robertkd
Been here a while!
Electronics Engineer from sunny Queensland
Posts: 111
|
Post by robertkd on Apr 10, 2009 0:26:41 GMT
Well that makes more sense with the underside of the PCBA consider if the supply is nominally 40V i.e. not producing any audio you have 2 OPA Suns which want 50ma each that is 100ma total there for voltage drop Vdrop =I * R == .1A * 330R = 33V leaving you with 7V But for 1 OPA Sun .05A * 330R = 16V leaving you with 24V (oops wish I could add up ) So he is correct for one OPA Sun module the possible scenario put a series R of 330R (270R) for each OPA Sun strangely that is equivalent to mmm 160R ;D But the Zener diode might need to be upgraded for reasons discussed previously! the reason I tend to lean toward the lower value resisor is to allow for when the amplifier is producing audio power, the supply voltage will sag this is quite normal and depending on the power output can fall close to the rated 28V there for the supply to the OPA Suns will also fall,... Stay tuned ;D
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Apr 10, 2009 12:35:13 GMT
Soooo,
current thinking is the 330R/5w (or possibly 270R) resistors should be OK subject to some sort of change to the zeners?
Thanks, Chris
Oh, as for the evidence of heat in the area, the amp is over 20 years old, so maybe to be expected?
|
|
robertkd
Been here a while!
Electronics Engineer from sunny Queensland
Posts: 111
|
Post by robertkd on Apr 10, 2009 12:46:49 GMT
Soooo, current thinking is the 330R/5w (or possibly 270R) resistors should be OK subject to some sort of change to the zeners? Thanks, Chris Oh, as for the evidence of heat in the area, the amp is over 20 years old, so maybe to be expected? gezz Chris have you read any of my posts? Robert
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Apr 10, 2009 13:00:03 GMT
Sorry, maybe I misunderstood your previous post, I took it as offering the possibility of a simple resistor and zener change as apposed to "half wave rectifier", which, frankly, went over my head Chris
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Apr 10, 2009 13:22:02 GMT
Chris What Robert is saying is that you would need a series 330 ohm PW5 resistor to supply just ONE module. You could use an individual 330 ohm resistor for each module, or you would need a 150 ohm PW5 resistor in series to feed both modules. (or 2 x 330 ohm PW5 resistors in parallel.) You could use 150 ohm 3W , but they aren't as readily available . You should also use 1W 15V zener diodes. This would need to be done to both -VE and +VE supply rails. Alex
P.S. Ideally, both of the zener diodes should have something like 1,000uF 25V electrolytics in parallel with them too, due to the high current demands of these modules, and the apparent lack of reasonable bypassing in the existing setup.
|
|
robertkd
Been here a while!
Electronics Engineer from sunny Queensland
Posts: 111
|
Post by robertkd on Apr 10, 2009 13:37:51 GMT
Chris, I was trying to explain that given the rated 50mA standing current for a single OPA Sun sure 330R would be fine. My understanding is that you are going to run 2 OPA Suns therefore we need to allow for a total current of 100mA Ok the complication comes about from the fact that at idle the amplifiers supply is about +/- 40V and under load I would expect the supply voltage to fall to mid to low 30V or so, less when driven hard. now to provide a +/- 14V supply from worst case supply voltage of 30V you need to have a 160R series resistor in each supply in place of the 1KR. so 2 choices use one resistor nearest prefered value 150R OR parallel 2 330R 5 Watt resistors per supply rail. the maths for worst case supply rail voltage at 30V with a load of .1A (2 OPA SUN duals) voltage dropped across resistor 30V -14V (ZD) - 16V current drawn with 150R = 106ma resistor power dissapation 16V * 0.106A = 1.7W zener diode power dissipation 14V *0.006A = SFA or 84mW assume OPA's draw 50mA each then total 100mA leaving zener current 6mA The maths for system at idle suppy +/- 40V 40V-14V(ZD) -= 26V =174mA assume OPA's draw 100mA then zener current = 74mA zener power dissipation 0.074*14V = 1.036W Resistor power dissipation 26V * 0.174mA 4.5W (very frelling hot!) Ok what this all says is that it is possible but not a real practical solution, let me think on it, my concern is that unless you implement the resistor and zener supply properly it would be a disaster. an alternative is to make another supply using a couple of diodes and a couple of filter capacitors and who knows As an aside I measured the current drawn by my OPA Sun at +/- 11V and it draws 30mA at idle Robert
|
|
robertkd
Been here a while!
Electronics Engineer from sunny Queensland
Posts: 111
|
Post by robertkd on Apr 10, 2009 13:47:53 GMT
Chris What Robert is saying is that you would need a series 330 ohm PW5 resistor to supply just ONE module. You could use an individual 330 ohm resistor for each module, or you would need a 150 ohm PW5 resistor in series to feed both modules. (or 2 x 330 ohm PW5 resistors in parallel.) You could use 150 ohm 3W , but they aren't as readily available . You should also use 1W 15V zener diodes. This would need to be done to both -VE and +VE supply rails. Alex P.S. Ideally, both of the zener diodes should have something like 1,000uF 25V electrolytics in parallel with them too, due to the high current demands of these modules, and the apparent lack of reasonable bypassing in the existing setup. Alex, thanks for that my biggest problem here is the whole shooting match could had disastrous consequences should a regulator (eh zener) go out of regulation and cause feedback via modulating the supply voltage, I'm sure that would sound great Then the other side at system idle you may need to keep a fire extinguisher next ot the amplifier in case the PCB caught fire I'm thinking a half wave rectifier and 317/337 regulators Robert
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Apr 10, 2009 13:56:22 GMT
Robert and Alex, Thank you for the clarification. Holding fire Chris
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Apr 10, 2009 14:00:59 GMT
Chris What Robert is saying is that you would need a series 330 ohm PW5 resistor to supply just ONE module. You could use an individual 330 ohm resistor for each module, or you would need a 150 ohm PW5 resistor in series to feed both modules. (or 2 x 330 ohm PW5 resistors in parallel.) You could use 150 ohm 3W , but they aren't as readily available . You should also use 1W 15V zener diodes. This would need to be done to both -VE and +VE supply rails. Alex P.S. Ideally, both of the zener diodes should have something like 1,000uF 25V electrolytics in parallel with them too, due to the high current demands of these modules, and the apparent lack of reasonable bypassing in the existing setup. Alex, thanks for that my biggest problem here is the whole shooting match could had disastrous consequences should a regulator (eh zener) go out of regulation and cause feedback via modulating the supply voltage, I'm sure that would sound great Then the other side at system idle you may need to keep a fire extinguisher next ot the amplifier in case the PCB caught fire I'm thinking a half wave rectifier and 317/337 regulators Robert Robert To be on the safe side, you would really need to use 15V 1N5352B 5 WATT ZENER DIODES . TBH, as I have said previously, this is really a bit on the messy side for a novice to implement, and the heat risk is real . Your other option is far more preferable, but not readily implemented by a novice. Sorry Chris, but that's the way I see it. Better to be safe than sorry ! Alex
|
|
robertkd
Been here a while!
Electronics Engineer from sunny Queensland
Posts: 111
|
Post by robertkd on Apr 10, 2009 14:11:10 GMT
mmm well you know 50W zeners use a 5/32 UNC mounting stud got a bottoming tap handy I know your chomping at the bit but would rather you get the best outcome rather then some half baked system. These modules do have a distinctive sound but they are not a simple plug in replacement in some circumstances and I wonder how many have before got p-poor results because of the circuit topology they where plugged into
|
|
insomniac
Been here a while!
Team Zopiclone
Posts: 938
|
Post by insomniac on Apr 10, 2009 14:12:38 GMT
Chris, depending on the available free space inside the case it may be better and safer to fit a dedicated PSU for the OPA's. Kingwa's modification PSU's would be suitable for this and you could have it built to suit your requirements. These PSU's also produce some heat and need appropriate heatsinking or mounting to casing to keep them cool. If you decided to go down this route you could cut the supply tracks to the OPA sockets and feed in the new regulated supply. Of course Robert and Alex have came through with various ways and means to get the job done whilst also considering the very important safety aspects involved. Just another idea from my less experienced and more "modular" viewpoint.
|
|
robertkd
Been here a while!
Electronics Engineer from sunny Queensland
Posts: 111
|
Post by robertkd on Apr 10, 2009 14:17:52 GMT
Chris, depending on the available free space inside the case it may be better and safer to fit a dedicated PSU for the OPA's. Kingwa's modification PSU's would be suitable for this and you could have it built to suit your requirements. These PSU's also produce some heat and need appropriate heatsinking or mounting to casing to keep them cool. If you decided to go down this route you could cut the supply tracks to the OPA sockets and feed in the new regulated supply. Of course Robert and Alex have came through with various ways and means to get the job done whilst also considering the very important safety aspects involved. Just another idea from my less experienced and more "modular" viewpoint. Funny you should say that I've been looking at the SC HA supply PCBA and musing as to removing a couple of diodes from the bridge rectifier maybe larger filter capacitors, just need to get a 28V -0- 28V source to evaluate Robert
|
|
insomniac
Been here a while!
Team Zopiclone
Posts: 938
|
Post by insomniac on Apr 10, 2009 14:35:24 GMT
Another good option with the modded SC PSU Robert I have one of Kingwa's 2008ed PSU's in front of me here... +/- 18v 160ma... 8x MUR diodes... 2x 3300uF audio grade filter caps... I'll up a few pic's if you like.
|
|
robertkd
Been here a while!
Electronics Engineer from sunny Queensland
Posts: 111
|
Post by robertkd on Apr 10, 2009 14:48:32 GMT
Ah Dave you mean where is this em compass HA/DAC
|
|
insomniac
Been here a while!
Team Zopiclone
Posts: 938
|
Post by insomniac on Apr 10, 2009 14:53:01 GMT
Yes Robert... that's the one... a very nice piece of kit too.
Compass HA build in progress here ;D without DAC.
Again very nice class A discrete headphone amp...
|
|
robertkd
Been here a while!
Electronics Engineer from sunny Queensland
Posts: 111
|
Post by robertkd on Apr 10, 2009 15:01:53 GMT
Yes Robert... that's the one... a very nice piece of kit too. Compass HA build in progress here ;D without DAC. Again very nice class A discrete headphone amp... krap I do not need another HA or DAC www.audio-gd.com/enweb/DAC-Compass.htmmmm looks bloody brilliant Dave progress reports are now mandatory you do realise that Robert
|
|
insomniac
Been here a while!
Team Zopiclone
Posts: 938
|
Post by insomniac on Apr 10, 2009 15:13:23 GMT
Yes Robert... that's the one... a very nice piece of kit too. Compass HA build in progress here ;D without DAC. Again very nice class A discrete headphone amp... krap I do not need another HA or DAC www.audio-gd.com/enweb/DAC-Compass.htmmmm looks bloody brilliant Dave progress reports are now mandatory you do realise that Robert Sure Robert.. progress reports will be available from time to time This is the Compass HA board that I am using in conjunction with the 2008ed PSU: www.audio-gd.com/enweb/EARpic/Ear.JPGThis amp uses parallel connection volume control so the alps pot is not in the signal path and does not need replaced by an expensive attenuator: You sure you don't need another amp or dac ;D
|
|
robertkd
Been here a while!
Electronics Engineer from sunny Queensland
Posts: 111
|
Post by robertkd on Apr 10, 2009 15:28:34 GMT
Dave, clearly I have enough Headphone Amplifiers and DAC's my wife often remarks that I have only one pair of ears and many headphones, obviously the headphone amplifier to DAC to headphone ratio is a cotangent variable which means there is neither a correct ratio nor incorrect ratio. Therefore one is always on the look out for possible solutions to complex polynomials, I think that's correct Robert
|
|
insomniac
Been here a while!
Team Zopiclone
Posts: 938
|
Post by insomniac on Apr 10, 2009 15:45:30 GMT
clearly I have enough Headphone Amplifiers and DAC's my wife often remarks that I have only one pair of ears and many headphones, Gather all your wife's shoes, except for one pair, place in a pile in the garden and douse liberally with petrol... When she comes screaming tell her... "what's wrong, you only got one pair of feet" Sounds good to me... but many ratio combinations must be tested just in case
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Apr 10, 2009 17:24:01 GMT
Just got back to see the latest.
That's spooky, while I was out I thought to ask if the Jaycar psu could be hooked in somehow! OR, if another suggestion was beyond my feeble electronic abilities, if I could pay someone to knock it up as a module and tell where to splice it in ;D
Chris
|
|
|
Post by clausdk on Apr 10, 2009 18:38:54 GMT
Would it not be easier just to jump off a high building ??
|
|
robertkd
Been here a while!
Electronics Engineer from sunny Queensland
Posts: 111
|
Post by robertkd on Apr 10, 2009 22:36:46 GMT
one I prepared earlier ;D
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Apr 11, 2009 1:23:27 GMT
Just got back to see the latest. That's spooky, while I was out I thought to ask if the Jaycar psu could be hooked in somehow! OR, if another suggestion was beyond my feeble electronic abilities, if I could pay someone to knock it up as a module and tell where to splice it in ;D Chris Chris IF there was sufficient room internally, it may be easier for someone like yourself to fit a parallel 15-0-15VAC 20VA toroidal transformer, and use the PSU PCB as in the SC HA thread. Failing that, I would be inclined to put the Audio-gd modules to one side for use in a project where they would work much better. BTW, does anybody know the peak current drawn by the Audio-gd modules, as distinct from their idling current ? As they are unlikely to be in true Class A, their demands may be much higher on volume peaks, and the present zener situation would be completely inadequate. Dave Do you know if they are in Class A, or whether they are in Class AB instead ? What is the maximum current draw of these units ? TBH, I would imagine that they would need a power supply like in the SC HA to really shine. (not necessarily with a JLH either)
|
|
insomniac
Been here a while!
Team Zopiclone
Posts: 938
|
Post by insomniac on Apr 11, 2009 1:46:40 GMT
Dave Do you know if they are in Class A, or whether they are in Class AB instead ? What is the maximum current draw of these units ? TBH, I would imagine that they would need a power supply like in the SC HA to really shine. (not necessarily with a JLH either) Alex, as far as I'm aware only the OPA Moon is in class A. The maximum current draw should not exceed the Audio-gd specified values..ie.. approx 28ma per single module, 56ma per dual (Earth and Moon) and 25ma single/50ma dual for the Sun v2. I would say the better and cleaner the power supply the better the performance, just like most audio op-amp ic's. A dedicated low noise PSU just for the op-amps... even better. The modules perform well in many CD players without PSU supplementation though so I don't think their current draw is overly excessive.
|
|
robertkd
Been here a while!
Electronics Engineer from sunny Queensland
Posts: 111
|
Post by robertkd on Apr 11, 2009 1:58:47 GMT
The modules perform well in many CD players without PSU supplementation though so I don't think their current draw is overly excessive. Dave you'd be right in most case with CDP's and the like as they usually use linear regulators and have enough capacity to support and additional OPA module or 2,.. What makes Chris's particular situation is the single dual supply which can be anything from +/-40V and the rather modest low current supply for the front end electronics. But I think I have a solution Robert
|
|