Hello SSE builders,
I have built my SSE with both solid state and tube rectification. As per the PCB, I have added D3 and D4 in series from the HT secondary of the power transformer (Hammond 274BX) as well as the inrush current limiter. I am thinking of bypassing D3 and D4 by means of a jumper. I understand that these diodes are there to protect the GZ34 but I feel that kind of renders the tube rectifier useless as all rectification is done by d3 and d4. With the diodes in place, even if the GZ34 develops a short from anode to cathode, everything would work as usual. One would not even notice the issue. The board allows to switch to SS rectification anyway (which I tend to prefer but that is another story)
So how much damage can be expected of a rectifier tube failing? Anybody with real-world experience of the damage done by a failed rectifier? I understand the power supply caps may fail but if that is the extend of the damage, I am willing to live with the risk. I have put a 2A fuse on the primary. Do you think that is sufficient to protect the power transformer?
While on the topic, if a GZ34 arcs once, is it considered dead? My Sovtek arced the first time I powered on the amp but it was due to an accidental short on the chassis. After fixing that, it seems to work just fine.
I have built my SSE with both solid state and tube rectification. As per the PCB, I have added D3 and D4 in series from the HT secondary of the power transformer (Hammond 274BX) as well as the inrush current limiter. I am thinking of bypassing D3 and D4 by means of a jumper. I understand that these diodes are there to protect the GZ34 but I feel that kind of renders the tube rectifier useless as all rectification is done by d3 and d4. With the diodes in place, even if the GZ34 develops a short from anode to cathode, everything would work as usual. One would not even notice the issue. The board allows to switch to SS rectification anyway (which I tend to prefer but that is another story)
So how much damage can be expected of a rectifier tube failing? Anybody with real-world experience of the damage done by a failed rectifier? I understand the power supply caps may fail but if that is the extend of the damage, I am willing to live with the risk. I have put a 2A fuse on the primary. Do you think that is sufficient to protect the power transformer?
While on the topic, if a GZ34 arcs once, is it considered dead? My Sovtek arced the first time I powered on the amp but it was due to an accidental short on the chassis. After fixing that, it seems to work just fine.
If you intend to keep using the GZ34, I would leave D3 & D4 in place. Removing them is just asking for trouble, especially if you're using week tubes like the Sovtek or JJ GZ34's. They are not entirely useless with SS diodes in series with the plates. You still get the benefit of slower B+ rise with the tube rectifier. 
jeff
jeff
D3 and D4 were added around 2012, a time period when the 5AR4's produced by the three major current production manufacturers had quality problems. The SSE with a 375 volt transformer like the 274BX runs the 5AR4 near its limits. Sometimes a new tube would spark out on first power up, and occasionally two or three were needed to find a good one. Fortunately, things have improved a bit, but the quality of most new production tubes is not near what was made 50 years ago.
Diodes D3 and D4 do not provide all of the rectification in the amp, in fact they provide virtually none. As you already know they can be bypassed and you probably would not notice. If you can hear the difference between a tube and SS rectifier, then by your own admission, bypassing the tube would be noticed.
In any conventional recitfier / cap AC to DC conversion the rectifiers only conduct for a brief moment near the peak of the incoming sine wave. This is known as the conduction time or conduction angle.
Due to the low forward voltage of a SS diode this time is short with an abrupt turn on and in some cases a "complicated" turn off that can create noise pulses when interacting with imperfections in the power transformer. When conducting slightly, or near it's maximum spec a SS diode will drop a near constant 0.75 volts (for silicon). It's "dynamic resistance" is low, 1 ohm or less, and nearly constant during conduction, and high (megohms) when not conducting.
A vacuum tube rectifier behaves different from a silicon diode. As the tube becomes forward biased it starts to conduct slowly with no abrupt turn on. Its dynamic resistance is still usually well over 50 ohms at full conduction and increases again as the tube nears turn off. It turns off slowly which will usually not provoke ringing in the power transformer.
If you wire a 50 ohm resistor in series with a 1 ohm resistor and ask the pair to do a job, which one does all the work? The diodes in series with the tube work the same way. Under forward bias (conducting) conditions the tube does 98% of the work and provides 100% of the soft conduction benefits that create the tube rectifier sound. During reverse bias (not conducting) the silicon diodes provide an extra 1000 volts of breakdown voltage capability that prevent a tube arc from continuing if one should start.
At turn on the tube rectifier is asked to provide all of the B+ current the amp needs AND fill up all of the empty filter caps at a time when it has been rudely awakened from a cold sleep. If the coating on the cathode is not uniform, it will not heat evenly. The thin spots will heat quicker than the thick spots and therefore start conducting first. If the cathode is not centered perfectly in the plate structure, conduction will be heavier in the closest areas, and its reverse breakdown voltage will be below spec. If the current density exceeds the capability of the cathode coating at a point anywhere on the cathode a tube arc may start. This arc may continue indefinitely if the reverse breakdown voltage is exceeded, which is usually the case once an arc ionizes the path from plate to cathode. The extra reverse bias protection provided by the SS diode is what stops the arc and saves the tube.
In my experience over 50+ years an old school DH rectifier tube like a 5U4 or 5Y3 can spark at every turn on and still live long and prosper. A IDH tube like the 5AR4 or 6BY5 will usually spark at every turn on once it has arced over once. If yours behaves properly at turn on when observed in a dark room, I would continue to use it, bit keep an eye on it at turn on for a while.
A dead rectifier tube should blow the fuse. Some tubes may just erupt in internal fireworks for some time before the fuse blows with a 2 amp fuse. Try a 1.6 amp slow blow fuse in your amp and see if it holds in normal use. I had several hundred 1.6 amp fuses so that's what I put in my SSE's. I lived in South Florida at the time. Lightning hits the power lines every day in the summertime in Florida. This resulted in a blown fuse every few weeks, but I ran my SSE hard at nearly 100 mA on the KT88's.
I just finished my Tubelab SSE build. When the solid state rectifier jumper is out, there is no voltage at B+. When I insert the jumper then there is 450V at B+. Is my Sovtek 5AR4 blown? I just bought it new from Partsconnexions . Is there a way to trouble shoot the 5AR4 to see if it is working? I checked and there is continuity between pin 2 and 8. I dont have diodes D3 and D4 mentioned above since the schematic does not show where and how to wire them. Is there a more robust tube? is the 5U4 or 5Y3 more robust? Could someone please sketch how to wire up D3 and D4 to protect the 5AR4. Thanks
Place a 1N4007 or UF4007 diode in the D3 and D4 positions on the PC board near the rectifier tube socket. Put a CL-140 Inrush Current Limiter in the TR-1 position and you will have B+. These components were added back in 2012 to make life much easier on the 5AR4 at a time when all 3 of the major new production tube suppliers were selling some tubes of questionable quality.
You can put a jumper wire in place of each part and get the board to work. This reverts the board to the version sold from 2007 through 2012.
The rectifier tube will be much happier with the components installed. The SSE runs the rectifier tube pretty hard. A 5Y3 will NOT live long in a typical SSE. They do work fine in 6V6GT builds that incorporate a lower voltage power transformer for about 320 volts of B+. A 5U4 can work, but the robustness of the rectifier tube comes down to the way it was assembled. The cathode / filament must be evenly coated and perfectly centered to get uniform current flow throughout the tube.
The SS rectifier switch just puts SS diodes in parallel with the existing vacuum tube rectifier. No interruption in current flow occurs, so there should be no ugly consequences from flipping the switch while the amp is operating. There WILL however be an audible POP SOUND when turning the switch on due to the abrupt rise in B+ voltage and the momentary increase in output tube current while the cathode bypass caps charge to the newly needed bias voltage. This pop may NOT be a good idea for some speakers. The momentary increase in tube dissipation may not be suitable for amps that are already running at or near maximum plate dissipation.
I have flipped the switch while the amp was on several times in my SSE without issue, but there is a surge in voltage and current which does pose some risk. Venture here at your own risk.
Do NOT operate the triode / UL / pentode (if equipped) switch while the amp is operating. This will cause a total interruption in output tube current which will create a collapsing magnetic field in the OPT leading to a very loud POP in the speaker and the possibility of an arc or other failure.
I have flipped the switch while the amp was on several times in my SSE without issue, but there is a surge in voltage and current which does pose some risk. Venture here at your own risk.
Do NOT operate the triode / UL / pentode (if equipped) switch while the amp is operating. This will cause a total interruption in output tube current which will create a collapsing magnetic field in the OPT leading to a very loud POP in the speaker and the possibility of an arc or other failure.
