Hi anatech,
will let you know when being back. Pentode vs. beampower is an old debate. Depends on the circuit and purpose.
Hi anatech,
remembered that it was the preamp of Ampex 350 and I downloaded the service manual. Thr tube I mean is 12SJ7 pentode.
Hi radiogaga,
Beam power tubes are more efficient and have lower plate resistance. The heater current is lower and they don't cook their bases to bits over time. I have every single time measured better performance and they sounded better both to me, and the customer involved. One thing to be aware of is that they can both be marked incorrectly. A skinny bottle is a pentode (EL34) for sure, the fat one will be a beam power tube (6CA7). Don't forget that the Beam Power tube is a later refinement over the Pentode tube, so what i have said shouldn't be surprising in any way.
Anyway, to each his own. I know enough now to use 6CA7 tubes in place of an EL34 unless you are looking at a regulator pass tube or something special. But in those cases I will experiment so see if the improved Beam Power type can be used. Having said that, I do prefer the 6BQ5 over the 6AQ5, so the situation is reversed. I have not investigated if there is a real improved beam power tube for the 6BQ5 (that might be exciting, but I don't think a direct replacement in Beam Power exists).
The 12SJ7 is an early high gain Pentode, so you would have had to probably redesign the socket wiring for a newer design with lower noise. This tube was used for high gain low noise applications, so at the time it may have been the best choice. At least it had DC heaters! They have a whole string of these in series to achieve the desired gain, wow! Two for the tape head amp, but notice that the first gain stage is selected tubes. That means they knew they were at the limits for the tube type and needed to pick out quiet ones. I would suggest that if they wanted higher performance but the same sound, you have have to redesign for a more modern 9-pin tube type. The chemistry for the cathode would be better and probably good for lower noise. The new tubes might have higher plate resistance, but given they are pentodes it likely would not matter since that is the characteristic of a pentode. It really just depends on what they want.
In your shoes, you might want to stick to RCA 12SJ7 or similar since these tended to be lower noise, Raytheon is another good choice. The problem is that these were used in consumer equipment and radios mostly, so a commodity item. You would need to use a tube destined for military or scientific equipment. That's when noise and other parameters become important instead of price.
-Chris
Beam power tubes are more efficient and have lower plate resistance. The heater current is lower and they don't cook their bases to bits over time. I have every single time measured better performance and they sounded better both to me, and the customer involved. One thing to be aware of is that they can both be marked incorrectly. A skinny bottle is a pentode (EL34) for sure, the fat one will be a beam power tube (6CA7). Don't forget that the Beam Power tube is a later refinement over the Pentode tube, so what i have said shouldn't be surprising in any way.
Anyway, to each his own. I know enough now to use 6CA7 tubes in place of an EL34 unless you are looking at a regulator pass tube or something special. But in those cases I will experiment so see if the improved Beam Power type can be used. Having said that, I do prefer the 6BQ5 over the 6AQ5, so the situation is reversed. I have not investigated if there is a real improved beam power tube for the 6BQ5 (that might be exciting, but I don't think a direct replacement in Beam Power exists).
The 12SJ7 is an early high gain Pentode, so you would have had to probably redesign the socket wiring for a newer design with lower noise. This tube was used for high gain low noise applications, so at the time it may have been the best choice. At least it had DC heaters! They have a whole string of these in series to achieve the desired gain, wow! Two for the tape head amp, but notice that the first gain stage is selected tubes. That means they knew they were at the limits for the tube type and needed to pick out quiet ones. I would suggest that if they wanted higher performance but the same sound, you have have to redesign for a more modern 9-pin tube type. The chemistry for the cathode would be better and probably good for lower noise. The new tubes might have higher plate resistance, but given they are pentodes it likely would not matter since that is the characteristic of a pentode. It really just depends on what they want.
In your shoes, you might want to stick to RCA 12SJ7 or similar since these tended to be lower noise, Raytheon is another good choice. The problem is that these were used in consumer equipment and radios mostly, so a commodity item. You would need to use a tube destined for military or scientific equipment. That's when noise and other parameters become important instead of price.
-Chris
Hi anatech,
thank you for the information. Since I do not much on tube amps anymore I just replace tubes in musical amps when defective. For the studio, how it is now they are happy with the sound of the Ampex 350. These are only used as mic preamps, not as tape recorder. I just wanted let you know part of my experience with vintage gear.
Sven
Hi anatech,
yes, it is helpful to get knowledge from experience others. I have a long term project and I am not sure to do it. This is a Class-A PP power amp with pentodes stripped to triodes. According to the literature this should be the topology with the least distortions but not much power and no efficiency. Just for gaining an own experience with such a concept instead of only reading about it. I know there are the 300B fans out in the scene and also ready to buy amps. But this is all to expensive for me. And probably the 300B hype does not offer in real what is talked about.
Nope.
Ultralinear.
Never use an old, obsolete tube or design. Tubes have been more fully developed since the 300B (ancient design!), and amplifier design has progressed to much higher quality since engineers were actually designing with the 300B (I mean real engineers).
Using a tube as a triode is not a really good idea, speaking from experience. Some manufacturers only made the output stage switchable between two modes due to market pressure (to get sales folks). The lore is that triode mode has the least distortion. That's because of the very low transconductance, but then you have to drive it harder. Well, what about the preceding stages? Not only that, distortion climbs rapidly as output swing of any tube (signal and driver stages) goes up. They love to talk about lower distortion of an output stage in isolation or driven specially and without feedback. Believe me, you really want feedback! They never compare as a system, but that is what you are using.
So what does an obsolete tube get you? Reduced supplies for those tubes and greatly increased prices while also delivering reduced performance. What a wonderful combination! The basic 6L6GC design (including the excellent 7581A) is really as far as power tube design went. Everything else is a variation on the theme. Beam power tubes. Even the newest "designs" are merely variations on the 6L6 basic design. You'll find more exotic designs for large RF power tubes, but they are really not suitable for domestic use. I've seen (in use) tubes that were moved on carts like torpedos, little ones mind you. The KT88 / 6550A is probably the highest power tube I would consider using. My preference is for one pair of output tubes per channel. Higher power requires higher B+ voltages which increases failure modes and expense way too fast, and high current plus current sharing when using multiple tubes is really not a joke. Need more power? Go solid state.
Technology did progress steadily before approximately the 1980's, plus or minus a few years. Since then only a few brands have improved design for quality, the rest for less expensive manufacture and shorter service life. (that's true folks). So what I am saying is that technology has improved throughout the tube era, that means designs, materials - everything. So going back to an earlier design makes absolutely zero sense, especially also accepting an old, embryonic active device design. Just think about that, investigate a little and use common sense. Yes, engineering was also used to make the devices lower cost, but they did not sacrifice reliability unless the tube was listed as an economy tube (6DJ8 comes to mind, the 6922 was the opposite as a high quality, high performance tube). However beware of tube brands that were specifically marketed for low cost and manufactured inexpensively. It wouldn't surprise me to see line rejects in that mix as well. Those merely had to function, that's all. I know because I worked at a place that began selling that garbage. Counterfeit tubes are probably also out there in force, mostly as NOS stock commanding higher prices.
-Chris
Ultralinear.
Never use an old, obsolete tube or design. Tubes have been more fully developed since the 300B (ancient design!), and amplifier design has progressed to much higher quality since engineers were actually designing with the 300B (I mean real engineers).
Using a tube as a triode is not a really good idea, speaking from experience. Some manufacturers only made the output stage switchable between two modes due to market pressure (to get sales folks). The lore is that triode mode has the least distortion. That's because of the very low transconductance, but then you have to drive it harder. Well, what about the preceding stages? Not only that, distortion climbs rapidly as output swing of any tube (signal and driver stages) goes up. They love to talk about lower distortion of an output stage in isolation or driven specially and without feedback. Believe me, you really want feedback! They never compare as a system, but that is what you are using.
So what does an obsolete tube get you? Reduced supplies for those tubes and greatly increased prices while also delivering reduced performance. What a wonderful combination! The basic 6L6GC design (including the excellent 7581A) is really as far as power tube design went. Everything else is a variation on the theme. Beam power tubes. Even the newest "designs" are merely variations on the 6L6 basic design. You'll find more exotic designs for large RF power tubes, but they are really not suitable for domestic use. I've seen (in use) tubes that were moved on carts like torpedos, little ones mind you. The KT88 / 6550A is probably the highest power tube I would consider using. My preference is for one pair of output tubes per channel. Higher power requires higher B+ voltages which increases failure modes and expense way too fast, and high current plus current sharing when using multiple tubes is really not a joke. Need more power? Go solid state.
Technology did progress steadily before approximately the 1980's, plus or minus a few years. Since then only a few brands have improved design for quality, the rest for less expensive manufacture and shorter service life. (that's true folks). So what I am saying is that technology has improved throughout the tube era, that means designs, materials - everything. So going back to an earlier design makes absolutely zero sense, especially also accepting an old, embryonic active device design. Just think about that, investigate a little and use common sense. Yes, engineering was also used to make the devices lower cost, but they did not sacrifice reliability unless the tube was listed as an economy tube (6DJ8 comes to mind, the 6922 was the opposite as a high quality, high performance tube). However beware of tube brands that were specifically marketed for low cost and manufactured inexpensively. It wouldn't surprise me to see line rejects in that mix as well. Those merely had to function, that's all. I know because I worked at a place that began selling that garbage. Counterfeit tubes are probably also out there in force, mostly as NOS stock commanding higher prices.
-Chris
All true. Since it is just a hobby and a fun project for me I will use the tubes I have in my stock. Power tubes will be original russian GU50 which were derived from the german LS50. Of course driving power triodes requires huge voltage swing and some current capability because of the miller effect. There I need to think about how to handle this. Maybe with bootstrapping the plate voltages. All this will be experimental. Also because calculating a circuit in advance may help but the actual physical circuit and especially the tubes do not behave according to the calulations accurace. Ultralinear is much more effective but has drawbacks in distortion behaviour. And of course feedback in order to get a stable gain and less distortion even when tubes parameter change over time. Passive components will be good new ones, no paper or fishy oil. Further just point-to-point wiring, no PCB which is worst for power tube circuits. Many newer built amplifers have PC boards but there are open soldering points too after years what I experienced at repairs. And I still use the good old lead solder which I am used to since 50 years. OK, silver blended might be a bit better but not valuable for me. Regarding newer produced tubes I do not think that these are better than the Mullard and Telefunken and RCA tubes at their times. When reading old original literature on tube development so the chemistry was really complicated. Hardware tooling is more accurate today but the secret lies in the building and treating of the cathode. Also grid and plate material is needed to be perfect to withstand and dissipate the heat of power tubes.
- Sven
