Well the switch for a fast bleeder, no need to disconnect a slow bleeder which can stay there permanently.
An amplifier without any bleeder resistor would still be perfect, any technician who relies on bleeder resistors is an idiot.
I worked in factories and relays always fail, it is almost the number one issue in todays world. Second is fuses and lose wires. PLC bugs.
An amplifier without any bleeder resistor would still be perfect, any technician who relies on bleeder resistors is an idiot.
I worked in factories and relays always fail, it is almost the number one issue in todays world. Second is fuses and lose wires. PLC bugs.
?Zoran said:
Very few amps need a 'standby' switch. It can be omitted, so it doesn't do any harm.
So what? Power transformers still need to be kept away from the audio circuits. Having more of them means more to keep away.
Hum and noise are decreased by careful layout, and understanding grounding.
Stability is increased by understanding stability.
Safety is not affected by how many transformers you have.
Sequencing is rarely needed for domestic audio. Sequencing does not avoid the need for bleeder resistors. No effect on valve life.
Pure nonsense.
Last edited:
It is totally different context when we consider relay in on/off operation 1000 times per day, and few times switching only... Off-course that stability and life time with thousands times switching is questionable.
But in operation mode of only few times switching, good quality relays, and chosen for needed purposes are reliable.
DF96
Please do not say for something that You didn't try, implement and listened to - that is nonsense...
...
Some (many) things that You wrote is not true. For instance, about You said "sequencing" - Yes it is saving tube life. (look at the manual of Quad II inspired of that they used fast hating GZ32 tube, not recommend to turn on/off amp (without stand by option) more than 12 times a day...
...
Another thing, You say that is not an issue to have very long cathode hating wires (because transformers should be far from tube device) and give advantage to long-wires separate power supplies...
...
Next, Concept of bleeder resistor is to spend energy from the capacitance bank - NOT to be present all of time in listening. If It is present, it acting like shunt element in PS. If You want shunt it should be designed in that way. These bleeder resistors have an impact in sound quality in line, tube systems. And they are additionally source of hating in device already produce heat
...
When You have In one common core primary of say 120 or 220V power line, and 2 x 300V
for secondary, that is step-up ratio. Step down is with secondary of 5-10V. It is obvious that huge difference exist. And it is also different multi-mutual coupling factors introduced
.
Please we can talk personal experiences with aim to improve things if we can somehow.
(I dont want to answering to Your "questions" any more.)
cheers
Recent experience
Just my 2cts.
Using the Isotek mains filter high current outputs for my 300B SET's really killed the dynamics. Before I would not have expected that this could happen since the SET has got a very big over-dimensioned power supply and the expectation was that any small restriction in line (due to a filter) would be compensated by the huge PSU buffer.
NOT
With this in mind I wondered about the GZ34 rectifier that has been in the amp for approx 2 years and I swapped it for a fresh GZ34. Amazing! So wear of the rectifiers always has to be taken into account and even though some brands have got an excellent reputation on life-time ("they are still working after ages") does not guarantee they contribute to the best sound.
The only thing I can think of controlling this is to regularly also have the rectifier tubes on the tube tester.
The other thing is the thought of cathode stripping of the GZ34 since after all the rectifier tube conveniently provides the slow start of the HV for the other tubes, but the poor thing itself get's the full wack of voltage being cold. So, a standby-switch between the HV connections of the rectifier tube after all? Or, go for the EML tubes since these seem to have a built in slow start for this (if I got this right)?
(But then again EML doesn't have a GZ34/5AR4 version (yet))...
Just my 2cts.
Using the Isotek mains filter high current outputs for my 300B SET's really killed the dynamics. Before I would not have expected that this could happen since the SET has got a very big over-dimensioned power supply and the expectation was that any small restriction in line (due to a filter) would be compensated by the huge PSU buffer.
NOT
With this in mind I wondered about the GZ34 rectifier that has been in the amp for approx 2 years and I swapped it for a fresh GZ34. Amazing! So wear of the rectifiers always has to be taken into account and even though some brands have got an excellent reputation on life-time ("they are still working after ages") does not guarantee they contribute to the best sound.
The only thing I can think of controlling this is to regularly also have the rectifier tubes on the tube tester.
The other thing is the thought of cathode stripping of the GZ34 since after all the rectifier tube conveniently provides the slow start of the HV for the other tubes, but the poor thing itself get's the full wack of voltage being cold. So, a standby-switch between the HV connections of the rectifier tube after all? Or, go for the EML tubes since these seem to have a built in slow start for this (if I got this right)?
(But then again EML doesn't have a GZ34/5AR4 version (yet))...
Well, if you have the chance, try a 866 or if you do not like Mercury, a 836.
You will see that they give your amp much more microresolutiin, voicing of an adult vs. kid, etcetc... like a different amp. fir tge driver take a Telefunken az1 mesh... and welcome to nirvana.
The gz34 is a nice starting point, been there, done that. two gz34 are better than one. if you want to stay with it. I built a bridge with four.
My current favorites are 872a for power and rgn2004 mesh for driver...
... have fun!
And allow you self some six chamber chokes with amorphous core from our friend 50AE...and forget Tango and Tamura after this
You will see that they give your amp much more microresolutiin, voicing of an adult vs. kid, etcetc... like a different amp. fir tge driver take a Telefunken az1 mesh... and welcome to nirvana.
The gz34 is a nice starting point, been there, done that. two gz34 are better than one. if you want to stay with it. I built a bridge with four.
My current favorites are 872a for power and rgn2004 mesh for driver...
... have fun!
And allow you self some six chamber chokes with amorphous core from our friend 50AE...and forget Tango and Tamura after this
If I would like to allow myself some...where do I find your friend 50AE?
Interesting experience with mercury rectifiers.
I read most of the posts because this is interesting to me and true. Not sure about hi-fi, but in guitars for sure, I do know that when I replace the 5AR4 rectifier in my Fender Super Reverb with a simple solid state plug in rectifier the string response is way, way different. Whether its "better" or "worse" depends on how you want to use it playing, the song, what you're after, are you comping or playing melody/chord, etc. The sag is basically gone with the solid state rectifier plugged in; so attack, rise, sustain, decay and the distortion/tone profile all along that way is different. Sensitivity and headroom are way different too. That's why I like that amp mod where you can simply switch between solid state or tube rectification. That mod keeps the tube warm but simply flips some diodes across it or not. The backup diodes are optional though:
Amp Mods
I dont know if such a setup can be made to quickly (and safely) switch between 4 or 5 tube rectifiers while the amp is playing a song? For Hi-Fi one would need to do immediate switching to detect subtle differences without a 10 minute wait between tests. This would eliminate much of the acoustical psychology and make the test more physical. With guitar playing both your hands and your ears tell you how blatantly and obviously different the rectifiers are. You have more sensory inputs. I'm not sure a listener in the audience would tell though in an A/B test even.
I think the key to setting up a successful test of this theory in hi-fi would be to devise a circuit that lets you switch rectifiers with a rotary switch while music plays to eliminate the acoustical psychology and memory aspects. And also to use headphones to further eliminate the room acoustics, reflections, resonances and listening position variables. In other words strip out as many variables as one can and make the switching of rectifiers very fast. Measuring voltages, frequency response, distortion obviously has its place too, but that's not what the OP is getting at I think.
Amp Mods
I dont know if such a setup can be made to quickly (and safely) switch between 4 or 5 tube rectifiers while the amp is playing a song? For Hi-Fi one would need to do immediate switching to detect subtle differences without a 10 minute wait between tests. This would eliminate much of the acoustical psychology and make the test more physical. With guitar playing both your hands and your ears tell you how blatantly and obviously different the rectifiers are. You have more sensory inputs. I'm not sure a listener in the audience would tell though in an A/B test even.
I think the key to setting up a successful test of this theory in hi-fi would be to devise a circuit that lets you switch rectifiers with a rotary switch while music plays to eliminate the acoustical psychology and memory aspects. And also to use headphones to further eliminate the room acoustics, reflections, resonances and listening position variables. In other words strip out as many variables as one can and make the switching of rectifiers very fast. Measuring voltages, frequency response, distortion obviously has its place too, but that's not what the OP is getting at I think.
Last edited:
Sagging Power Supplies and non-HIFI guitar amps...
I am a competent circuit designer, WindCrest (OK, OK, that can be hotly debated, but it is at least sometimes confirmed here by others, LOL!). The first time I got ahold of a legendary Guitar Amp schematic, I kind of just sat there, looking, looking, scowling, looking, calculating, looking some more. Eventually calling my guitarist friend (who was lusting for the thing), asked him how it sounded when played hard. He kind of just couldn't stop creaming in his shorts.
I was incensed. The design was almost the antithesis of good hi-fi amplifier design.
Too much gain all over the place.
But this creates 'adjustable' distortion.
Absolutly ridiculously high power supply impedance.
But this creates non-linear sustain-of-note distortion effects.
Indefensible use of series resistors on cathode heaters.
But starved cathodes increase harmonic distortion.
Introduction of all nature of intentionally nonlinear germanium diodes in tone section.
Again … to increase harmonics, change things up a bit.
It was like an amplifier made of gerbils, wheels, springs and packing tape; Yet … my friend was saving up the $3,000 it'd cost to buy one. Not even new. Used. With bad tubes. Three thousand smackeroons!
________________________________________
Needless to say, I had to go hear one. It was a real ear-opener. After listening to a quite-talented guitarist wail away on the thing with a trio of excellent custom-wound pickup guitars, I was a convert. Clearly not Hi Fi. But just as clearly, awesomely musical. In a way that almost defied description.
________________________________________
So… after that I went back and bought an updated schematic of the amp that was demoed. And I put together a most difficult successive-approximation circuit simulation program, to see if I could mathematically produce anything like the harmonics and over-time macro-structures this amp produced. After many weeks of fiddling with the program, after actually recording my friend's guitar pickups with a digitizer and using them for infinite-impulse-response input to the algorithms, I started to see how it worked.
Basically, the core to everything is to have a power supply that is weak, and a chain-link-fence of successively weaker stages, to drive stages ever closer to the input jack. That stage has the weakest B+ supply of them all.
Your observation that changing out the 5AR4 for a solid-state plug-replacement totally changed everything is completely congruent to what my algorithms showed. That the much, much lower VFORWARD voltage drop of the silicon rectifiers, and especially their lower transfer resistance, was removing one of the most important “weak power supply” weaknesses at the source! Sturdied up, your harder-to-trick power supply is more capable of delivering power without as much DC 'mushiness', which of course is exactly the opposite of the design.
So… it makes sense.
Oddly enough.
Just Saying,
GoatGuy ✓
I am a competent circuit designer, WindCrest (OK, OK, that can be hotly debated, but it is at least sometimes confirmed here by others, LOL!). The first time I got ahold of a legendary Guitar Amp schematic, I kind of just sat there, looking, looking, scowling, looking, calculating, looking some more. Eventually calling my guitarist friend (who was lusting for the thing), asked him how it sounded when played hard. He kind of just couldn't stop creaming in his shorts.
I was incensed. The design was almost the antithesis of good hi-fi amplifier design.
Too much gain all over the place.
But this creates 'adjustable' distortion.
Absolutly ridiculously high power supply impedance.
But this creates non-linear sustain-of-note distortion effects.
Indefensible use of series resistors on cathode heaters.
But starved cathodes increase harmonic distortion.
Introduction of all nature of intentionally nonlinear germanium diodes in tone section.
Again … to increase harmonics, change things up a bit.
It was like an amplifier made of gerbils, wheels, springs and packing tape; Yet … my friend was saving up the $3,000 it'd cost to buy one. Not even new. Used. With bad tubes. Three thousand smackeroons!
________________________________________
Needless to say, I had to go hear one. It was a real ear-opener. After listening to a quite-talented guitarist wail away on the thing with a trio of excellent custom-wound pickup guitars, I was a convert. Clearly not Hi Fi. But just as clearly, awesomely musical. In a way that almost defied description.
________________________________________
So… after that I went back and bought an updated schematic of the amp that was demoed. And I put together a most difficult successive-approximation circuit simulation program, to see if I could mathematically produce anything like the harmonics and over-time macro-structures this amp produced. After many weeks of fiddling with the program, after actually recording my friend's guitar pickups with a digitizer and using them for infinite-impulse-response input to the algorithms, I started to see how it worked.
Basically, the core to everything is to have a power supply that is weak, and a chain-link-fence of successively weaker stages, to drive stages ever closer to the input jack. That stage has the weakest B+ supply of them all.
Your observation that changing out the 5AR4 for a solid-state plug-replacement totally changed everything is completely congruent to what my algorithms showed. That the much, much lower VFORWARD voltage drop of the silicon rectifiers, and especially their lower transfer resistance, was removing one of the most important “weak power supply” weaknesses at the source! Sturdied up, your harder-to-trick power supply is more capable of delivering power without as much DC 'mushiness', which of course is exactly the opposite of the design.
So… it makes sense.
Oddly enough.
Just Saying,
GoatGuy ✓
silicon rectifiers in tandem with tubes rectifiers is what i use if ever,
design and build of power transformer then becomes easy and insulation requirements not as strict as when doing center tapped with higher voltages...
i did a pp amp with 16 kt88's, and i used a flatpack bridge rectifier rated for 1000v 16 amperes, i would have used 25 or even 50 if i had them then, they are dirt cheap, i will never hesitate to use...
design and build of power transformer then becomes easy and insulation requirements not as strict as when doing center tapped with higher voltages...
i did a pp amp with 16 kt88's, and i used a flatpack bridge rectifier rated for 1000v 16 amperes, i would have used 25 or even 50 if i had them then, they are dirt cheap, i will never hesitate to use...
Real time switching is not useful at all as far as i am concerned. It has zero value for the development of one's system and stunts hearing acuity. Cannot imagine it is even useful for newbies. If you are not willing to invest time and effort in training your aural memory hifi becomes a pointless pursuit.
It's quite true what Mr. GoatGuy and others say about tube rectification in guitar amplifiers. Anyhow, these amplifiers are quite the opposite of HiFi reproduction. They're sound producers. In HiFi in it's correct sense you do NOT want the PSU to sag, because you do NOT want distortions at all. Otherwise, what you really want is an effects box.
Best regards!
Best regards!