Has anyone ever plugged a set of Ei KT90 into a Simple SE? I happened across a pair and I'd like to try them out in triode mode. They're the type that has all the little weld spots down the edge of the plates, so I think they're the "good" kind. I found a set of triode plate curves for them. Can anyone confirm if these are close to how they actually perform?
Based on that curve data, I'm guessing with a 560 ohm cathode resistor I'll get about 72 mA through the KT90 for an overall dissipation of ~27 watts. The KT90 should be able to handle a lot more. I'd like to consider using a smaller cathode resistor, but it's a pain to change it out and things are already getting close to the limit of my Hammond 374BX power transformer (rated 175mA total at 50 Hz?).
Any comments?
edit... Just thinking out loud - I bet what I would need to do to make this work is to drop down to 510 ohm cathode resistors, and go to solid state rectification to get the extra volts on the B+.
Based on that curve data, I'm guessing with a 560 ohm cathode resistor I'll get about 72 mA through the KT90 for an overall dissipation of ~27 watts. The KT90 should be able to handle a lot more. I'd like to consider using a smaller cathode resistor, but it's a pain to change it out and things are already getting close to the limit of my Hammond 374BX power transformer (rated 175mA total at 50 Hz?).
Any comments?
edit... Just thinking out loud - I bet what I would need to do to make this work is to drop down to 510 ohm cathode resistors, and go to solid state rectification to get the extra volts on the B+.
rknize said:
Me? No. The first time I fired up my Simple SE, it had a pair of 6CA7EH in it. I thought they sounded rather good. I've got some minor complaints about the stability of their grids (tiny jingling noises), but I wouldn't say I strongly dislike them.
I did have some problems with a set of EH EL34 and glowing screens. Maybe that's what you are thinking of?
In my experience with small signal stuff, diodes go bad with smoke and fire. Aka - open. But you would still have a really bad smell to investigate !! Then I'd worry about the life expectancy of the other one...
if you parallel them make sure both of them can take the full load. One will win with a better Vf and do all the work.
Sounds like you found yet another tube that will work in the SE... I'm interested in hearing what you think...
if you parallel them make sure both of them can take the full load. One will win with a better Vf and do all the work.
Sounds like you found yet another tube that will work in the SE... I'm interested in hearing what you think...
+1 on the parallel silicon. If one won't do it, then two seem to fail just as easily probably for the reasons Strat mentioned (one will do the work, fail, then the second takes over and also fails). If there is some series resistance in the circuit, then splitting that resistance between the two diodes is probably enough to allow them to share the load a bit better.
I have blown up way too much silicon not to have an answer here. A silicon diode will almost always fail to a short first. Then if the short circuit current is high enough, it may blow open. Look at the IXYS diode failures in the Simple SE's. The symptoms were always instantly blown fuses, due to a shorted diode. The same goes for BJT's. Mosfets often fail due to gate oxide damage.
The reasons that a silicon diode will fail are varied. They blow due to over voltage, over current, or over temperature. Often more than one of the above. In the power supply section of a tube amp most diodes are killed by excessive voltage spikes.
The power transformer is a magnetic device. The core is capable of storing considerable energy for brief periods of time. Ordinarilly the energy is transferred from primary to secondary in a controlled manner by application of a sine wave. If the power is interrupted suddenly by turning off the power switch exactly at the peak of the sine wave, the voltage can rise to extremely high levels as the magnetic energy stored in the core gets turned back into electrical energy. This is exactly the same principle used in automobile ignition coils to generate 20,000 volt sparks from a 12 volt battery.
This voltage spike usually contains little energy. In a tube amplifier where there are multiple secondaries all connected to loads, this energy is usually dissipated harmlessly in the tube filaments, leaving little to cause harm. It is still possible for destructive voltage levels to appear on the high voltage secondary. This is directly related to how the power transformer is made and the order in which the windings are placed on the core. It does seem that the current production transformers produce a lot more "spike" energy than older transformers. I have measured 2400 volt spikes on the HV secondary winding of a Hammond 274BX. Where does the 2400 volts go?
All silicon diodes have a reverse breakdown voltage spec. This is usually quite conservative. As the breakdown voltage is exceeded one or more electrons will jump through the depletion region this usually causes an "avalanche effect" leading to a lot of electrons breaching the depletion region. If the current density is high enough the diode will be permanently damaged. Some diodes are designed to operate in the avalanche region. We know them as zener diodes. Other diodes are designed and characterized such that they can withstand a certain amount of energy under avalanche conditions. They are avalanche rated.
FRED diodes are a special type of diode (Fast Recovery Epitaxial Diode) that produce far less noise than conventional diodes so we like to use them in our audio designs. The early ones like the IXYS diode were not avalanche rated, but worked OK until they changed the recipie. The Fairchild "Stealth" diode is avalanche rated and I have yet to see one fail. They absorb the spike energy and dissipate it as heat.
SO.....to sum this up use a Fairchild ISL9R8120P2 (in stock at Mouser) in your tube circuits including the Simple SE and they won't blow up. Use two in series if you really want to be sure.
If you are going to switch a Simple SE over to fixed bias (it's possible I have done it) you won't need a cathode resistor. Put a 100 or 125 mA fuse in series with a 10 ohm resistor where the cathode resistor used to be. The fuse will save your tubes and the 10 ohm allows for bias current measurement.
The reasons that a silicon diode will fail are varied. They blow due to over voltage, over current, or over temperature. Often more than one of the above. In the power supply section of a tube amp most diodes are killed by excessive voltage spikes.
The power transformer is a magnetic device. The core is capable of storing considerable energy for brief periods of time. Ordinarilly the energy is transferred from primary to secondary in a controlled manner by application of a sine wave. If the power is interrupted suddenly by turning off the power switch exactly at the peak of the sine wave, the voltage can rise to extremely high levels as the magnetic energy stored in the core gets turned back into electrical energy. This is exactly the same principle used in automobile ignition coils to generate 20,000 volt sparks from a 12 volt battery.
This voltage spike usually contains little energy. In a tube amplifier where there are multiple secondaries all connected to loads, this energy is usually dissipated harmlessly in the tube filaments, leaving little to cause harm. It is still possible for destructive voltage levels to appear on the high voltage secondary. This is directly related to how the power transformer is made and the order in which the windings are placed on the core. It does seem that the current production transformers produce a lot more "spike" energy than older transformers. I have measured 2400 volt spikes on the HV secondary winding of a Hammond 274BX. Where does the 2400 volts go?
All silicon diodes have a reverse breakdown voltage spec. This is usually quite conservative. As the breakdown voltage is exceeded one or more electrons will jump through the depletion region this usually causes an "avalanche effect" leading to a lot of electrons breaching the depletion region. If the current density is high enough the diode will be permanently damaged. Some diodes are designed to operate in the avalanche region. We know them as zener diodes. Other diodes are designed and characterized such that they can withstand a certain amount of energy under avalanche conditions. They are avalanche rated.
FRED diodes are a special type of diode (Fast Recovery Epitaxial Diode) that produce far less noise than conventional diodes so we like to use them in our audio designs. The early ones like the IXYS diode were not avalanche rated, but worked OK until they changed the recipie. The Fairchild "Stealth" diode is avalanche rated and I have yet to see one fail. They absorb the spike energy and dissipate it as heat.
SO.....to sum this up use a Fairchild ISL9R8120P2 (in stock at Mouser) in your tube circuits including the Simple SE and they won't blow up. Use two in series if you really want to be sure.
If you are going to switch a Simple SE over to fixed bias (it's possible I have done it) you won't need a cathode resistor. Put a 100 or 125 mA fuse in series with a 10 ohm resistor where the cathode resistor used to be. The fuse will save your tubes and the 10 ohm allows for bias current measurement.
Ok, just curious here, but if the current is enough to blow the diode then why wouldn't it be enough to open the short?
Normally I don't investigate the death throws of a diode, but I'm curious... under what conditions would a diode die, but leave a short circuit?
adding to my analog book of knowledge - which is painfully incomplete.
Normally I don't investigate the death throws of a diode, but I'm curious... under what conditions would a diode die, but leave a short circuit?
adding to my analog book of knowledge - which is painfully incomplete.
oldmanStrat said:
If memory serves, break-down occurs (secondary breakdown?) at the junction that causes permanent damage to the depletion layer, or something like that. So the first failure mode is for the diode to become a short. Then it goes up in smoke and becomes an open circuit if enough current can pass through it. In a bridge or full wave rectifier, this current path will be through one of the other diodes, destroying it as well. In a power supply, the dead-short on the transformer will hopefully blow the fuse on the other side.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.