newbe questions again
What does dropping the screen by 6V do ( 1.5K resistor ) as the bias has only changed by a volt or two?
Still trying to understand why CCS vs CVS. For a CVS, would not the 317 in a standard configuration be easier than the RLD method? How do either of these relate to regulated B+ as far as why one picks one technique over another? Have not dug that out of the books yet.
What does dropping the screen by 6V do ( 1.5K resistor ) as the bias has only changed by a volt or two?
Still trying to understand why CCS vs CVS. For a CVS, would not the 317 in a standard configuration be easier than the RLD method? How do either of these relate to regulated B+ as far as why one picks one technique over another? Have not dug that out of the books yet.
On my amp each heater winding feeds 2 output tubes and 1 pre-tube. However, you can see I had to add a 0.47R 5W resistor to one heater channel to even up the voltages, I did that shortly after I got the amp, about 2 years ago. Is it possible that one winding is for pre-tubes and one for the outputs? How is yours wired? I always thought the guy winding the transformer messed up the number of turns... But then, there is/was one hum balance pot at each input tube so this suggests that each channel has its own seperate heater windings.
Ian.
Ian.
The PT heater windings should feed one channel each - as originally wired.
Otherwise the load will be substantially different on the two.
I also recommende to keep the heater hum-cancelling pots.
The HT winding from the PT tend to generate considerable noise to the heaters. This is probably due to the voltage doubler PS.
This noise can often be he heard as bzzz noise on the speakers, and is mostly elliminated when the heater GND reference is centered - by ear.
SveinB
Otherwise the load will be substantially different on the two.
I also recommende to keep the heater hum-cancelling pots.
The HT winding from the PT tend to generate considerable noise to the heaters. This is probably due to the voltage doubler PS.
This noise can often be he heard as bzzz noise on the speakers, and is mostly elliminated when the heater GND reference is centered - by ear.
SveinB
I just checked mine again, and you are correct, one heater winding per channel. I just assumed the other way because one of the twisted pairs in mine goes to a preamp tube, and one to a power tube on the other channel.
I decided not to waste the Dale resistors and bought 3 180 ohm 5W to put in parallel. I'll see how long that lasts. I had done my power calculations based on 230v. If the voltage rose to 240, the resistor rating would have been exceeded (7w).
Global feedback experiments have proved fruitless in improving damping factor or sound quality. There must be weird phase changes with frequency that mean simple feedback is not going to work. Without decent test equipment, I am just guessing though.
Gary
Sorry tvr, missed your post yesterday. From one newb to another, all I know is what I have read, and that the 1K5 resistor's purpose is to limit max current, to keep the screen grid within its 1.3W max dissipation, and help prevent an unsheduled light show.
I ordered The Book yesterday...theory seems to work as far as electron flow is concerned, ripple currents, power supplies, OPT's, grounding etc; but the whys and wherefores of what makes any particular amp sound good seems to be the total sum of all the parts and how they are connected. How to know what works best given a particular chassis and iron (like this Mengyue) must take years and years of experience. Hence my approach of trying anything I can and see what happens to the sound. So far, the voltage tripler with Suppo front end and LM317 CCS on the output tube cathodes has given very nice results. More things to try soon. Why is it that the stock mains caps (X2 type) sound better than one brand of dedicated audio coupling cap I tried - just an example of many questions I could ask.
Thanks Svein, will definitely keep that in mind.
Gary, whatever wattage you calculate for the resistor, multiply it by at least 3 otherwise it will get too hot. 15W should just cut it. On my amp where the heater windings are a bit low, I hope to be able to run it on 240V with a resistor between the secondary winding and the rectifiers. 240V will bring the heaters to 6.4V vs 5.7V on 220V. The B+ on 240V went to 292V and B2 was 430V, pushing it a bit. Nothing let go. I couldn't detect any difference in sound quality at the higher voltages.
I'll have a look with a basic 2 channel scope and sig gen on the amp this weekend and see if I can see any phase changes at different freq. I have noticed the amp seems capable of lower freq than I am used to. I'm happy without feedback at the moment, the speakers used are 4" full-range Lotus (Scottmoose)design, they are 4 ohm speakers.
I tried the LED's on the cathodes, sounds much better than the LM317's, tighter bass, clean and extended top end, vocals are good. Inspired by SY's RLD. I put 3 parallel strings of [6 x LED's plus a 4.7 ohm resistor in series] for each output tube cathode. I see the LED's modulating at medium levels, maybe that indicates the onset of class B, as in when one tube goes into cut-off? Cathode currents are well balanced at quiescent conditions. Sound quality is best yet.
Ian.
Ian.
Ah, we might be getting down to some real data now. I meant to ask what impedance speakers you were using. I have 8 ohm units. So your amp is naturally going to be more damped than mine by virtue of the different load presented to the output stage. Hence the more colouration I hear due to the lesser damping factor (a factor of 4 times less). I see the same thing on my SE amps.
I've just been playing with an SE design using the 6N3P input triode driving a 6P1P. Seems to work well as a basic common cathode stage, although the best results (given the transformers were made to suit ECL86) come with EL84 as the output.
I haven't done a lot more with the Mini because of playing with the SE amps, but the resistors in the mains are still holding up. Have you tried allowing a little more anode current through the 6N1's? I reckon that they would be down in the non-linear range with the <1mA that you seem to be running, hence why it sounds better with higher B+. I had to get a surprising amount of current through the 6N3 before it sounded right.
Gary
I have beat my head against the wall, and read everything I can find. I just plain don't get a ccs in the cathode of a PP output stage. All I can visualize a constant current transferred to the plate. It would totally ignore ( or actually track) the grid. I can only see if the 317 implementation is more like a DC current servo and the bypass cap allows the tube to work as if it had a standard resistor instead.
One valve has to go into cut-off for class B to happen. I guess the only way to see that is with a scope. The bias on the output tubes determines the class A/class B transition. An LM317 needs ~2.5V headroom so for cut-off, can it actually happen? For an easy 1 page read, check the valvewizard on push-pull outputs Valve Wizard - How to design valve guitar amplifiers
The three amps are quite different, and will sound different. They shouldn't really be compared with each other.
I would be suspicious of the EL34 amp which is listed as being parallel push pull or single ended, depending on which auction you look at.
The specs aren't right for a single ended amp.
Your choice depends largely on how much power you need.
I would be suspicious of the EL34 amp which is listed as being parallel push pull or single ended, depending on which auction you look at.
The specs aren't right for a single ended amp.
Your choice depends largely on how much power you need.
My choice would be the 4xEL84 Meng for your power needs, but I would bet that the best sounding of the three (or the easiest to MAKE sound the best) would be the EL34 SE. The one with the 8 output tubes just scares me
I am a little suspicious of the power claims for the EL34 SE though. I have seen it advertised as 25wpc, which is way too high. I would take all Meng power claims with a grain of salt.
I own a Meng 6P1 PP (modified now), by the way.
Be prepared to ground the chassis to the mains input socket at least, and note the 220v input means just that (not 230 or 240).
Gary
Yes, I just started working through Valve Wizards pages too. Good stuff, as is some of the basic articles on the TubeCAD site.
I'm not familiar with the HK amp (Harmon Kardon?), but if it has classic status, just do the recap and maybe check the resistors (especially cathode ones) and leave it be and use it as a starting blueprint to build a clone that you can play with to your heart's content.
Gary
Well, I see the upside of CCS in the output stage as inherent DC current balance through the output transformer (so long as the screen currents are balanced as well, but that's another discussion). To me, that is an important thing because unbalanced OPT currents must be sonically detrimental.
One thing which I find interesting, after reading Jung's CCS article on the 317, is some debate about whether and how much capacitor bypass is needed, given the 317 is inherently a good audio-band ripple killer (better with 2 317 in cascade). From what I read, most noise that would escape the 317 is supersonic, meaning little bypassing needed. Maybe the best bypass would be a small film cap. Might have to try it and see if I can hear any difference.
I am also going to try an alternative to boosting the B+ for the input stages. This might prove sonically good, and be a lot simpler than having yet higher voltages (and the extra PS noise of a tripler) flying around this little amp.
I still think the PS can be improved sonically as well, but haven't got the parts yet to try alternatives (UF4007 to replace 1N4007, with snubber capacitors and balancing resistors in the existing doubler). There just has to be a lot of SS rectifier switching noise running around this amp
Gary
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.