PANDA Headphone amp kit, build thread, upgrades & tweaks.

[Deleted]

New question my good sirs.
I know it has been brought up before... but how would one go about making a BALANCED Panda?
Is it possible without too much hazzle?
Reason being I have two Panda boards I got mostly without components so I have sourced the components and can now make two identical Panda's.
This made me wonder if I should embark on the great "balanced" adventure for my Beyerdynemic T1.
I am thinking specifically about the wiring for the headphones and the source (which would be balanced as well).

Any thoughts guys?

[joethearachnid]

I think we concluded that a balanced Panda could drive twelve HE-6s at once with the fury of a thousand suns. Or something like that. Are you sure you want to have that kind of power?

-JoetheArachnid

[Deleted]

Oh, but what else to do with them?
Maybe I could decrease gain and just sit back and enjoy the surplus of power.
Also, it would be a cheap way for me to go balanced as my Beta22 is currently broken into a 2-channel and a 2-channel with virtual ground version and I am not quite sure I will have the heart to butcher those to get balanced. Honestly, I have not got a balanced source so it would be convenient to have just standing by till I get that balanced source.

"...fury of a thousand suns", somehow that appeals to me.

[Deleted]

Why does balanced exist for headphones and what improvements does that give and why ?

T1 balanced on Pandas ?

That's with 120 Ohm out (ideal for T1) so each amp 60 Ohm output R.
max power = 650mW into 600 Ohm and the max rating of the T1 = 300mW.
You would be able to reach 130dB.
Enough to damage your ear and drivers.

1 Panda will reach 150mW into 600 Ohm b.t.w. enough to reach 124dB
The pain treshold (means really uncomfortably loud) is around the same SPL.
I doubt most people even don't like 100 to 110dB on their ears.

You can also build a 2nd Panda and put it up for sale ?

[Deleted]

Balanced for headphones? I honestly do not know, but I am reluctant to dismiss it without being able to at least say I tried it. It could be just another way to get money from audiophiles, but it could actually make a difference. Anyways, with both a stock Panda and a balanced one I would be able to compare the two, instead of comparing brand X headamp with brand Y BALANCED headamp. Does this make sense?

Is there money in building these things and selling them?
I reckon getting just the components worth can at times be challenging...

I am aware that I could do some serious damage to my ears with such a setup, but that is not my intention. I just thought that since I have already got ONE working Panda and 2 spares I could put those extras to good use and make a cheap balanced headamp since it would be much cheaper than the balanced Beta22 (of which I will soon have both a 2-channel and a 2-channel with virtual ground finished, but no balanced version... yet).

[PinkFloyd]

You can balance it with these:



I agree with Frans, this "balanced" thing is totally unnecessary for the home headphone amp user.... it sounds so appealing though doesn't it? "balanced".... yes, that means a lot.... Dual mono sounds so yesterday.

If it rocks your boat though then go for it man

[clausdk]

Apr 27, 2012 18:56:52 GMT @mcfrede said:

Balanced for headphones? I honestly do not know, but I am reluctant to dismiss it without being able to at least say I tried it. It could be just another way to get money from audiophiles, but it could actually make a difference. Anyways, with both a stock Panda and a balanced one I would be able to compare the two, instead of comparing brand X headamp with brand Y BALANCED headamp. Does this make sense?

Is there money in building these things and selling them?
I reckon getting just the components worth can at times be challenging...

I am aware that I could do some serious damage to my ears with such a setup, but that is not my intention. I just thought that since I have already got ONE working Panda and 2 spares I could put those extras to good use and make a cheap balanced headamp since it would be much cheaper than the balanced Beta22 (of which I will soon have both a 2-channel and a 2-channel with virtual ground finished, but no balanced version... yet).

[ukemaxxer]

Jan 10, 2011 20:15:59 GMT PinkFloyd said:

Panda 4 basically has larger power caps (4700uF) bypassed with film and polystyrene caps.... UF4001 shunted with silvered Mica and polystyrene caps in place of the 22nF ceramic types..... I will get around to housing it one of these days....

I have just bought myself one of these from an ebay seller, before I build it, I want to gather some extra bits to be able to do the mods you have on yours. I have already ordered some 22pf polystyrenes to replace the 8 ceramics, from the pictutes, I can't make out what values you have across your IN4001 diodes, or the ones across your power caps. If you could let me have those values, I can get some ordered before I start assembly. Don't fancy ripping it all back apart at a later date.
I will be running this with my Sennheiser HD595's so I reckon around 100R at the output, it will be fed from a Beresford TC7510 DAC, unless I build something else!

Cheers
Richard

[Deleted]

Hi Richard,

Welcome to the forum.

The HD595's work best from a LOW Ohmic source and will sound absolutely boomy, muddy and fat on 100 Ohm output R.
I suggest no output R or max 10 Ohm for THIS particular headphone.

You should be aware that the Panda without output R's can supply 2.4W in the HD595 and when clipping 5W which could potentionally destroy the drivers if unattended and accidentally playing at full power.

slowing down the diodes can be done with 10nF to max 100nF.

Polystyrenes behave very badly as decoupling caps when they are used stand-alone.
Also they are not recommended to replace ceramics in certain applications but in this (relatively) slow design they aren't very critical.
Since they are in parallel to film caps that are many factors bigger in value (and thus lower in ESR) they do not do anything but ease your mind.
Whack in there whatever you like (value) and the result will be the same , an improved sonic character.

[ukemaxxer]

Hi solderdude, thanks for that, saved me from making my first mistake.
The 595's are not in their first flush of youth, but I'd rather not kill them just yet.

[Deleted]

Jun 22, 2012 22:18:06 GMT ukemaxxer said:

Hi solderdude, thanks for that, saved me from making my first mistake.
The 595's are not in their first flush of youth, but I'd rather not kill them just yet.

I would be more concerned about permanently damaging hearing due to excessive SPL.

[PinkFloyd]

Jun 22, 2012 17:23:10 GMT ukemaxxer said:

Jan 10, 2011 20:15:59 GMT PinkFloyd said:

Panda 4 basically has larger power caps (4700uF) bypassed with film and polystyrene caps.... UF4001 shunted with silvered Mica and polystyrene caps in place of the 22nF ceramic types..... I will get around to housing it one of these days....

I have just bought myself one of these from an ebay seller, before I build it, I want to gather some extra bits to be able to do the mods you have on yours. I have already ordered some 22pf polystyrenes to replace the 8 ceramics, from the pictutes, I can't make out what values you have across your IN4001 diodes, or the ones across your power caps. If you could let me have those values, I can get some ordered before I start assembly. Don't fancy ripping it all back apart at a later date.
I will be running this with my Sennheiser HD595's so I reckon around 100R at the output, it will be fed from a Beresford TC7510 DAC, unless I build something else!

Cheers
Richard

Still sounds good to my old ears

[joethearachnid]

Hi Richard,

As a long-(ish) time Panda and HD 595 owner, and having used them together quite a bit I can say that I do enjoy the combo a lot. The HD 595 has a lot of detractors on this forum and elsewhere, and though I can see where they're coming from I also find that the 595+Panda combo is the most (ugh, I hate myself for saying this) 'musical' combo I own. Not the most accurate but plenty of pleasant distortions and a warm, enveloping tonality. I do actually tend to run my Panda at a 110R output most of the time and have never found it to affect my experience negatively.

I have a 0ohm output on my Panda as well that I installed for use with orthodynamics and the like, so if it would be helpful to you I can do some direct comparisons between the two outputs with the 595s and furnish you with my findings.

-JoetheArachnid

[ukemaxxer]

Hi Joethearahnid,

Thanks for the info, I'm on a few forums and do notice that a lot of people don't like the 595. Each to their own, I tried a pair of Grado SR80's after glowing recommendation before I got the Senns, I just could not do with them at all. Hated the sound, hated the comfort, for what they cost, they looked and felt cheap. Loved the feel and sound of the 595 since I got them.

I'm still waiting for some stuff from Farnell so haven't started building the Panda yet, having a hard time finding an enclosure I like in the UK though, all seem to be plastic or 70's style instrument chassis. Any recommendations gladly received. The Hifidiy.net one looks good, just the shipping from China that's the bugbear.

Cheers
Richard

[Deleted]

Jun 24, 2012 0:13:32 GMT joethearachnid said:

The HD 595 has a lot of detractors on this forum and elsewhere,

-JoetheArachnid

Don't know why, I think they're a really nice phone, I love mine.

[joethearachnid]

Experimenting today, there is actually a far more significant difference between 0ohm and 110ohm outputs that I had realised, especially on the 595 (changing the output resistance on the C.H.AMP by a similar factor has a less noticable difference). The bass is far more present and prominent on the higher-ohmic output, thought the treble is toned down somewhat. The 110ohm is a far warmer, smoother presentation overall.

I do remember that when I first started using the Panda I found the bass somewhat overwhelming at first, so it's definitely worth your while experimenting with output resistances to find one that suits you. Something like 50 ohms might give you the best balance.

-JoetheArachnid

[Deleted]

Jun 25, 2012 23:02:04 GMT joethearachnid said:

Experimenting today, there is actually a far more significant difference between 0ohm and 110ohm outputs that I had realised, especially on the 595 (changing the output resistance on the C.H.AMP by a similar factor has a less noticable difference). The bass is far more present and prominent on the higher-ohmic output, thought the treble is toned down somewhat. The 110ohm is a far warmer, smoother presentation overall.

I do remember that when I first started using the Panda I found the bass somewhat overwhelming at first, so it's definitely worth your while experimenting with output resistances to find one that suits you. Something like 50 ohms might give you the best balance.

-JoetheArachnid

68 ohms is often a good compromise if you own different types of headphones .

[Deleted]

Jun 25, 2012 23:02:04 GMT joethearachnid said:

Experimenting today, there is actually a far more significant difference between 0ohm and 110ohm outputs that I had realised, especially on the 595 (changing the output resistance on the C.H.AMP by a similar factor has a less noticable difference). The bass is far more present and prominent on the higher-ohmic output, thought the treble is toned down somewhat. The 110ohm is a far warmer, smoother presentation overall.

I do remember that when I first started using the Panda I found the bass somewhat overwhelming at first, so it's definitely worth your while experimenting with output resistances to find one that suits you. Something like 50 ohms might give you the best balance.

-JoetheArachnid

Some headphones really need the 0 Ohm or close to it, others 30 Ohm or 68 or 100 to 120.
As most HP's nowadays (Ian noticed it too) are designed to be driven from portable gear (and thus low Ohmic as current is no issue) a lot of these are designed to be driven from 0 Ohm, but don't necesarily sound best on 0 Ohm.

Each headphone (and person using it) has a preference and, technically as well, most headphones seem to have an optimum in performance at certain output resistances.

You will have to determine that optimum yourself.
If you like bigger bass and less highs a higher resistance can be beneficial with certain headphones that react well (it's why my designs have more than one output resistor).
The HD595 (and the rest of that series) react quite strongly and as they are quite potent in the highs a resistance won't hurt, but the bass can become overwhelming/boomy/too warm and mids can become muffled.

It's why I always say .. experiment yourself, preferably blind.
It's your ears and your gear.

Therefore it is hard to create a list of headphones which perform optimal at a given output resistance, as taste also comes into play.

[ukemaxxer]

Got the Panda finished, on setting it up, I am unable to get the voltages across the 1k5 resistors below 1.55v.
I know in the thread it was advised to set them at 1.5v instead of 1.3v, so I think it should be OK. Just wondered if anyone else had this issue.

Listening to it all afternoon anyway and it sounds good so far.

[krojules]

hello,

I'm a new one in this forum and I'm a French Guy so all my apologizes in advance for my poor English level...

Well, I performed this headphone amplifier since a long time now and I realise today that i never checked th DC offset.

For fixing my mistake I've checked offset this morning and I saw problem with it.

By using the DC offset method in this thread I can reach the 0mV for the left side but I can't obtain better than 150mV for the right side. And the worst is the following, not in positive but in negative (-150mV).

If someone have an issue for my problem.

krojules.

[Deleted]

Welcome Krojules,

check:
voltages (with respect to ground) on Emitter Q3, Q12, Q22, Q28 (at least the parts from the right side)
be carefull NOT to accidentally short the pad to the groundplane directly around the pin !
the voltage should be somewhere around +16V and -16V when one of them is MUCH higher that transistor is gone.

check the input FET's Q6, Q8, Q15, Q16 one of them might be too far 'off' spec wise.
check this by noting the voltages across R1, R2, R27, R28 (not to ground but across the Resistor)

[krojules]

Ok,

I've already checked R1 R2 R27 and R28 I adjusted to 1.35v.

I found Q3 and Q22 but I'm not sure where the emitter is but I have around 17 volts between ground and a pin.

I did not find Q12 and Q28 on the schematic.
Do you have more informations about these components ?

Thanks for your fast reply.

[Deleted]

Q3 and Q22 should have 2 pins close to 17V and 1 (the middle one) close to 22 to 25V (depending on trafo)
Q12 and Q28 are the same type of transistor in the middle of the board and should have around -17V on the outer pins and around -22 to -25 on the inner pin.

[krojules]

Hello solderdude,

Here is the results for the Q components

Q3: 16.9/20.6 and 17.56 for center pin
Q12: -16.95/-17.59 and 21.4 for center pin

Q22: 17.15/17.78 and 19.9 for center pin
Q28: -17.12/-17.77 and -21.3 for center pin

No noticeable things between good side and problem side ?

Do you think the blue adjustable resistance (with little screw) could be damaged ?


I don't know what I have to do after measured these components.

I thank you.

krojules

[Deleted]

krojules

You could also try setting the trimpot back to centre, and check the voltage drops across the 1K5 drain resistors, as well as across the pairs of emitter resistors (top and bottom) of the 2nd stage.
My apologies if Frans has already covered these.
Alex