Building a symmetrical psu B1 buffer - Page 91

Salas · 29th January 2010, 06:02 PM

Quote:

Originally Posted by pengboon

Curious to know the reason too.

Counteracts ringing when feeding cable lines, especially long. I.e. perform damping.

jormajj · 30th January 2010, 05:56 PM

When using two Hypnotize PCBs to build a balanced stereo buffer, can the PCBs be stacked with only one having the regulator components or do each PCB need to have regulator of its own ?

Salas · 30th January 2010, 06:21 PM

They could, especially when you can assure very short and sturdy connections of output voltages and ground between them if stack one on top of the other, maybe with pins. I don't know if they will retain all their subjective performance as if with individual regs, but electrically is most certainly feasible.

njepitt · 31st January 2010, 12:04 AM

Hi Guys,

I hope I have managed to score 100 2SK170s, so at least for the time being the issue of finding substitute parts has gone away... We'll see if/when they arrive...

On a related issue, I have a couple of questions about building the PSU for the B1 active crossover, which will use multiple copies of a symmetrical B1. These use BF862 in place of 2SK170 in the filters, but that's irrelevant to the questions, and since I plan on using the circuit Salas posted in #11 and #12 here, this thread seems the right place to ask.

1. Instead of building one monster PSU for all 14 buffer circuits or building multiple copies of the whole PSU circuit, somewhere (I can't find it right now) someone suggested splitting the circuit, using one large CCS block feeding several shunts. (Following a suggestion of AndrewT, perhaps four shunts - one for each channel high-pass and low-pass, although it might be fewer or more). Looking at the schematics, if I do this I would put all the leds and so forth in the large block, and only have multiple copies of the part formed by R3,R6,Q2,Q3,Q4 and C2, right? (I'm looking at the +ve regulator here - obviously the -ve one is done in the same way.)

2. If I use a nicely heatsinked IRFP9240 for Q1, to handle all the current, can I be cheap and use something more modest for Q4 without compromising SQ too much? I am thinking of something like IRF9620, IRF9630 or similar T0-220 device. If the performance is really much better with the IRFP9240 I'll pay, but it's a large cost so I'd like to be sure there's a real advantage. (Before you all think I'm too much of a cheapskate, please bear in mind that down here IRFP9240 will cost me something like $7 or $8 each, IRFP240 are similar, whereas their smaller brothers are much cheaper. I would need something like 10 of them...)

Cheers

Nigel

Salas · 31st January 2010, 12:58 AM

All those things sound feasible, like using one heavy CCS and to star out in multiple shunts with TO-220 Mosfets etc, but I am not sure if it will not be a PITA to layout, check it for oscillations. If you can, and having an oscilloscope, try, but we DCB1S builders have no prior experience of laying out such a scheme and basically you are on your own for practicalities. The error amp and shunt part is what lies under, and to the right of the 3 leds controlled input Mosfet. Also the locality is one positive thing, low divided current through lower Ygfs Mosfets is a minus.

pengboon · 31st January 2010, 02:18 PM

For the Decoupling A (0.15uf) and B (0.1uf), is it OK to use 0.22uf to substitute?

Salas · 31st January 2010, 11:57 PM

Decoupling what components?

pengboon · 1st February 2010, 12:26 AM

Quote:

Originally Posted by Salas

Decoupling what components?

IN the BOM, it says near "BC517" and near "L7812ACV". Not sure what it actually decouples as I have not received the PCBs and its not stated in the schematics.

Salas · 1st February 2010, 12:32 AM

That was a remedy for those without a Darlington like BC517, when some prototypes showed some erratic NPN relay driver and relay coil ''chatter'' mishaps. A 0.1uF to 0.15uF cap across B,C of the NPN under the coil could shut down the coil buzzing, soldered underneath and directly to the NPN's pads, hidden from view. No further report of chatter was mentioned after specifying BC517 relay driver. If you just happen to have a BC550 for that or other, the 0.22uF cap will still save the day if you chance on an upsetting relay coil. The 7812 is the 3 terminal 12V reg for the delay and relays.

umut1001 · 6th February 2010, 10:25 AM

hi.As a power supply capacitor people suggest 4700uf caps.Is 3300uf 50v elna silmic 2enough for it ? Does sound quality decrease ?thanks

6th February 2010, 10:25 AM	#910
umut1001 diyAudio Member Join Date: Mar 2007 Location: Istanbul	hi.As a power supply capacitor people suggest 4700uf caps.Is 3300uf 50v elna silmic 2enough for it ? Does sound quality decrease ?thanks __________________ My New and vintage capacitor store: http://stores.ebay.com/Paradise-of-H...dio-components

Thread Tools	Search this Thread
Show Printable Version Email this Page	Search this Thread: Advanced Search

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
New-building of my B1 buffer	Babowana	Pass Labs	328	20th February 2010 01:14 AM

New To Site?	Need Help?
Register to Participate Search Privacy Statement Contact Us	Frequently Asked Questions Did you forget your password? Mark Forums Read


	Please consider donating to help us continue to serve you. Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving

30th January 2010, 05:56 PM	#902
jormajj diyAudio Member Join Date: Nov 2002 Location: Finland	When using two Hypnotize PCBs to build a balanced stereo buffer, can the PCBs be stacked with only one having the regulator components or do each PCB need to have regulator of its own ?

30th January 2010, 06:21 PM	#903
Salas diyAudio Chief Moderator Join Date: Oct 2002 Location: Athens-Greece	They could, especially when you can assure very short and sturdy connections of output voltages and ground between them if stack one on top of the other, maybe with pins. I don't know if they will retain all their subjective performance as if with individual regs, but electrically is most certainly feasible.

31st January 2010, 12:04 AM	#904
njepitt diyAudio Member Join Date: Aug 2008 Location: Brasilia	Hi Guys, I hope I have managed to score 100 2SK170s, so at least for the time being the issue of finding substitute parts has gone away... We'll see if/when they arrive... On a related issue, I have a couple of questions about building the PSU for the B1 active crossover, which will use multiple copies of a symmetrical B1. These use BF862 in place of 2SK170 in the filters, but that's irrelevant to the questions, and since I plan on using the circuit Salas posted in #11 and #12 here, this thread seems the right place to ask. 1. Instead of building one monster PSU for all 14 buffer circuits or building multiple copies of the whole PSU circuit, somewhere (I can't find it right now) someone suggested splitting the circuit, using one large CCS block feeding several shunts. (Following a suggestion of AndrewT, perhaps four shunts - one for each channel high-pass and low-pass, although it might be fewer or more). Looking at the schematics, if I do this I would put all the leds and so forth in the large block, and only have multiple copies of the part formed by R3,R6,Q2,Q3,Q4 and C2, right? (I'm looking at the +ve regulator here - obviously the -ve one is done in the same way.) 2. If I use a nicely heatsinked IRFP9240 for Q1, to handle all the current, can I be cheap and use something more modest for Q4 without compromising SQ too much? I am thinking of something like IRF9620, IRF9630 or similar T0-220 device. If the performance is really much better with the IRFP9240 I'll pay, but it's a large cost so I'd like to be sure there's a real advantage. (Before you all think I'm too much of a cheapskate, please bear in mind that down here IRFP9240 will cost me something like $7 or $8 each, IRFP240 are similar, whereas their smaller brothers are much cheaper. I would need something like 10 of them...) Cheers Nigel

31st January 2010, 12:58 AM	#905
Salas diyAudio Chief Moderator Join Date: Oct 2002 Location: Athens-Greece	All those things sound feasible, like using one heavy CCS and to star out in multiple shunts with TO-220 Mosfets etc, but I am not sure if it will not be a PITA to layout, check it for oscillations. If you can, and having an oscilloscope, try, but we DCB1S builders have no prior experience of laying out such a scheme and basically you are on your own for practicalities. The error amp and shunt part is what lies under, and to the right of the 3 leds controlled input Mosfet. Also the locality is one positive thing, low divided current through lower Ygfs Mosfets is a minus.

31st January 2010, 02:18 PM	#906
pengboon diyAudio Member Join Date: Dec 2005 Location: Singapore	For the Decoupling A (0.15uf) and B (0.1uf), is it OK to use 0.22uf to substitute?

31st January 2010, 11:57 PM	#907
Salas diyAudio Chief Moderator Join Date: Oct 2002 Location: Athens-Greece	Decoupling what components?

1st February 2010, 12:32 AM	#909
Salas diyAudio Chief Moderator Join Date: Oct 2002 Location: Athens-Greece	That was a remedy for those without a Darlington like BC517, when some prototypes showed some erratic NPN relay driver and relay coil ''chatter'' mishaps. A 0.1uF to 0.15uF cap across B,C of the NPN under the coil could shut down the coil buzzing, soldered underneath and directly to the NPN's pads, hidden from view. No further report of chatter was mentioned after specifying BC517 relay driver. If you just happen to have a BC550 for that or other, the 0.22uF cap will still save the day if you chance on an upsetting relay coil. The 7812 is the 3 terminal 12V reg for the delay and relays.