Home |
Search |
Today's Posts |
#1
|
|||
|
|||
211 Ultra Linear PP output stage??
Hi RATS!
I visited the Jolida webpage and seen they have a big monoblock called JD-3000A. This amp using 211 output tubes in Push Pull Ultra Linear output stage. DHT tube in Ultra Linear output stage, don't make any sense!! http://www.jolida.com/catalogue/jd3000a.html Cheers! |
#2
|
|||
|
|||
Tube747 wrote: Hi RATS! I visited the Jolida webpage and seen they have a big monoblock called JD-3000A. This amp using 211 output tubes in Push Pull Ultra Linear output stage. DHT tube in Ultra Linear output stage, don't make any sense!! http://www.jolida.com/catalogue/jd3000a.html Interesting. You're right, Ultra-Linear can't possibly apply to triodes! The only possible way it could make sense is if the UL taps were somehow returned to the screens of the EL34 drivers. But without a schematic or more info from Jolida, I guess we'll never know. Cheers, Fred -- +--------------------------------------------+ | Music: http://www3.telus.net/dogstarmusic/ | | Projects, Vacuum Tubes & other stuff: | | http://www.dogstar.dantimax.dk | +--------------------------------------------+ |
#3
|
|||
|
|||
Fred Nachbaur wrote: Tube747 wrote: Hi RATS! I visited the Jolida webpage and seen they have a big monoblock called JD-3000A. This amp using 211 output tubes in Push Pull Ultra Linear output stage. DHT tube in Ultra Linear output stage, don't make any sense!! http://www.jolida.com/catalogue/jd3000a.html Interesting. You're right, Ultra-Linear can't possibly apply to triodes! The only possible way it could make sense is if the UL taps were somehow returned to the screens of the EL34 drivers. But without a schematic or more info from Jolida, I guess we'll never know. Cheers, Fred Its very easy to do a UL 211 amp. Just use CFB on the OPT, a la Quad II, and have EL34 drivers, in PP, and instead of bypassing the screens of the drivers to the cathodes, for PP pentode driving, bypass the screens to the cathode FB windings of the OPT, rather like I have done at http://www.turneraudio.com.au/htmlwe...bbsfbinteg.htm But I really don't think my arrangement is any better sounding than many well done other varieties, such as having tubes such as EL34 or EL84 as trioded drivers in an LTP, with plain triode output stage, and some global FB. I might add that using a dozen KT88 produces an AB amp capable of 0.1% thd at 200 watts, which 1/10 the Jolida figures. From a single 211, 25 watts is possible, so two in all pure class A1 give 50 watts, so to get 200 watts, the class A% in the AB power must be kept low. The 13E1 also produces 25 watts SE tetrode in plain UL mode, and in class A1, two could make more than 200 watts AB1, but about only 50w in pure class A1 PP. The 211 requires a HV supply, wheras KT88 require no more than +500v. Many transmitter tubes can be driven AB2, and hence have the load voltage swing more like what a tetrode AB1 might produce, so amazing power from PP transmit tubes is possible. QUE208 is another gut buster graphite anoded tube.... 813 x 2 are good for enormous powers. Patrick Turner. -- +--------------------------------------------+ | Music: http://www3.telus.net/dogstarmusic/ | | Projects, Vacuum Tubes & other stuff: | | http://www.dogstar.dantimax.dk | +--------------------------------------------+ |
#4
|
|||
|
|||
"François Yves Le Gal" wrote: On 14 Sep 2003 02:57:25 GMT, (Tube747) wrote: DHT tube in Ultra Linear output stage, don't make any sense!! They use EL34s as drivers and *could* have UL taps on the output iron returning to these. But this would be quie a strange topology. IMO, it's rather a typo or some creative marketing droid at work. Definately marketing. UL is simply the use of local NFB. Its achieved by screen taps off the OPT, or CFB windings. But UL really is a fib in the case of the 211 PP amp. As I explained in anothyer post, UL feedback is possible from the CFB windings to the driver pentode screens, since 211 have no screens. Also possible is the use of a centre tapped OPT speaker winding arranged so balanced NFB can be applied to the driver screens. Many ways of doing it are possible, but plain old UL she ain't. The OPT needs to be pretty darn good for balanced screen FB. Jolida have no monopoly on great sound. The amps are put together in china for very low $$, and sold at prices comparable to western made, ( if they have not been all driven bankrupt.) The 60w/channel sell for around aud $3,700, or usd $2, 400, in the shop here, and 45% is shop markup, 10% is GST, and some % for freight, and the chinese maybe get $100, because the american partner wouldn't bother unless he made the lion's share. The 200 w monos are listed at usd $6,000, so they'd sell in Oz for about aud $12,000, but I dunno if that's for one mono, or a pair. They aim for the top end of the market, and its here that the dedicated amateur can easily compete, providing no damn middlemen are involved. 200 watt amps using tubes are pretty easy to build, and they won't cost 6 or 12 grand, whatever it is. When I had to service a 60/60 Jolida, which had an open primary on an OPT, I found the good channel's OPT had not been wound with any RDH4 awareness. The power supply electros had to be altered to allow B+ = +530v, instead of the design centre of 460v, since the power trannies had 220v primaries, but we regularly get 250v here. Factory bias had to be reset to a low bias current. To me, its just another cheap quality chinese made amp, without the finnesse of really good quality, and done with no respect that class A is the best. They just have large power ability, because class B action% is large, class A% is small. I do hope they have improved; they'd benefit from it. Patrick Turner. |
#5
|
|||
|
|||
You could do Triode UL by way of cathode FB as suggested by
Patrick T or you could do something I have posted on ABSE. See Triode UL. In this circuit you won't need a separately wound cathode winding. You can use an ordinary UL output transformer. The limitation is that it needs high mu triodes, something that can run in Class A2. Several tubes to do just that were developed in the 30's. Typical of these were the 6AC5 & 25AC5. The driver tube would be a medium mu triode such as the 6P5. There are several high mu transmitting triodes that could use this circuit to advantage. In the 30's double triodes which combined the driver with the output such as the 6B5, 25B5 & 6N6G were produced for Class A2 operation of the output triode. I didn't have any of these tubes in my stash so I improvised. You can use any of 6F6, 6K6, 6V6 & others with G1 & G2 tied together for the high mu triode. I found the 6F6 to work well with a mu of about 50 in this service. The driver can be any triode with a mu of 5 to 20. The pair form a composite tube with characteristics much like a pentode whose triode mu is in the range of 5 to 20, same as it's driver. The pentode mu of the composite is similar to that as a true pentode. If you try it have fun. I did!!! John Stewart Tube747 wrote: Hi RATS! I visited the Jolida webpage and seen they have a big monoblock called JD-3000A. This amp using 211 output tubes in Push Pull Ultra Linear output stage. DHT tube in Ultra Linear output stage, don't make any sense!! http://www.jolida.com/catalogue/jd3000a.html Cheers! |
#6
|
|||
|
|||
john stewart wrote: You could do Triode UL by way of cathode FB as suggested by Patrick T or you could do something I have posted on ABSE. See Triode UL. In this circuit you won't need a separately wound cathode winding. You can use an ordinary UL output transformer. The limitation is that it needs high mu triodes, something that can run in Class A2. Several tubes to do just that were developed in the 30's. Typical of these were the 6AC5 & 25AC5. The driver tube would be a medium mu triode such as the 6P5. There are several high mu transmitting triodes that could use this circuit to advantage. In the 30's double triodes which combined the driver with the output such as the 6B5, 25B5 & 6N6G were produced for Class A2 operation of the output triode. I didn't have any of these tubes in my stash so I improvised. You can use any of 6F6, 6K6, 6V6 & others with G1 & G2 tied together for the high mu triode. I found the 6F6 to work well with a mu of about 50 in this service. The driver can be any triode with a mu of 5 to 20. The pair form a composite tube with characteristics much like a pentode whose triode mu is in the range of 5 to 20, same as it's driver. The pentode mu of the composite is similar to that as a true pentode. If you try it have fun. I did!!! John Stewart Ha! Cute. Rather like a tube darlington stage with internal feedback. ;-) Cheers, Fred -- +--------------------------------------------+ | Music: http://www3.telus.net/dogstarmusic/ | | Projects: http://dogstar.dantimax.dk | +--------------------------------------------+ |
#7
|
|||
|
|||
Behold, Fred Nachbaur signalled from keyed 4-1000A filament:
snip Ha! Cute. Rather like a tube darlington stage with internal feedback. ;-) I can imagine that grid current though :-o -- Gregg *It's probably useful, even if it can't be SPICE'd* http://geek.scorpiorising.ca |
#8
|
|||
|
|||
Its very easy to do a UL 211 amp.
Just use CFB on the OPT, a la Quad II, and have EL34 drivers, in PP, and instead of bypassing the screens of the drivers to the cathodes, for PP pentode driving, bypass the screens to the cathode FB windings of the OPT, rather like I have done at http://www.turneraudio.com.au/htmlwe...bbsfbinteg.htm But I really don't think my arrangement is any better sounding than many well done other varieties, such as having tubes such as EL34 or EL84 as trioded drivers in an LTP, with plain triode output stage, and some global FB. I might add that using a dozen KT88 produces an AB amp capable of 0.1% thd at 200 watts, which 1/10 the Jolida figures. From a single 211, 25 watts is possible, so two in all pure class A1 give 50 watts, so to get 200 watts, the class A% in the AB power must be kept low. The 13E1 also produces 25 watts SE tetrode in plain UL mode, and in class A1, two could make more than 200 watts AB1, but about only 50w in pure class A1 PP. The 211 requires a HV supply, wheras KT88 require no more than +500v. Many transmitter tubes can be driven AB2, and hence have the load voltage swing more like what a tetrode AB1 might produce, so amazing power from PP transmit tubes is possible. QUE208 is another gut buster graphite anoded tube.... 813 x 2 are good for enormous powers. Patrick Turner. -- +--------------------------------------------+ | Music: http://www3.telus.net/dogstarmusic/ | | Projects, Vacuum Tubes & other stuff: | | http://www.dogstar.dantimax.dk | +--------------------------------------------+ Patrick Tuner, You could be right. The Jolida 3000A using two EL34 and two 12AT7 in the driver stage, I guess in differential mode. |
#9
|
|||
|
|||
Patrick Turner wrote:
Here is a note I sent to another back about a month ago explaining how tubes used in this experiment originated. The circuit is indeed a strange looking animal in the present day. In the 30's several options were explored to getting good audio & lots of it. This is one of them. The 59 is one of the tubes developed to fill several quite different applications. See 59 Specs at ABSE. It is the precursor of the 6F6 which I used experimentally. These tubes can be operated with normal negative grid bias or pushed into Class B2 where both G1 & G2 are paralleled & run at zero bias so that part of the time the grids are +ve. Normally you would need a triode connected driver to push the grids as you can see in one of the diagrams. The 59 can do that too. The combined dissipation for the grids in the 59 is 1.5 Watts Max. That means you could also run Class A2 where the control grid is always +ve. That is what I did. This kind of circuit was used with tubes such as the 6AC5G & 25AC5G while driven by a cathode follower such as a 76 or 6P5G. Cathode current for the driver tube flows through the control grid of the output tube. Sometimes a resistor of 25K is connected between that lead & the common buss. Tubes such as the 6AC5G are very high mu (125) & need +ve grid bias to operate properly in a linear mode. No fear of the 6F6 expiring in this circuit. Just looking at what I measured while in operation shows the plate current was only 13.65 ma while running at 258 volts. If I was to build an amplifier on this principle I would want at least double that. To get there I would need a driver of lower mu than the 6SN7 to pull the grids even more +ve. Perhaps I will try that sometime!!! After that it was only a short step to combining both functions in one envelope. Tubes that do that are the 6B5, 6N6G, 25B5 & 25N6G. Pinout for the octal versions (G) was designed to be the same as octal pentodes such as the 6F6 & 6K6GT. That way a designer could easily sub the oddball tubes into his circuit. Didn't even need a cathode resistor & it's bypass cap. For whatever reason, they didn't catch on!!! john stewart wrote: You could do Triode UL by way of cathode FB as suggested by Patrick T or you could do something I have posted on ABSE. See Triode UL. In this circuit you won't need a separately wound cathode winding. You can use an ordinary UL output transformer. The limitation is that it needs high mu triodes, something that can run in Class A2. Several tubes to do just that were developed in the 30's. Typical of these were the 6AC5 & 25AC5. The driver tube would be a medium mu triode such as the 6P5. There are several high mu transmitting triodes that could use this circuit to advantage. In the 30's double triodes which combined the driver with the output such as the 6B5, 25B5 & 6N6G were produced for Class A2 operation of the output triode. I didn't have any of these tubes in my stash so I improvised. You can use any of 6F6, 6K6, 6V6 & others with G1 & G2 tied together for the high mu triode. I found the 6F6 to work well with a mu of about 50 in this service. The driver can be any triode with a mu of 5 to 20. The pair form a composite tube with characteristics much like a pentode whose triode mu is in the range of 5 to 20, same as it's driver. The pentode mu of the composite is similar to that as a true pentode. If you try it have fun. I did!!! The biasing of your basic "triode UL" amp doesn't include for the input tranny to be biased negatively, or positively, depending on tube choice, and AB1, or AB2.whichever it has to be. For normal triodes, or trioded tubes like KT88, the CT of the driver triodes would need a -ve supply, and Rk loads. The circuit works as shown. Bias is by way of continuous grid current in the output tubes. The 6AC5 & 25AC5 were designed from the beginning to work in that rather strange (to us) way. Same for the 6B5, 25B5, Etc family, although in their case the biasing is internal. The basic biasing arrangement is shown in RCA RC-14. The result is Class A2 operation. See 6AC5 at ABSE. I didn't have a 6B5 (or 6N6G, 6AC5, 25AC5, 25N6G, Etc) to play with so I thought why not hook up an old pentode as a triode (not the usual way) & drive it with a CF. I used a 6F6 as the pentode & paralleled both sections of a 6SN7GT to run as the CF. The results in the form of my notes last year are at ABSE. The output transformer is a Hammond 125E which has a CT so I was able to try UL & triode operation of this strange pair as well. The 6SN7 plates act like the screen grid in an ordinary pentode. Looking down through the 6SN7 plate the control factor (mu) is 1/mu of the 6SN7. That is about 1/17. The results show the 6F6 mu in this configuration to be in the range of 45 to 50. See 6SN7 6F6 Amp at ABSE. And the feedback to the input triode driver cathode followers is via the plates of the driver, and methinks this would have little FB effect, compared to taking FB to a screen someplace, which is the essence of UL. FB thru the input plate is by way of the "reach through" of the plate to the input cathode and is about 1/mu. In the final versions of tubes made to operate in this condition the input plate was brought out to the pin 4 of their octal socket, so that a very easy switch with several octal power pentodes such as the 6F6 was possible. Examples of this group are the 6N6G & 25N6G. Another advantage was no requirement for a seperate bias resistor or network. AFAIK, no one bothered with the UL hookup since all these tubes soon became obsolete. The input voltage to the "compounded" tubes would be high, no? Say you have +200 vrms at the anodes of the main power triodes, then you have +100v at the UL OPT taps This is applied to the anodes of the drivers, which have to make say 40v at their cathodes to propel the output triodes if the gain is say 5. So the drivers have 140v across them, and if their gain is internally 20, or near their mu if the cathode load is high, then 7 v is needed from g-k, so 47v input is required to get the output desired. If a gain change of -30% occurs when the load falls 50%, then I leave it to you guys to work out if the anode FBG to the drivers boosts the driver output to reduce the sag in OPV gain when load falls, ie, compensate for lower gain, and make Ro of the amp effectively lower, which is the function of NFB. In my EL84 drivers with balanced screen FB, sort of UL, if one stretches the definitions, the gain of the EL84 is quite high, and only about 3vrms is needed at the EL84 grids. The CFB in the 6550 OPV AND tyhe screen FB to the drivers gives about the same result of 20 dB or normally applied global FB to a normal UL amp. If the output tubes in my case were 211, the CFB in the output stage wouldn't have much effect, because the reduction in gain is marginal, and triodes already have a low Ro. But it is possible to still get a sizable benefit from delivering FB to the driver pentode screens. Its also possible to use balanced FB to the cathode circuit of the balanced drivers, which could be in pentode mode of triode mode. But this strays too far from the definitions of UL, its simply plain old loop series voltage FB, used in 1,001 amps. I had another customer phone me about a shorted OPT in a Jolida today. Patrick Turner. John Stewart Tube747 wrote: Hi RATS! I visited the Jolida webpage and seen they have a big monoblock called JD-3000A. This amp using 211 output tubes in Push Pull Ultra Linear output stage. DHT tube in Ultra Linear output stage, don't make any sense!! http://www.jolida.com/catalogue/jd3000a.html Cheers! |
#10
|
|||
|
|||
My copy of the original test results were not very clear, so here
they are again, a little more formal & I hope easier to follow. The internal plate resistance of the composite tube drops as more FB is applied as we would expect. The tests could have used some refinements but demonstrate the principal quite well, IMO. Included this time is the test setup I used to determine the control exerted through the 6SN7 Plate lead. All posted at ABSE. Cheers, John Stewart |
Reply |
Thread Tools | |
Display Modes | |
|
|
Similar Threads | ||||
Thread | Forum | |||
common mode rejection vs. crosstalk | Pro Audio | |||
Topic Police | Pro Audio | |||
Artists cut out the record biz | Pro Audio | |||
Thordarson 22S65 Ultra Linear Output Transformer Color Code | Vacuum Tubes |