Completely separate. Transformer+two stages of regulation. The DCB1 is just sitting on an antistatic bag with its dedicated transformer near by. The RCA ground tabs are soldered to a copper bar, all ground wires are connected to the centre of the bar, bar to SE point via a ground loop breaker.
LS 'out' goes to B1 'input', B1 'output' to amp.
Last edited:
Its weird saying you can read any output voltage before the click, since before click there is no connection between the Jfets and output socket. I would expect that DC peak and slow DC die out if it was connected all the time. Check the continuity between 220R out to ouput socket when off, maybe its stuck. Can you also use 4 corner makeshift lifts instead of having it on the bag?
I'm almost done as well. Got the last components yesterday, soldered everything but the 4x FETS in the audio path.
I've got ~1.9V LEDs, and I'm running 2x51Ohm resistors at R1.
The problem is, the 3 LED string doesn't light up. Around R1 I measure 6.9V. The 5 LED string has no problems and my output seems right - 10.3/-10.2 Volt.
I'm not sure what can be wrong yet, I'm sure the leds are in the correct orientation, etc.
I've got ~1.9V LEDs, and I'm running 2x51Ohm resistors at R1.
The problem is, the 3 LED string doesn't light up. Around R1 I measure 6.9V. The 5 LED string has no problems and my output seems right - 10.3/-10.2 Volt.
I'm not sure what can be wrong yet, I'm sure the leds are in the correct orientation, etc.
That's totally intentional. I'll heatsink them and if it isn't enough, I'll mount them to the case, which is a gigantic aluminium heatsink. I planned to parallel a 3rd 51r with them when all is done if I can handle the heat.
Plus, I thought I'll sink some of that current in a headphone amp which should go in the same enclosure, that'll draw some current away from the shunt. I'll see how it goes.
The leds are all mounted right. They actually light up when I check for continuity with my DMM. The 2x2sk170 CSS transistors for them are also intact, since current flows if I check for continuity between the CSS and the last LED - it still lights up.
The voltage drops are off however, they're random values between 0.3 and 0.9V for all 6 leds there. Now that is weird. The leds in the 5-strings all drop 1.94V with their CCS (I'm guessing they're a little on the high side, but that's all good). I'm too sleepy to play with electricity right now, but I'll be on it tomorrow again.
They can be lighting up on outside sourced power but are they orientated as they must in circuit? If you put a 9V battery's (-) at the near ground foot (100R point) of the plus (IRFP9240s side) Led's CCS Jfet and (+) on first Led, do they light up dropping 5.4-6V? Reverse for negative polarity's side test. Also look closer you did not mix 240 with 9240 on the wrong side. Labelling is thin many times.
Hi,
what are the voltage drops across the gate resistors of the series FETs and the shunt FETs? A high Vdrop would indicate a high gate current (=damaged FET).
Check the voltage across those 6 non lit LEDs again.
If you are very careful you can short one LED at a time and see what happens to the voltages across the unlit LEDs.
what are the voltage drops across the gate resistors of the series FETs and the shunt FETs? A high Vdrop would indicate a high gate current (=damaged FET).
Check the voltage across those 6 non lit LEDs again.
If you are very careful you can short one LED at a time and see what happens to the voltages across the unlit LEDs.
Last edited:
Hi again,
After swapping the Mosfets, to ensure they're ok (they were) I decided to approach the best way I know when I'm *certain* something should work. I pulled out the 3LED strings and soldered a freshly matched set. Problem solved.
I've no idea what exactly went wrong with the previous ones though, for all I know they still work. They're going in the scrap bucket though.
I'm happy to report that all is in order. Voltages are +10.22/-10.32 and the drop across the 2x51r is 2.02V. The heat is reasonable even without heatsinks, but they'll be installed soon. I'll try to install a happier LED in the positive rail (green probably, I'll measure a bunch) to make it symmetrical.
Thanks Salas and AndrewT, as always
After swapping the Mosfets, to ensure they're ok (they were) I decided to approach the best way I know when I'm *certain* something should work. I pulled out the 3LED strings and soldered a freshly matched set. Problem solved.
I've no idea what exactly went wrong with the previous ones though, for all I know they still work. They're going in the scrap bucket though.
I'm happy to report that all is in order. Voltages are +10.22/-10.32 and the drop across the 2x51r is 2.02V. The heat is reasonable even without heatsinks, but they'll be installed soon. I'll try to install a happier LED in the positive rail (green probably, I'll measure a bunch) to make it symmetrical.
Thanks Salas and AndrewT, as always
Right now, for driving my LM3886 Gainclone. It's got no input caps, but it's got no gain at DC and I've a DC protect installed after it so I don't worry too much.
Input, in most cases, will be Bantam or Grub DAC. The Bantam has coupling caps, while the Grub doesn't. Not an issue, at any rate.
Input, in most cases, will be Bantam or Grub DAC. The Bantam has coupling caps, while the Grub doesn't. Not an issue, at any rate.
Swapping the positions of the FETs didn't really change anything, so I'll leave them the way they are. Offsets will wander between 2 and 3mV, but I'm currently happy.
After a close but short encounter with 220V I got to listening. First of all - it's dead quiet. No noise whatsoever and I'm using it with a horrible source right now - the built-in sound card of a laptop. From what I can notice, the sound is definitely quite a bit better. That sound card is noticeably worse than my Bantam DAC, when used without the DCB1. Right now the sound is splendid, even though it's sitting 'naked' on my floor, with crappy wires to connect it all.
I intend to swap a led from each 5-string and raise the supply voltage a volt, to something a little over 11. I've got 20 after rectification so I've enough headroom.
Oh and btw - the 4 2sk170s in the audio path are running quite nice, with their 14mA IDSS That's what I had, that's what I used...
I've got some pics, I'll get them online and put some links, when I start boxing it. It's glowing nice and warm with its Amber ultra-high brightness LEDs.
After a close but short encounter with 220V I got to listening. First of all - it's dead quiet. No noise whatsoever and I'm using it with a horrible source right now - the built-in sound card of a laptop. From what I can notice, the sound is definitely quite a bit better. That sound card is noticeably worse than my Bantam DAC, when used without the DCB1. Right now the sound is splendid, even though it's sitting 'naked' on my floor, with crappy wires to connect it all.
I intend to swap a led from each 5-string and raise the supply voltage a volt, to something a little over 11. I've got 20 after rectification so I've enough headroom.
Oh and btw - the 4 2sk170s in the audio path are running quite nice, with their 14mA IDSS
That's what I had, that's what I used...I've got some pics, I'll get them online and put some links, when I start boxing it. It's glowing nice and warm with its Amber ultra-high brightness LEDs.
Last edited: