Hoffman Amplifiers Tube Amplifier Forum
Amp Stuff => Tube Amp Building - Tweaks - Repairs => Topic started by: pbman1953 on January 01, 2015, 08:45:02 am
-
Within the last 2 years New Sensor has offered the new KT series tubes with the Tung Sol name-
KT 90
KT 120
KT 150
I have a Sunn 2000s which is able to handle, as stock, either the 6550 and KT88. Would anyone know if any of the above would work in the amp?
If not, does anyone have a preference from 6550 to KT88? I've read that it's an "ok" thing to have a pair of each in the amp. Actually I do have one 2000s with 2 bias pots,one per tube pair. Thanks to the forum I installed the second so I wouldn't have to worry about needing a matched quad. Pairs seem to be more common. At the moment I have the amp with the Tung Sol 6650.
Thanks
-
Within the last 2 years New Sensor has offered the new KT series tubes with the Tung Sol name-
KT 90
KT 120
KT 150
I have a Sunn 2000s which is able to handle, as stock, either the 6550 and KT88. Would anyone know if any of the above would work in the amp?
What is the goal of replacing the existing 6550/KT88's with any of the new tubes? More power output?
The Sunn 2000s schematic (http://el34world.com/charts/Schematics/files/sunn/Sunn_2000s.pdf) shows the output tubes are running in Ultralinear mode, with ~500vdc on the plate and screen (variable depending on line voltage). For any proposed tube swap, you should first check that the tubes can handle this voltage, with maybe some extra margin for safety. The KT90 (http://www.eierc.com/rc/KT90.htm), KT120 (http://www.tungsol.com/tungsol/specs/kt120-tung-sol-specs-curves.pdf) and KT150 (http://www.tungsol.com/tungsol/specs/kt150-tung-sol.pdf) data sheets all show maximum voltage ratings well above this, so voltage is not a problem.
Next, you should figure out if the amp can handle the tube swap; specifically, bigger output tubes often have higher heater current draw. The 6550 data sheet (http://www.tungsol.com/tungsol/specs/6550-tung-sol.pdf) indicates a heater current of 1.6A per tube. Data for the bigger tubes show heater currents from 1.6A to 2A per tube, for a potential 1.6A extra overall sucked from the heater winding. I don't know if the Sunn has that extra unused capacity available.
You might be able to offset the extra VA drawn from the PT by replacing the parallel GZ34 rectifiers with a solid-state rectifier.
Let's assume the Sunn can handle the extra heater current for the bigger tubes (hopefully, someone with experience with this particular amp will say for sure). Again, what's the goal of using the new tubes?
If it's more output power, you may have a problem. The tubes don't "make power"; they control the power available from the power supply, while working against the output transformer primary impedance. When an input signal is applied to the output tubes, they will respond by attempting to suck more & less current from the power supply through the primary impedance. The tube's characteristics will set an upper limit of how much current the tube can pull from the supply. However, even if the tube has very ample current capability, if the supply voltage is low and/or the primary impedance is high, they will limit the current actually drawn to a level below what the tube is otherwise capable.
-
Example 1:
Say the supply voltage is 500v and the primary impedance is 1kΩ; also, the tube is capable of passing 50mA peak, and 2x tubes per side of the push-pull output stage. The maximum peak current the power supply/OT could deliver is 500v/1kΩ = 500mA, but the tubes can only manage 50mA x 2 = 100mA peak. Output power will be tube-limited.
Example 2:
Say the supply voltage is 500v and the primary impedance is 1kΩ; also, the tube is capable of passing 350mA peak, and 2x tubes per side of the push-pull output stage. The maximum peak current the power supply/OT could deliver is 500v/1kΩ = 500mA, and the tubes will try to draw 350mA x 2 = 700mA peak. Output power will be supply/OT-limited. To enable all the peak current the tubes could deliver, the supply would need to be at least 700v (700v/1kΩ = 700mA), or the OT primary impedance would need to be reduced to ~700Ω (500v/700Ω = 714mA).
The above examples assumed no complicating factors like power supply sag, or minimum plate voltage across the output tubes. But they should also show you the tubes are not the only consideration; supply voltage and OT primary impedance are every bit as important. Also implied, but not stated was that the power supply needs to be able to deliver the extra current for the bigger tubes and lower OT impedance. We can't take this for granted, because the high-voltage winding is rated for a set amount of current; drawing more current will cause the supply voltage to sag, limiting current output to some value near the rated current.
Practical Issue: Transformers are likely the most-expensive component on the amp, and power supply capacity costs $$$. A manufacturer building in excess unused power supply and heater capacity into an amp is increasing their production costs a LOT, while not having any extra advertise-able output power to justify a higher selling price. As a result, it's very unlikely your amp will realistically support the extra heater and B+ current draw needed to allow bigger tubes to make more real power. Additionally, you'd probably have to substitute the OT with another of lower primary impedance to allow the tubes to draw more current at the same supply voltage.
All of this adds up to likely having to buy different power and output transformers, in addition to the bigger tubes, if you want an actual power output increase.
That said, if the heater winding will support the bigger tubes, you could use them to get the same power output; there may/may not be a sense of lessened distortion at the top of the volume control range. The KT120 & KT150 claim very slightly higher transconductance than the 6550, and may hit the amp's maximum output power at a slightly lower driving signal level, which might add a slight sense of "authority" to the amp's power output.
If it were my amp, I'd stick to the stock output tubes. Maybe build a new amp if I wanted to take advantage of the capabilities of the bigger tubes.
-
Thanks for the in depth details. I think you're right with staying with what Sunn designed. Do you have an opiion on using the 6550 vs. the KT88?
-
FWIW, I tried the old Ei KT90's and was not impressed - but they are long gone. I'm not sure how close the Tung Sol (EH) tubes are to that design. I hope Ed chimes in here as he HAS tried the new 90 and 120 (not sure about the 150). He was VERY impressed with the performance of these tubes. The only thing I would check would be head clearance. They are taller than the KT88 so you may want to put a ruler back there.
Jim
-
Sorry, just saw your question about tube preference. I would take a worn out and and sputtering KT88 over a new 6550 any day of the week! I like the softer characteristics and distortion of the KT88 MUCH more than the 6550. Be wary of the EH tubes. It appears that their KT88's are just re-bottled versions of their exact 6550 design and are not even close to the original MO Valve design.
YMMV
Jim
-
I also have wondered about SS plug ins instead of the GZ34s on Sunns.
I can see that there would be no 5V current used, due to no 5V heaters.
That should make the PT run a bit cooler.
Also, the B+ will be up a litttle.
My question is this; since there is no current being drawn by the 5V secondary,
there must be more available in the primary,
so, does that mean the B+ can now draw more current?
-
I also have wondered about SS plug ins instead of the GZ34s on Sunns.
I can see that there would be no 5V current used, due to no 5V heaters.
That should make the PT run a bit cooler.
Also, the B+ will be up a litttle.
My question is this; since there is no current being drawn by the 5V secondary,
there must be more available in the primary,
so, does that mean the B+ can now draw more current?
Not really. While it may put less stress on the primary, the secondaries are discrete windings. The high voltage secondary is wound for a given spec as are the 6.3VAC and 5VAC secondary windings. Exceeding any particular secondary's current capacity will cause problems for that winding whether or not the other windings are within spec or even wired at all. Think of it like this: if you wire up a bunch of tubes to the 6.3VAC winding and exceeded its capacity it will likely burn open before too long - even if the HT winding is disconnected or in standby. Not using the 5VAC winding may make the PT run cooler and prevent a global failure but you still must be mindful of the individual winding specs. Disconnecting one secondary does not magically make the others more robust.
To me the benefit of using a solid state rectifier has always been higher B+ because the diodes do not have the same large forward voltage drop of the tube rectifier and there is no tube sag with a SS rectifier.
-
I also have wondered about SS plug ins instead of the GZ34s on Sunns.
...
Also, the B+ will be up a litttle.
I wonder if the B+ rise would be anything noticeable. They are 2x parallel GZ34's to start with, so very little voltage drop.
Regarding any potential B+ current increase:
Take volts * amps for any winding which will be un-loaded (in this case, the 5v winding * total current draw of the GZ34's, or 20VA). Now divide this VA value by the a.c. voltage of the winding from which you want extra current draw.
- In the case of the high voltage winding, calculate a probable a.c. voltage (maybe 500v/1.414 = ~354vac). Use this value to figure available current increase (20VA/354v = ~31mA).
This should give an indication if the tube removal will help. You can see you only gain ~31mA of extra reserve on the high voltage winding; however 20VA yields an extra 3.17A available for draw from the 6.3vac winding (the V in the VA matters).
It is my understanding the overall VA rating of the power transformer is the overriding factor, such that windings will behave as above. That said, if you try to pull amperes from a winding sized for milliamperes, expect a burnt winding. But the wire gauge suitable for 4-6A in the transformer is probably still heavy enough to carry a few extra amperes. But everything is always individual-part specific. We don't know what tradeoffs the transformer manufacturer made.
-
I did a little internet searching.
I found that the later version of the 200S was the 190B.
Same thing, but with SS recto.
They increased the power claim to 150 watts.
OK, we all know that factory power claims are somewhat arbitrary, but maybe they had some degree of reason to make that claim.
Also, since the Sunns were basicly Dynacos anyway, they always had pretty conservative power ratings to begin with.
My guess is that there may be some exrta power in the P/S that different tubes could make use of.
If there is any question about having enough heater current a seperate heater tranny could cure that and take more heat from the PT.
Hey, I get that the secondarys are what they are, but still I feel like it's not the extrra current that will do damage,
it's the heat from that extra current, so anything that removes heat has to benefit the longivity of the PT.
-
I did a little internet searching.
I found that the later version of the 200S was the 190B.
Same thing, but with SS recto.
They increased the power claim to 150 watts.
I dunno... This page (http://johnkvintageguitars.homestead.com/FS190B.html) says the 200S and 190B were 2x output tube amps with an 85w rating; we're talking the 4x output tube 2000s.
I'd assume Sunn's stated ratings are accurate for clean output power, not measured bent output power with the volume full-up (which are different for a number of reasons).
My guess is that there may be some exrta power in the P/S that different tubes could make use of.
...
Hey, I get that the secondarys are what they are, but still I feel like it's not the extrra current that will do damage,
it's the heat from that extra current, so anything that removes heat has to benefit the longivity of the PT.
My point in the earlier posts was that:
1) A manufacturer generally won't pay extra for something (PT capacity), then not use it. They wouldn't stay in business long wasting money in that manner (homebrew is different; you're not making 1000's of the same item & multiplying cost over-runs by that quantity).
2) The tubes are only one of 3 factors (really 4) that need adjustment to increase output power. The others are Supply Voltage, Supply Current and Primary Impedance.
So if the PT can support the heaters, the tubes won't make extra power.
If you want to double the output power of amp X, you have several options:
1. Keep supply voltage the same. Halve the primary impedance, and double the # of tubes (or double the plate current capability by using bigger tubes). The PT must be able to deliver double the original supply current while maintaining the original voltage.
2. Double the supply voltage. Keep the same OT primary impedance and individual tube peak current. This option is usually not feasible because we are unlikely to start with a design where the output tubes were utilized so far below their capabilities (especially with regard to supply voltage; high supply voltage in the original design would lead to smaller peak currents, more likely within the tube's capability).
3. Raise supply voltage and reduce OT primary impedance, resulting in some increased peak current; the factors of increase/decrease are something less than x2 and /2. This might be harder to do because OT impedances are only available in so many typical values.
Option 1 is the one seen most often, as you'll note by looking over schematic from most any amp company.
I'm assuming more power was the original intent, since we were considering swapping 42w dissipation tubes for 50w, 60w or 70w plate dissipation tubes.
-
> to double the output power
> Double the supply voltage. Keep the same OT primary impedance
No; that would 4X the power.
Double the voltage and double the impedance gets 2X power.
This is "never" practical because designers work to very-near the device voltage limits. Even if you find 6L6 working at 350V, and know that 6L6GC are rated 500V, that's not double voltage.
-
> to double the output power
> Double the supply voltage. Keep the same OT primary impedance
No; that would 4X the power.
I keep forgetting those √2, don't I? :l2:
Thanks for the follow-up! Yes, what I said here would double the voltage and double the current, for 4x power. I should have said, "Run √2 times the voltage, keep the OT primary impedance the same for double output power."
-
halve the Z and increase B+ by 1/2 - 3x power.
--pete
-
Chiming in!
Check to insure you have sufficient heater current available. I restored a 200s and had to put the filter caps in series because of the voltage increases with current to achieve a voltage I was comfortable with. I ended up with a SS rectified and loaded voltages of 550 on the plates if the 6550. Change to KT88 about the same.
I have tried KT120 and KT150. The 150 is sterile and does not distort pleasantly at all. Probably a great HIFI tube, not sure. I would have tried them in HIFI, but my Dyanco amps cannot handle the heater draw. The 120 are a different story. Very nice tube. My plate voltage dropped to 520. I use a lot of KT66 and KT88. I have some original KT88 and KT66 GEC.
I hate to admit, but in listening to the new production tubes the new Gold Lions sound better to me. When I gave the customer back the amp it had Valve Art KT88 in it. The reason is I have found them to be the most durable in the KT88's. If it were mine I would be running the KT120's in it, but it was a restoration so I kept them the same.
The difference in the heater draw on the KT120 is not a concern on the SUNN 200s.
On another note I have a friend who has been using a pair of KT120 with 569 plate voltage for almost a year running the pair wide open 5 nights a week. I know of no new production tube that will run being pushed that hard. The KT120 has a very smooth distortion and you will notice much better Bass response.
I hope this helps. There is room in the head for the 120 and enough room inside to place series filter caps inside and leave the old one to keep the original appearance. At lease that is what I did.
On Biasing the KT120, I could not find the proper recommendation. If you bias them like a KT88 they will be cold. I needed -80vDc. then 42ma looked good on the scope.