Allow me to add some confusion:
http://www.dissident-audio.com/PP_6L6/Page.html
Penthode LTP PI, just one voltage gain stage before the power stage.
Yves.
http://www.dissident-audio.com/PP_6L6/Page.html
Penthode LTP PI, just one voltage gain stage before the power stage.
Yves.
Johan Potgieter said:Just one thing bothering me in the given GU50 amplifier circuit: Surely R12 and R13 cannot be only 22K? That will give a highly unsatisfactory load to the phase inverter. I would believe that to be a misprint for 220K, but such a value would somewhat exceed the maximum specified for fixed bias operation of the power tubes.
Perhaps phase splitter anode loads of 47K/56K working into 100K grid resistors? Mmmmmm...
It is defenitely 220K. The tubes have no problems with such values.
By the way, when I see that transformers I drool...
Wavebourn said:
They do look compact, indeed, but is there any logical reason for them to be any better than Hammmond or Lundahl, to name just a few?? The datasheet does not reveal any specific infomation outside of normal specs??
I looked around at the "Experience Electronics" website....http://www.experience-electronics.de/
They have some kits based on the Circlotron principles, which I had forgotten. Anyone have experiences with the Circlotron??
Sy,
I did not look too close, but 6L6Gs did look like that lo-o-ong ago. I still have the first one I bought! (I did not blow it up? NO! - the humble beginnings of a good engineer? )
AuroraB,
Somewhere you asked about the difference between UL/triode performance. To sum up (with apology to those who have already read this from me elsewhere): UL maintains the best of pentode and triode operation - close to the high output and efficiency of pentodes, with Rp (internal tube or plate resistance) almost as low as that of a triode. (The term "ultra-linear" is not really applicable, I fear; improved linearity is not so much a feature.)
I could never see the reason to want to switch over to triode operation (ideally one cannot just switch G2s - there are other things to optimise as well.) At least one manufacturer admitted that they included such a switch because of public preference only. Yes, the design is a little more critical than for a triode. But consider the practical situation: Typically a pair of EL34s (triode) will give 12W, with 1% distortion. UL-connected they will give 25W with say 1,5% distortion. So at 12W the UL topology would only have about 0.7% distortion, thus better, and you will still have another 3 dB to go, etc. From my side I fear the triode mode of operation, though very good (I have used it often), has picked up some legend status over the years. Hope this helps.
Regards.
I did not look too close, but 6L6Gs did look like that lo-o-ong ago. I still have the first one I bought! (I did not blow it up? NO! - the humble beginnings of a good engineer?
)AuroraB,
Somewhere you asked about the difference between UL/triode performance. To sum up (with apology to those who have already read this from me elsewhere): UL maintains the best of pentode and triode operation - close to the high output and efficiency of pentodes, with Rp (internal tube or plate resistance) almost as low as that of a triode. (The term "ultra-linear" is not really applicable, I fear; improved linearity is not so much a feature.)
I could never see the reason to want to switch over to triode operation (ideally one cannot just switch G2s - there are other things to optimise as well.) At least one manufacturer admitted that they included such a switch because of public preference only. Yes, the design is a little more critical than for a triode. But consider the practical situation: Typically a pair of EL34s (triode) will give 12W, with 1% distortion. UL-connected they will give 25W with say 1,5% distortion. So at 12W the UL topology would only have about 0.7% distortion, thus better, and you will still have another 3 dB to go, etc. From my side I fear the triode mode of operation, though very good (I have used it often), has picked up some legend status over the years. Hope this helps.
Regards.
At first glance, yes, it's interesting. However, I'm not sure what is the point of it. The signal goes through one less capacitor from the bottom half of the concertina splitter. I don't think that can help much with LF stability, because there is still the cap from the top half of the splitter in the signal path. Also, with this arrangement the 6N1P is robbed of some of its plate-cathode voltage. It does force you to use a bigger common cathode resistor for the 6N1P, which could be better for balance; or, if preferred, a CCS could easily be accommodated.
Yvesm said:
talking about LTP some moons back Wayne posted this ckt. post 205-206
http://www.diyaudio.com/forums/showthread.php?s=&threadid=69042&perpage=10&highlight=&pagenumber=21
it used a 6SL7 and 6SN7 LTP. I hope it does NOT add to the confusion. For me it was a boon as both tubes are available where i shop.
input spliiters scheme that works
Take a look at Lynn Olsons Web Page http://www.nutshellhifi.com/
and click on the first mention of Karna that is highlighted, then scroll down to the first schematic and click on it to get a full schematic of Lynn's latest circuit. The input splitter from Tribute is likely the best, most expensive and hardest to obtain of any. Other folks make good ones too, but go on and read the whole article, not for the DHT propaganda, no matter how good they are, but for the power and signal theories and how they are implemented. Then go and find his Amity circuit on the same site. Both amplifiers are leading edge performers and the information in the articles can keep you from a lot of "not quite what you wanted" activities. These are public domain designs with no strings attached.
Bud
Take a look at Lynn Olsons Web Page http://www.nutshellhifi.com/
and click on the first mention of Karna that is highlighted, then scroll down to the first schematic and click on it to get a full schematic of Lynn's latest circuit. The input splitter from Tribute is likely the best, most expensive and hardest to obtain of any. Other folks make good ones too, but go on and read the whole article, not for the DHT propaganda, no matter how good they are, but for the power and signal theories and how they are implemented. Then go and find his Amity circuit on the same site. Both amplifiers are leading edge performers and the information in the articles can keep you from a lot of "not quite what you wanted" activities. These are public domain designs with no strings attached.
Bud
I am currently testing this differential driver arrangement with high GM pentodes(6688, E180F, D3a), triode connected. For now, the outcomes are very good but I am still in the testing phase. The harmonic content is PP like, meaning odd harmonics higher than even ones. The output stage is a 5998 double triode biased at -40V. Total harmonic distortion is 0,9% at full power, that is 10W with no feedback applied. I cannot figure out how Allen Wright achieved this harmonic contend with this topology since differential topology as PP output stage will cancel even harmonics, so the remaining spectrum will be dominated by odd ones. Next experiment will be to try pentode driver in pentode mode for even more gain.
Attachments
I asked Allen the question a while ago on AA, and finally got his answer today. Apparently, it was a fluke. When Allen first measured the distortion spectrum, he had unbalanced plate loads in his LTP cascode first stage. After he realized what was going on and fixed it, the even harmonics disappeared.
Western Electric developed the "Harmonizer" cirduit for their output sections, Used an RC network around the tube to eat up odd order distortion by dumping even order in, as a local feedback loop. I have no idea how you could implement this scheme for your circuit, but it is somethng to dig up and evaluate.
Bud
Bud
Hey-hey!!!,
If one needs EL/KT sized input swing, a single stage willdo fine. A pentode LTP riding a CCS will do quite well. I have found best performance when anode swing is kept above the g2 voltage. The g2 current does nothing for the loads but does count in the sum the CCS is regulating. A cascode does this a bit better, but one still has the 'here be monsters' effect when anode voltage goes below the upper grid ref.
I built a front end for a Dyna Mk.III with a cascaded pair of 6SU7>6SN7 diff amps. Loop NFB to the grounded grid of the 6SU7. With a CD player I had more gain than I knew what to do with. 6BL7>7236 was a selection that offered more useful gain structure. Better drive too...
I penned an amp for a friend, Mullard style. Input V-A triode running open loop and direct connected to a resistive LTP. 6BK7B>6SN7>807's, with loop NFB to the grounded grid of the 6SN7. There's ways I like better than the loop NFB, but this design suited my friend quite well. He's OK with building it, and there's no Terrible Sand to offend his Priests.
cheers,
Douglas
If one needs EL/KT sized input swing, a single stage willdo fine. A pentode LTP riding a CCS will do quite well. I have found best performance when anode swing is kept above the g2 voltage. The g2 current does nothing for the loads but does count in the sum the CCS is regulating. A cascode does this a bit better, but one still has the 'here be monsters' effect when anode voltage goes below the upper grid ref.
I built a front end for a Dyna Mk.III with a cascaded pair of 6SU7>6SN7 diff amps. Loop NFB to the grounded grid of the 6SU7. With a CD player I had more gain than I knew what to do with. 6BL7>7236 was a selection that offered more useful gain structure. Better drive too...
I penned an amp for a friend, Mullard style. Input V-A triode running open loop and direct connected to a resistive LTP. 6BK7B>6SN7>807's, with loop NFB to the grounded grid of the 6SN7. There's ways I like better than the loop NFB, but this design suited my friend quite well. He's OK with building it, and there's no Terrible Sand to offend his Priests.
cheers,
Douglas
Bandersnatch said:
Yes, it is silly, like Fourier transformations.
I may agree that lowering output stage gain and making it more linear will increase impact of a phase splitter, but I disagree that some linear feedback may change spector reducing power of odd harmonics generated by symmetrical output stage.
Sometimes black is blask, white is white, believe me...
Dear Wavebourn,
Ummm, nobody claimed it was linear FB that did it. It looks to me that it is taking advantage of non-linear behaviour to work. Why you think the Fourier Transform is silly raises some more questions about your means of analyzing the WEHE performance. Just because the Earth looks flat is no reason to believe that it is.
cheers,
Douglas
Ummm, nobody claimed it was linear FB that did it. It looks to me that it is taking advantage of non-linear behaviour to work. Why you think the Fourier Transform is silly raises some more questions about your means of analyzing the WEHE performance. Just because the Earth looks flat is no reason to believe that it is.
cheers,
Douglas
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.