurable only.

I think it's safest to assume that the caps have been left out for a reason, and stop trying to improve a design that's already impeccable in any practical sense

Sounds pretty safe to me. What’s a $1 in caps and 5 minutes of soldering anyways.

 

[YSC]

Sounds pretty safe to me. What’s a $1 in caps and 5 minutes of soldering anyways.

I would not try to tinker with that anyway, if it ain't broken don't mess with it, sometimes the PCB are for various models and some combinations will better have it or not, and main thing is the amp is basically as transparent as it could get, there's no point to try improve it where in reality you can't hear the difference, and you are risking to have the switch mode PSU having unintended load or even degrading the transparency, which can be audible.

 

[MAB]

Sounds pretty safe to me. What’s a $1 in caps and 5 minutes of soldering anyways.

The advice you were given is sound. You seem determined to forge ahead.
Increasing the filter capacitance causes larger in-rush current during turn-on. Do you understand the ramifications of this?
If you stuff tons of caps, especially as audiophile cap-swappers do with low ESR caps, you run the risk of self-oscillation. Do you understand the ramifications of this?
Do you know if the manufacturer spec'ed the Mean Well supply to handle the in-rush current with more cap? And lower ESR? Are you clear on the role of ESR and Capacitance in this?
Do you know if the observed onset of self-oscillation with more cap and backed off on the original design?
Can you measure these above phenomena?
And if you could do all the above, do you really think the power supply would help this device? The power supply is the furthest thing from the sound, and if grounded, shielded, and operating properly is not the limiting aspect of the device.

 

[RenEH]

The advice you were given is sound. You seem determined to forge ahead.
Increasing the filter capacitance causes larger in-rush current during turn-on. Do you understand the ramifications of this?
If you stuff tons of caps, especially as audiophile cap-swappers do with low ESR caps, you run the risk of self-oscillation. Do you understand the ramifications of this?
Do you know if the manufacturer spec'ed the Mean Well supply to handle the in-rush current with more cap? And lower ESR? Are you clear on the role of ESR and Capacitance in this?
Do you know if the observed onset of self-oscillation with more cap and backed off on the original design?
Can you measure these above phenomena?
And if you could do all the above, do you really think the power supply would help this device? The power supply is the furthest thing from the sound, and if grounded, shielded, and operating properly is not the limiting aspect of the device.

Yes to all of the above. Guys, you could just say you have no idea what the value of the caps would be.

 

[Rottmannash]

I believe they are all giving you good advice in order to ensure good working order of the amp. If you feel the amp is disposable then go ahead and solder some caps in-but if it starts smoking don't say you weren't warned.

 

[YSC]

Yes to all of the above. Guys, you could just say you have no idea what the value of the caps would be.

it's highly likely that the value of the caps should be : NOTHING.

But sure, it's your amp and you can do whatever you like, Just don't complain for reliability if anything goes wrong

 

[MAB]

Yes to all of the above. Guys, you could just say you have no idea what the value of the caps would be.

We are trying to prevent you from degrading or destroying your second headphone amp. I don't need a schematic to do that, but you do need one to make changes that won't degrade the gear. Yet you just asking for the cap value.

 

[RenEH]

We are trying to prevent you from degrading or destroying your second headphone amp. I don't need a schematic to do that, but you do need one to make changes that won't degrade the gear. Yet you just asking for the cap value.

Curious to make assumptions to imply I have destroyed a first headphone amp. I’m sure you have peer reviewed evidence you guys get really angry when you can’t answer a question.

All of you are assuming this thing wasn’t cost cut. That’s hilarious when the original design was to be fully balanced and we ended up with what we have. I won’t look up the posts about that, google it yourself. Next you’ll tell me companies don’t penny pinch components where they can.

You got more fake news about me destroying headphone amps? Riddle me this though. Even if I did degrade the audio quality, but it wasn’t audible then what does it matter. You all think the power supply won’t make an audible difference after all

 

[MAB]

Curious to make assumptions to imply I have destroyed a first headphone amp. I’m sure you have peer reviewed evidence you guys get really angry when you can’t answer a question.

All of you are assuming this thing wasn’t cost cut. That’s hilarious when the original design was to be fully balanced and we ended up with what we have. I won’t look up the posts about that, google it yourself. Next you’ll tell me companies don’t penny pinch components where they can.

You got more fake news about me destroying headphone amps? Riddle me this thought. Even if I did degrade the audio quality, but it wasn’t audible then what does it matter. You all think the power supply won’t make an audible difference after all

No. Not angry at all. I've said it several times, no idea on the cap values.
Your previous headphone amp blew up, you had a thread about it. Seems it didn't have output fault protection. I was honestly a bit disappointed in Schiit for the directionless responses, seemed like they didn't know their own schematic! I certainly don't know their schematic, but they should. That's quite ironic. I have a generally good opinion of them, but that wasn't a quality interaction. I'm not angry, but you certainly have a right to be irritated at least. Appears you replaced it with a Drop, I have the same unit. I also wonder (briefly) about the possible function of those missing caps. I wonder more about the 3300uF Nichicon caps that people report bulging.

Review and Measurements of Massdrop THX AAA 789 Amp

Hi Brought this beast! Looks very solid, sound as well :) Should I plug off it from power (green diode always on when plugged to socket) ?

www.audiosciencereview.com

I know you have speculated removing those caps is cost-cutting. I equally speculate that they either caused oscillation which would destroy the actual power supply caps, or caused a large inrush current, which does the same. The question isn't the value of the caps, it's the circuit schematic first.

The biggest cost cutting is the volume control. That fails regularly.

My biggest beef is these cheap(ish) devices are not particularly reliable, even if they have good performance. Like fault protection for common shorts at the headphone jack. Or cold solder joints, etc.

 

[RenEH]

No. Not angry at all. I've said it several times, no idea on the cap values.
Your previous headphone amp blew up, you had a thread about it. Seems it didn't have output fault protection. I was honestly a bit disappointed in Schiit for the directionless responses, seemed like they didn't know their own schematic! I certainly don't know their schematic, but they should. That's quite ironic. I have a generally good opinion of them, but that wasn't a quality interaction. I'm not angry, but you certainly have a right to be irritated at least. Appears you replaced it with a Drop, I have the same unit. I also wonder (briefly) about the possible function of those missing caps. I wonder more about the 3300uF Nichicon caps that people report bulging.

Review and Measurements of Massdrop THX AAA 789 Amp

Hi Brought this beast! Looks very solid, sound as well :) Should I plug off it from power (green diode always on when plugged to socket) ?

www.audiosciencereview.com

I know you have speculated removing those caps is cost-cutting. I equally speculate that they either caused oscillation which would destroy the actual power supply caps, or caused a large inrush current, which does the same. The question isn't the value of the caps, it's the circuit schematic first.

The biggest cost cutting is the volume control. That fails regularly.

My biggest beef is these cheap(ish) devices are not particularly reliable, even if they have good performance. Like fault protection for common shorts at the headphone jack. Or cold solder joints, etc.

If you are you going to go several years in someone forum history to come up with some kind of straw man personal attack you may as well actually read the post! The amp did indeed malfunction if shorted , but it was still working fine. As in it still works and was long given away. Going through those kind of lengths certainly seems “angry”.

I actually have already replaced the volume knob with a stepped resistor ladder attenuator already. It works much better.

IMO caps that bulge after 3 years seems like a sign to me they are being pushed way too hard. Almost like those missing caps would have shouldered some of its burden and seems like cost cutting.

 

[MAB]

If you are you going to go several years in someone forum history to come up with some kind of straw man personal attack you may as well actually read the post! The amp did indeed malfunction if shorted , but it was still working fine. As in it still works and was long given away. Going through those kind of lengths certainly seems “angry”.

You misinterpreted me.
I wasn't critical of you, your Magni didn't have output protection, and from your posts it seemed Schiit gave you the runaround. I actually read the post. I think it's lame that the amp blows if you don't get the headphone jack all the way in. I'm keenly interested in reliability of devices, and some of them are disappointing. I even have a favorable opinion of Schiit, compared to some of the other unreliable gear. But seems you got let down.

I actually have already replaced the volume knob with a stepped resistor ladder attenuator already. It works much better.

Very good. The volume knob is certainly cheap and fails.

IMO caps that bulge after 3 years seems like a sign to me they are being pushed way too hard. Almost like those missing caps would have shouldered some of its burden and seems like cost cutting.

Just put in the biggest caps that fit.

 

[oleg87]

It's on the PCB mate.

Putting stuff on the PCB is free

Not sure what problem you’re trying to solve - the nonexistent power supply noise?

 

[solderdude]

IMO caps that bulge after 3 years seems like a sign to me they are being pushed way too hard. Almost like those missing caps would have shouldered some of its burden and seems like cost cutting.

When you mean too high a voltage (too close to the rated value) then yes.
It looks from the picture there are 16V caps in there while the one tested by Amir has 25V rated ones in there ?
A modification that was done later on or lowering production costs a few cent by lowering the rated voltage to 16V in later versions ?

 

[RenEH]

Just put in the biggest caps that fit.

I might take this advice literally. Given the pcb has an outline of the size of the capacitor in diameter, and If we assume the cap was originally rated for 25 volts as well as assume that it would also be a Rubycon cap, then couldn’t one approximate the capacitance from what fits that criteria?

 

[Killingbeans]

If we assume the cap was originally rated for 25 volts

Too low margin for a 24v PSU, if you ask me.

 

[YSC]

Sorry but I literally don't get it why you would wanted so badly to add a not in the design caps in there... a PCB layout is there for potential usage of caps, but it also depends on a lot of other variables to make it really useful if not harmful for the end result, from what Amirm have measured there isn't any noise needing those caps to filter out, so why do so is a mystery to me, there's a slot there don't necessary means the component of the amp need that. say if could be useful if the design use another type of power supply, or certain chips, using the wrong combination IME is usually worse in result

 

[RenEH]

Too low margin for a 24v PSU, if you ask me.

I have measured the voltage across the via’s where the caps should be. It’s 12 volts. I have been told the circuitry near the 24v input is a rail splitter. I’m thinking this is in line with the original fully balanced intent of the amp.