Tweet
Hi guys, it's been a while since I posted anything, I haven't built much lately because I've been too busy playing the amps I built previously!
But I just finished up this Sunn 2000s clone, mostly following the old schematic:
Prowess Amplifiers - Sunn - Schematics - 2000S
I biased it up, plugged in, turned it on, it sounded great! Then on the third note I played, the V6 output KT88 flashed and died (smoked screen resistor). But let's ignore that for now and assume it was just a bad tube. The amp kept ticking and still sounded great. But I noticed a "put-put-put" sound when turning up the volume just a wee bit. The sound could be stimulated by playing notes at slightly lower volume or would start by itself at slightly higher volume. I guess this is what they call "motorboating". I have several ideas about what might be causing this, and wanted to run them by the forum before hacking away willy-nilly.
1) My feedback loop is wrong.
I'm using a Hammond 1650T OPT. I've hard-wired its secondary to 4 ohm and 8 ohm output jacks. With this series of OPT's, one needs to re-configure the connections of the secondary to realize a 16 ohm output. Well who needs 16 ohms on a bass amp? But on the original schematic, the feedback loop is taken from the 16 ohm tap. When I was planning the build, I scratched my head about this for five minutes before deciding to just build the original feedback circuit and just connect to the 8 ohm tap and tweak it later if there's a problem. Well I guess I have a problem. Should I change the 1Kohm/750pF that's right on the OPT to 500ohm/1.5pF? It connects to a 47 ohm in the cathode of the V2 pentode. Should I up that 47 ohms to 100 ohms? Either? Both? What about that 390pF that connects to V3's screen?
2) I have too much power supply filtering capacitance.
The original circuit has 30 uF in the first stage. I have effectively 50 uF (two 100 uF 350V in series with voltage divider, just what I had laying around). I'd have to acquire different electrolytics to come up with 30 uF that can withstand 550V+
3) I have too much bass response going into the power stage.
I wanted to split the four tubes into two push-pull pairs that could be biased independantly. I figure it's usually easier to scrounge up two matched pairs of big tubes than it is to shell out for a matched quad of big tubes. So as a consequence, each tube V3,V4,V5,V6 has it's own grid bias resistor and it's own coupling capacitor coming from the phase inverter. The original schematic has each half of the phase inverter driving a shared 100 kOhm grid resistor through a 0.25 uF coupling capacitor. Like a simpleton, I used those same values (well, 0.22 uF), but twice as many. So now I have each half of the phase inverter driving 50 kOhms through 0.44 uF. This really lowers the cutoff frequency. Perhaps it would be more sensible to switch to 0.1 uF / 180 kOhm for each tube.
4) Maybe the asymmetry in the output stage caused by the KT88's death caused the oscillation. But maybe the instability was already there and contributed to the V6's death? Either way, throwing KT88's at the problem will get old fast.
5) Something completely different, perhaps as a result of my different, more compact layout?
FWIW, I am using an Antek AN-4TK400 toroidal PT, putting the dual 400V secondaries in parallel, and using a diode bridge rectifier. The negative bias voltage is coming from a capacitor-coupled negative charge pump. 590V B+ with no load!!!! Down to 550V with all power tubes biased.
Cheers and happy Friday,
Chris
But I just finished up this Sunn 2000s clone, mostly following the old schematic:
Prowess Amplifiers - Sunn - Schematics - 2000S
I biased it up, plugged in, turned it on, it sounded great! Then on the third note I played, the V6 output KT88 flashed and died (smoked screen resistor). But let's ignore that for now and assume it was just a bad tube. The amp kept ticking and still sounded great. But I noticed a "put-put-put" sound when turning up the volume just a wee bit. The sound could be stimulated by playing notes at slightly lower volume or would start by itself at slightly higher volume. I guess this is what they call "motorboating". I have several ideas about what might be causing this, and wanted to run them by the forum before hacking away willy-nilly.
1) My feedback loop is wrong.
I'm using a Hammond 1650T OPT. I've hard-wired its secondary to 4 ohm and 8 ohm output jacks. With this series of OPT's, one needs to re-configure the connections of the secondary to realize a 16 ohm output. Well who needs 16 ohms on a bass amp? But on the original schematic, the feedback loop is taken from the 16 ohm tap. When I was planning the build, I scratched my head about this for five minutes before deciding to just build the original feedback circuit and just connect to the 8 ohm tap and tweak it later if there's a problem. Well I guess I have a problem. Should I change the 1Kohm/750pF that's right on the OPT to 500ohm/1.5pF? It connects to a 47 ohm in the cathode of the V2 pentode. Should I up that 47 ohms to 100 ohms? Either? Both? What about that 390pF that connects to V3's screen?
2) I have too much power supply filtering capacitance.
The original circuit has 30 uF in the first stage. I have effectively 50 uF (two 100 uF 350V in series with voltage divider, just what I had laying around). I'd have to acquire different electrolytics to come up with 30 uF that can withstand 550V+
3) I have too much bass response going into the power stage.
I wanted to split the four tubes into two push-pull pairs that could be biased independantly. I figure it's usually easier to scrounge up two matched pairs of big tubes than it is to shell out for a matched quad of big tubes. So as a consequence, each tube V3,V4,V5,V6 has it's own grid bias resistor and it's own coupling capacitor coming from the phase inverter. The original schematic has each half of the phase inverter driving a shared 100 kOhm grid resistor through a 0.25 uF coupling capacitor. Like a simpleton, I used those same values (well, 0.22 uF), but twice as many. So now I have each half of the phase inverter driving 50 kOhms through 0.44 uF. This really lowers the cutoff frequency. Perhaps it would be more sensible to switch to 0.1 uF / 180 kOhm for each tube.
4) Maybe the asymmetry in the output stage caused by the KT88's death caused the oscillation. But maybe the instability was already there and contributed to the V6's death? Either way, throwing KT88's at the problem will get old fast.
5) Something completely different, perhaps as a result of my different, more compact layout?
FWIW, I am using an Antek AN-4TK400 toroidal PT, putting the dual 400V secondaries in parallel, and using a diode bridge rectifier. The negative bias voltage is coming from a capacitor-coupled negative charge pump. 590V B+ with no load!!!! Down to 550V with all power tubes biased.
Cheers and happy Friday,
Chris
Comment