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

[Deleted]

but I can simply short the resistor on the back of the board?!

give it a try.
You can't blow anything up if done right and with empty reservoir caps (make sure they are empty when fiddeling with this circuit).

The circuit it is in fact a damn good PS rails noise filter but has a rather high output impedance of a few Ohms.
Every upside has it's downside...
BUT this kind of PS is WAY more quiet then other PS regulator circuits.
Even the best parallel regs can't beat this noise wise.
It can do well over 120dB noise suppression but it has it's drawbacks as it can't be used in digital circuitery or AB class amps or circuitery where currents are not constant.
But since this PS only powers the input stage and this is 100% class-A this simple circuit is MUCH better suited then a regulator with feedback or parallel reg which always give higher noise figures.

so simply short that 1k resistor and see what it does (improve or not) and take it out if you don't like it.

[dicky]

This is way too deep for me . How do I empty the reservoir caps?

[gommer]

Empty = no charge = 0V. So you should short them. To be on the safe side, you'd better do that with a resistor or a light bulb.

[Deleted]

Feb 17, 2011 20:47:11 GMT dicky said:

This is way too deep for me . How do I empty the reservoir caps?

Hi Dicky
You could try sticking your tongue on them!
Seriously though, you can discharge them with something like a 10K 1 watt resistor connected across them for a little while.
Don't try discharging them with a screwdriver.
High voltage DC is actually more dangerous than mains power.
Many high powered amplifiers have rail voltages like +-70V, and if you went between those you would have a rather nasty 140V DC with very short term high current capabilities, as well as longer term capabilities of a few amps.
Regards
Alex

[PinkFloyd]

Feb 17, 2011 20:47:11 GMT dicky said:

This is way too deep for me . How do I empty the reservoir caps?

I always do it with a screwdriver (with insulated handle)... just short the caps a few times until the sparks stop flying... you "can" do this with your tongue but I do not recommend this method

[Deleted]

Feb 17, 2011 21:47:12 GMT PinkFloyd said:

Feb 17, 2011 20:47:11 GMT dicky said:

This is way too deep for me . How do I empty the reservoir caps?

I always do it with a screwdriver (with insulated handle)... just short the caps a few times until the sparks stop flying... you "can" do this with your tongue but I do not recommend this method

Mike
You can get away with that method on a typical HA PSU, but do that with high powered amplifiers and you will end up with nasty little marks on the screwdriver's blade, and perhaps even damage to the tip of the screwdriver if it's a small one.
Alex

[PinkFloyd]

Feb 17, 2011 22:04:58 GMT @sandyk said:

Feb 17, 2011 21:47:12 GMT PinkFloyd said:

I always do it with a screwdriver (with insulated handle)... just short the caps a few times until the sparks stop flying... you "can" do this with your tongue but I do not recommend this method

Mike
You can get away with that method on a typical HA PSU, but do that with high powered amplifiers and you will end up with nasty little marks on the screwdriver's blade, and perhaps even damage to the tip of the screwdriver if it's a small one.
Alex

Yes, I did it once on a VERY large cap and it actually "welded" the screwdriver to the capacitor terminal Glad I didn't use my tongue on that one

[Deleted]

How do I empty the reservoir caps?

Alex has the best method, but I would use a 100 Ohm resistor so it won't take that long.
The easiest way is power it off... wait 30 seconds or so (the voltage will have dropped for the biggest part) and then short the big caps one by one so the last bit of remaining voltage disappears.
There should be around 10V Dc on it or less after 30 sec.

For power amps and CERTAINLY when tinkering with tubes, switch mode power suplies (dangerous !) or photography flash equipment these caps have lethal voltages that can remain on there for days after being powered off !

Also caps don't like to be shorted.
Low ESR caps will have 100's of amps flowing and simply don't like it that much.

So in short... if you have some resistors lying around between say ... 22 and 470 Ohm of 1W or more use these and alllow a minute to empty them properly and / or measure the voltage during or after the process or (less preferred method but better then not doing it) et them drain natually for 30 secs and then short them... watch out for those sparks.

Indeed you can things weld with big caps, in Elektor there was a circuit for it with several power MOSFETS and a bank of caps.

[Deleted]

Feb 18, 2011 6:09:38 GMT @solderdude said:

How do I empty the reservoir caps?

Alex has the best method, but I would use a 100 Ohm resistor so it won't take that long.
The easiest way is power it off... wait 30 seconds or so (the voltage will have dropped for the biggest part) and then short the big caps one by one so the last bit of remaining voltage disappears.
There should be around 10V Dc on it or less after 30 sec.

For power amps and CERTAINLY when tinkering with tubes, switch mode power suplies (dangerous !) or photography flash equipment these caps have lethal voltages that can remain on there for days after being powered off !

Also caps don't like to be shorted.
Low ESR caps will have 100's of amps flowing and simply don't like it that much.

So in short... if you have some resistors lying around between say ... 22 and 470 Ohm of 1W or more use these and alllow a minute to empty them properly and / or measure the voltage during or after the process or (less preferred method but better then not doing it) et them drain natually for 30 secs and then short them... watch out for those sparks.

Indeed you can things weld with big caps, in Elektor there was a circuit for it with several power MOSFETS and a bank of caps.

Frans
It is not unusual these days to have a 2K2 or higher value resistor, (according to the supply rail voltages) , permanently connected across each rail with a series red L.E.D.
Alex

[Deleted]

That would discharge to about 1.5V but you won't find this in poweramps, tube amps, SMPS at the high voltage rails.
Also these LED's are usually found in just 1 power rail.

Also be aware when discharging with low Ohmic low wattage resistors these resistors can go up in flames which isn't funny when you are holding them !

[Deleted]

Feb 18, 2011 6:29:16 GMT @solderdude said:

That would discharge to about 1.5V but you won't find this in poweramps, tube amps, SMPS at the high voltage rails.
Also these LED's are usually found in just 1 power rail.

Also be aware when discharging with low Ohmic low wattage resistors these resistors can go up in flames which isn't funny when you are holding them !

Frans
An example, and yes the warning about low wattage resistors is a good one. In this example, they are using 5W resistors, which is a little overkill, but there usually isn't much readily available over the counter between 1W and 5W.
Alex

Attachments:

[jonclancy]

Bleeder Resistor Calculator here:

www.welwyn-tt.co.uk/products/resistors/calculation-tools.asp

[Deleted]

Hi Mike,

Can you please tell me where you purchased those green 75 ohm resistors you sent me for my Panda, (can`t remember what make they were, Japanese sounding i think) as i would like to buy a pr of 120 ohm ones.

Thanks Mick.

[joethearachnid]

Not Mike, but I'm fairly sure that the resistors are Kiwame brand.
They can be had off eBay for some hefty prices: cgi.ebay.co.uk/2ps-Japan-Kiwame-Carbon-film-Resistor-5W-audio-/320604663879
Or from HifiCollective for possibly slightly less (don't know what kind of shipping they charge): www.hificollective.co.uk/components/kiwame.html

-JoetheArachnid

[Deleted]

Yes there the ones Joe,

Thank you very much.

Mick.

[PinkFloyd]

Hi Mick,

Yes, I got them from that hificollective joint.... don't have any at the moment but do have some nice 100ohm carbon composites if you want a couple?

[Deleted]

I'm getting a hum on the left channel only of my Panda. I can't detect it when music is actually playing. I don't honestly know if it's been there from the start but I don't think so. (It's been a while since I used it). I've disconnected the board and inspected all components with a magnifier and can see nothing wrong. Is there any component that could cause this whilst still working otherwise OK?

I've checked all earths - IEC to case, case to ground on the Panda board. I can't think of anything else to try. I have the 6db gain mod.

Syd

[Deleted]

Connect the whole board but do not connect the input wires.
turn the volpot to min setting.
Is the hum still there ?

Now short the input of the PCB (no input wires connected) and turn the volpot up while listening and see if the hum gets smaller or bigger.

a small hum (with no input wires connected) could point to an oscilation which is hard to find if you don't have an oscilloscope.

Also check the voltages on the emitters of the 2SD669 and 2SB649 (should be around 17 and - 17V)
If these are higher this could be a source of hum in one channel also.

Check the DC output voltages of both channels.

[Deleted]

Me too Syd!! Very, very low level on one side.

For me, it's more apparent with the HD650. Much less with K701.

Would a wandering 'offset' do this? (or would it just lower dynamic range?)

I notice it more now I'm in the country and there's no surrounding ambient noise. It's extremely low volume, but then again I have laser ears for this kind of thing. I can't help it!!!

[Deleted]

Hi Ian

Before I dismantled it I checked the offsets and they were spot on, as were the 1.5v readings on the resistors for what it's worth (as I have no idea what affects what!)

I get it on 650's but much worse on the little modified Koss.

It doesn't stop the enjoyment when music is playing but as it is supposed to be dead silent...

Syd

[Deleted]

Thanks Frans

I'll not be able to check these things 'til later tonight. I know already it hums with no I/Cs attached, regardless of volume setting.

I'll report back.

Syd

[Deleted]

I suggest you check the voltages on the emitters of the 2SD669 and 2SB649 (should be around 17 and - 17V)
If these are higher this could be a source of hum in one channel also.

make sure your testpins do not slip or make contact with the pad and the groundplane around these pads.

The Collector should have between 22 and 28V on them (+ and -) and is the middle pin of these thicker flat stand alone transistors.
1 of the outer pins is the Base (on the left facing the writing on the part) and should bearound 17.5V the emitter is the most right leg.
be carefull NOT to short any of these pins with the testlead.

[Deleted]

It's there with the volume off Frans and doesn't change with the volume pot.

As Syd has described - very low level and not heard when music is playing. More apparent on headphones with fat bass response. Ambient noise takes it out, it's that low. Also, just in one channel.

Syd, your hearing is good!!

[Deleted]

I vaguely remember Joe mentioning this too but seemed to come from a ground loop which is impossible if not connected to other gear.
I mentioned somewhere the diodes aren't properly biased and this (or not proper functioning of this circuit) could cause this.

Are you both talking about the same channel and could Mike check his amp (or others) too with very sensitive cans ?

It could also be a layout 'problem' in the PCB design if all amps exhibit this behaviour and it as low as the background noise as AC enters the PCB.

I serviced Mick's amp but didn't listen to it only repaired it and did not notice any oscilations or nasty things.
It is impossible to see this low level hum on an oscilloscope if you are not looking at it and I powered it with a DC power source at that moment ( I only had the bare PCB) so even if it is a PCB design flaw it would not have been evident even when I looked for it.

Another (remote ?) possibility is the transformer being too close to the PCB or HP wiring.
Using screened cable from the output to the HP plug might help.
Also the internal input wiring I recommend to screen and the input connectors should not be connected directly to metal casing (which also could induce low level hum )

[Deleted]

Frans

Pin 1 reads around 17.2 on three of the devices, Pin 3 reads around 17.9. One of the 669s reads 6.1 on Pin 1 and 6.3 on Pin 3. I guess that's the problem? The collector voltage on all four is 25V.

The output is 12.5 mv on the left channel and 9.5 mv on the right, assuming I'm measuring correctly.

Wasn't sure what you meant about shorting the input at the PCB. Tried various combinations but hum didn't change.

Syd