Building a symmetrical PSU B1 buffer

That particular relay proves a pig. For one it needs much current gain to be pulled. Its unlikely you have an audio side problem, it may still be the relay not contacting well on one side. But measure before the relay. Maybe there is something before input or some audio side njfet has given up the ghost like that inadequate or dead 546. You can look for DC offset and signal on the 220Rs. If you don't have a generator, a steady tone cd or a PC tone wav can let any dvm measure AC RMS. Or even some repetitive busy music. Then you will know if you have to bin that relay or look elsewhere. Is your soldering iron more than 25W BTW? I wonder about erratic relay occurences or thermal stress.

Measuring the DC offset on the signal input (220R)to relay and checking the Resistance across the relay contacts after the click should give a clue about the cause. I am wondering what is the sound on switch on?
what make of relay he is using would be a good idea. That is why I prefer standard Omron relays for reliablity

Kannan
 
diyAudio Chief Moderator
Joined 2002
Paid Member
If it aint the relay, DC can pop and hiss. Maybe Uriah is the only one that got heavily jinxed but is good we see some practical stuff. It will be reference for the building people. No worries. This behind enemy lines trooper is gonna make it home if we all pull together as a team.
 
Love ya Salas :) Okay I will do the tests. Just thought I would say that the pop could be from a cheap switch and I will try a ceramic cap on it. Weird that it goes to only the one channel though..
While listening a few minutes ago I found that its the R out that doesnt work. There is no DC on either outputs. 3mV is close enough to 0 for me, but nothing at all on the odd channel. However if I remove the left channel RCA so that there is no signal to the amp on L then I can hear a real faint sound of music through the R channel. Like if you turn off your preamp and turn the CDP and the Amp on you get that tiny bleed through. Its like that.
so I will run the tests but its near dinner here so we will see. I will get it tonight though.
Omron is the make of the relay. I got the same line of relays as recommended on the silkscreen.
Uriah
 
diyAudio Chief Moderator
Joined 2002
Paid Member
After dinner check if you actually have 5V across the diode, or relay draws much and the voltage is low so it goes bonkers and killed or challenged the 546, goes not contacting well, or whatever weirdo thing. In such a case you can parallel a resistor across the 2X600 to adequatly drop less and let it have 5V. With the Nais 5V the 2X600 worked swell, but maybe yours draws more.
For those having the TQ-2 5V alright, has been checked in the protos, but Omron 12V or TQ2 12V is for the BOM using the jumper. Lighter as a load, less parts to solder.
 
No change with 400R making 5V. Still only one channel.
BTW the soldering iron is 48W but I can select the temp. I have it set so that I can hit most leads and in less than a second I solder and have a good joint and remove the iron. This is at 360C or 680F. Typical wave solder machines are at about 470F but the lead is dipped for a good 5 seconds or more. They move at about the rate of a pizza oven. I have soldered a few components at this temp including LM317, LM334 in TO92 packages and they survived. This was P2P so they actually got the iron for longer than the PCB does. Maybe 360C is to hot but it has worked fine for the last few months since I got the iron. I didnt like lower temps because it just took to long with the iron on the device to get the solder flowing.
I had a tone CD... think I moved it to HDD and then copied over the CD. I could burn it again if I can find it. There are a few on the net that are free I can download to. But there is nothing but bleed through music at super low levels on the dead channel like the level of a hum on the speaker. You need your ear right up to it.
Uriah
 
diyAudio Chief Moderator
Joined 2002
Paid Member
I am betting its the 2sk170's (we have factored out about everything else :) ). What voltage should I look for on them?

2SKs can surely die some times during soldering, I have found killed or hissy ones without any particular soldering effort history. Unfortunately there is only the B+, B- across them that you can DC measure. Other thing is to follow AC signal, either with a scope or measuring AC RMS. Like seeing it passes the input 220R and if it makes it to the B1 side of the output 220R then if it crosses it alright. If it stops just after the dead channel's 2SKs, then its tell tail caput.Or a resistor maybe does not pass. Sure about your input signal? Does it make it to the B1 safely? The faint sound must be crosstalk, mainly in the relay and gnd.
 
diyAudio Chief Moderator
Joined 2002
Paid Member
I did that check you wanted. I put the DMM on the relay side of each of the signal out 220R's and got 0R when the relay clicked on both sides.
This was without the 400R. I will solder that sucker in later.
Uriah

OK then. We fixed the relay to click, we know it does contact both channels, we know how to have even bullseye spec V across its coil.

Lets move on from there. Lets find the one channel signal bottleneck. Trace signal or look in the respective 2sks? What you fancy for start?
 
I am burning test disc right now. Oops it just popped out. Its done. I have this one
http://virtuelvis.com/archives/2004/09/audio-test-cd
Tones from 20Hz to 20000Hz.
So I suppose I will take 1k and feed it in straight from CDP without Lightspeed in the way and amplifier removed. Then you want me to follow your schematic for the right channel and try to find that 1kHz by following the Vrms all the way through the right channel circuit. Is this right? I should expect it to die after the 2sk?
Uriah
 
diyAudio Chief Moderator
Joined 2002
Paid Member
Great! Play track 33 in repeat. DVMs are good on catching a steady tone if it is over 100mV RMS and sinewave.

Yes trace it as you say. I don't expect, lets see. Can be from a solder joint anywhere, to a dead 2SK or two.

P.S. POWER OFF OR DISCONNECT THE MAIN AMP FIRST. (Not 4U, for future reference, many will be made).
 
Okay
D G S D G S S G D S G D
0-2.2-2.2 2.2-0-0 0-0-2.2 2.2-2.2-0

I am not getting voltage at the 2sk end of the 220R. Maybe the resistor is bad. I can replace it with another. I have to pull another off of a board if you think this is the case. The problem could be that I had it mounted vertically on another board and when I removed it there was not much left of the nub but enough to stick through the board so it is mounted near vertically here to. I measured that thing though and got 220 from the bottom of the board. Maybe I measured the other one. Will have to check again.
Anything wrong with the voltages I have above?
Uriah
 
diyAudio Chief Moderator
Joined 2002
Paid Member
Shows that your CD gives 2V2 RMS 0dB and that it stops. Though not on the 2SKS. Jumper the resistor first with crock clip. Is it the input to R relay 220R?
 

Attachments

  • ury.gif
    ury.gif
    1.4 KB · Views: 1,024