Hi Ian,
Did you manage to compare this J&K OPT with Tango XE-20S that you have? What's the outcome? Please share with us.
I'm looking for OPT and might be getting them if it is as good as the Tango XE-20S. Tango is out of production now and the NOS Tango XE-20S is fetching such a high price.
Thanks!
Did you manage to compare this J&K OPT with Tango XE-20S that you have? What's the outcome? Please share with us.
I'm looking for OPT and might be getting them if it is as good as the Tango XE-20S. Tango is out of production now and the NOS Tango XE-20S is fetching such a high price.
Thanks!
B+ arrived at 428V resulting in only slightly higher voltages than your schematic. PS ripple is 350mV coming from a dual rail of 20H-15uF pio after the first 4,4uF cap common to both channels. I could cancel output AC to a minimum of 4,8mV, there's no hum from my 100+dB speakers. Bass extension is not the first quality of ISO XE-20S but it's okay compared to my 100W/8R semi-amp. Working well.
I have a pair of nice 2K5 outputs I'd like to try. Is it much trouble to re-render the schematic for 2K5?
I have a pair of nice 2K5 outputs I'd like to try. Is it much trouble to re-render the schematic for 2K5?
Hi GB, Rdc is 200 ohms.
The amp is playing for a couple of hours now and dynamics have improved. Bass is still a little boomy so there's room for improvement. Great on horns.
* There's 236V (420-184) over the Shuanwang 2A3. I could up the PS another 20 volt by picking a higher tab, what would raise dissipation just a little over the max. Any better for the damping factor or would this only be an theoretical effort?
The amp is playing for a couple of hours now and dynamics have improved. Bass is still a little boomy so there's room for improvement. Great on horns.
An externally hosted image should be here but it was not working when we last tested it.
* There's 236V (420-184) over the Shuanwang 2A3. I could up the PS another 20 volt by picking a higher tab, what would raise dissipation just a little over the max. Any better for the damping factor or would this only be an theoretical effort?
Last edited:
Nice iron, dude
The voltage across the 2a3 could be a little higher, but has no effect on damping. For that, you would need to wire in some feedback. This can be done without going around the OPT as well, which people regularly report not to have any sonic penalties. I will try to sketch up something later tonight..
The voltage across the 2a3 could be a little higher, but has no effect on damping. For that, you would need to wire in some feedback. This can be done without going around the OPT as well, which people regularly report not to have any sonic penalties. I will try to sketch up something later tonight..
Ok, here's a scheme for your 2.5K OPTs.
I used the datasheet operating point for the 2A3 and lowered the B+ to 425V (for increased current draw, 60mA thru the 2A3). Please double check after re-wiring.
Note the bootsrapping cap on the 6SL7 must be increased to 1uF to yield proper bass extension and damping down to 20Hz. The optimum value for the ultrapath cap is 39uF. This provides the flattest response down to 20Hz, and does not compromise damping.
I will sketch up a variant with some nfb later.
I used the datasheet operating point for the 2A3 and lowered the B+ to 425V (for increased current draw, 60mA thru the 2A3). Please double check after re-wiring.
Note the bootsrapping cap on the 6SL7 must be increased to 1uF to yield proper bass extension and damping down to 20Hz. The optimum value for the ultrapath cap is 39uF. This provides the flattest response down to 20Hz, and does not compromise damping.
I will sketch up a variant with some nfb later.
Attachments
Fantastic job GB, I'll build it in a couple of days. At the moment the Shuangwangs are substituted by Sovteks, so there's more listening to do
C1 is 220nF in Soulmerchants setup, could it be causing phase shift in the lows?
I'm asking because of my compact build it's hard changing parts.
* Ian gives the answer here: http://www.diyaudio.com/forums/tubes-valves/258790-3-direct-coupled-2a3-amps-7.html#post4290818
C2 (39uF) should be equal to the last cap on the PS, is it not? The nearest stock size would be 47uF electrolytic I guess...
I wondered what the setup procedure is. In my case the 2A3 is powered from a not center tapped 2,5Vac incorporating a hum pot [33R - 25R ww pot - 33R]. With the hum pot I can cancel out all residual 50Hz on the speaker output. Is the variable U5 to be used with an oscilloscope?
C1 is 220nF in Soulmerchants setup, could it be causing phase shift in the lows?
I'm asking because of my compact build it's hard changing parts.
* Ian gives the answer here: http://www.diyaudio.com/forums/tubes-valves/258790-3-direct-coupled-2a3-amps-7.html#post4290818
C2 (39uF) should be equal to the last cap on the PS, is it not? The nearest stock size would be 47uF electrolytic I guess...
I wondered what the setup procedure is. In my case the 2A3 is powered from a not center tapped 2,5Vac incorporating a hum pot [33R - 25R ww pot - 33R]. With the hum pot I can cancel out all residual 50Hz on the speaker output. Is the variable U5 to be used with an oscilloscope?
Last edited:
I double checked and indeed, when it comes down to damping, there seems to be little difference in using 1uF of 220nF. 1uF however gives a nearly perfect linear frequency response and less phase shift. With 220nF there are some dips or around -0.3dB in the low bass.Whatever that means to your listening setup.
39 µF is a standard size. Jantzen cross cap are available for less than 10€ each in that size. There are also some motor run caps in 39 or 40 µF.
First , null out any filament hum. Then null out any residual hum using the "U5" trimmer in the L-W network. Then, optionally go back to the filament humbucking pot. If you can't hear any residual hum, it's probably good enough. Otherwise, a simple multimeter connected across the speaker terminals could probably be used as well. Connect a speaker or dummy resistor before you go!
Here's a possible way to include some nfb, without going 'round the opt. Damping factor rises to ~4. Phase shift is reduced as well, but beware the frequency response can get worse if your opts are not top quality! If you're using rather mediocre opts, it's probably better to include them in the feedback loop.
Attachments
Member
Joined 2009
Paid Member
Well, I'm confident in my tranny builder's knowledge, no worries there, but I think the feedback is complicating the topology which main feature was independence from PS quality. If you predict stellar results I'm happy to build it but not for the experiment's sake
Perhaps Ian has some experience to share... cuz whatever I tried the bass response remained not linear. A further dropping of the 6SL7 current raises ri while amplification diminishes, not helping the triode factor.
Right now I'm attracted to Kevins SRPP setup.
Perhaps Ian has some experience to share... cuz whatever I tried the bass response remained not linear. A further dropping of the 6SL7 current raises ri while amplification diminishes, not helping the triode factor.
Right now I'm attracted to Kevins SRPP setup.
Keep in mind that "resistance at the plate" is with self bias and not cathode bypass. The normal "Rp" as shown on data sheets is, I believe, calculated with a bypassed cathode.
Sheldon
There's exactly one more resistor in the NFB circuit, not really complicated. OK, the signal passes a cap as well, you might feel critical about that. The L-W compensation however is not impaired in any way.
The effect is kind of hard to predict, it probably depends on your speaker combination above all other things. But since you were talking of boomy bass, that's one area most likely to improve..
There's a reason I used 35uF film for C2 - it was closest to optimal value. If I had 40uF in the right size I would have used that instead. ASC X386 40uF caps exist - but I didn't have the smaller 330VAC ones in my supplies stock... I would not consider 47uF electrolytic.
Increasing the bootstrapping cap to 1uF didn't make much of a difference on my testing speakers.... I guess I could try it again later though.
Sorry I haven't been following up on this thread... I have been away at a conference last week. I will hopefully pick it up again in the next day or two.
I'm not fond of the use of R6 in goldbeer's schematic... Admittedly, my version has all triodes tied at the B+... but it seems to work nicest for me this way
I don't think the feedback loop is something you really want to do either...
Ian
Last edited:
Just re-calc the load line and adjust the anode load of the 6SL7 accordingly to get correct bias.
Ohh it looks like RCA already has the 2.5K load line on the specs sheet...
I prefer the 3.5K load line for 2a3.
Well, I'm confident in my tranny builder's knowledge, no worries there, but I think the feedback is complicating the topology which main feature was independence from PS quality. If you predict stellar results I'm happy to build it but not for the experiment's sake
Perhaps Ian has some experience to share... cuz whatever I tried the bass response remained not linear. A further dropping of the 6SL7 current raises ri while amplification diminishes, not helping the triode factor.
Right now I'm attracted to Kevins SRPP setup.
Perhaps it would help to post your power schematic? I will hopefully post mine later today after a bit more testing.
I have found small changes in the power supply have a significant impact on the sound. I am testing a flywheel power supply (much like Sheldon's).
Kevins circuit is a nice one too. I just don't have a 100Hy choke at the moment...
Ian
An externally hosted image should be here but it was not working when we last tested it.
At first the four 15uF 440Vac were used two per channel, one on B+ and one for cathode decoupling. Later I doubled two of them for B+ and tried many cap sizes for cathode decoupling. Right now there's 150uF in that position and I did some experimenting with serial resistance what softened bass but this is not the way to go of course. Bass is like a mule is kicking a concrete stable wall: too high dynamics and too loud overtones compared to a 100W/ch SS.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.