Can YOU hear the difference in Tube Rectifiers?
- Thread starter Wharfcreek
- Start date
I believe that to fully understand the difference one needs to have at least a modicum of electronics knowledge and understanding, we all know that changing from tube rectification to SS in a Fender or the like will result in a huge difference as noted by WC. Why? Because the SS won't allow the output to "sag" like a tube rectifier will, that is quantifiable and audible. And this is more likely to be heard in a git amp as it is almost always operated at the edge of its power for musical reasons, not like a tube audio amp being used for musical reproduction reasons. Are there differences in SS and vacuum diode operation? Of course, they are two totally different electron devices. Is that audible? That is the real question, something must be going on if switching noise is being dealt with by circuitry changes (snubbers), diode doping or totally different semiconductor topology (SiC diodes, etc). Is that audible? Of course, not using the same type of rectifier (5V4 instead of 5AR4) is just bullshit, you cannot make any value judgements by that means, anyone saying otherwise is just talking out of their ass. But what about same types, different brands? One thing for sure, the new stuff is not made as good or reliable or as long lasting as the old stuff, how much of that also affects the sound is another question. Two things are for certain, 1) different people like different things sound-quality-wise in their systems, speed, clarity, fullness, etc. so obviously one person's important like may be another's less important and vis-versa. And 2) the most important difference is how a system gets the details, the details are where the REAL difference lies in a really good system and not-so good, and how ALL the components, electronic and otherwise, work separately and together will have a major impact on that...
part of my own unwillingness to hear rectifiers is wallet self-defense. The one amp I own with 5AR4's uses a pair of them. I'm running $35 worth of Sovtek rectifiers in there currently. I do not wish to believe that $200 worth of Mullard rectifiers would sound 5x better. Its just too damn expensive of an experiment for me.
The other stuff mostly all uses a 5U4GB, and those things are cheap as chips. I've got boxes of them, so no worries about old being better than new.
I've never spent any significant time switching rectifiers out, but the one that I clearly remember was going from a 5U4GB to a 5U4g. The amp sounded noticeably worse at higher volumes, and I could measure significant voltage sag on the B+ line. I will not make any claims to have ever tested that tube, but it sure acted like it was about half dead. I stuck another 5U4GB in there, and the issues went away. It was a Magnavox console amp, so its not like I was running something with unusual demands.
I should see if I still have that tube, and test it if I do. i'm sort of curious if the tube tester will bear out the verdict of my voltmeter and my ears.
The other stuff mostly all uses a 5U4GB, and those things are cheap as chips. I've got boxes of them, so no worries about old being better than new.
I've never spent any significant time switching rectifiers out, but the one that I clearly remember was going from a 5U4GB to a 5U4g. The amp sounded noticeably worse at higher volumes, and I could measure significant voltage sag on the B+ line. I will not make any claims to have ever tested that tube, but it sure acted like it was about half dead. I stuck another 5U4GB in there, and the issues went away. It was a Magnavox console amp, so its not like I was running something with unusual demands.
I should see if I still have that tube, and test it if I do. i'm sort of curious if the tube tester will bear out the verdict of my voltmeter and my ears.
yep, the 5AR4 is a tough one in vintage, don't know if there is a difference sound-wise, but compared to the new issues, the vintage, and this is true for all vintage tubes, don't hold up near as well and can really go bad fast. This is particularly an issue for the 5AR4, however there is an old ham trick that can save those chinese firecrackers, so I have been told. The other thing that I find crazy is the price of the "select' new issues, they are costing almost as much as vintage, and I just don't think they have the cohones!
LOL. At least your honest.
I have never considered myself a collector of tubes but at one point i had more than 30,000 tubes. During the last two and one half decades i have sold most to people and manufactures in Japan and Hong Kong. I seemed to always be at the right place at the right time as far as finding tubes at one point in my life.
Back on subject, i have listened to every 5AR4 tube that i know of and after 1000`s of hours of listening there are differences between brands, sometimes slight and sometimes obvious. A lot does depend on the equipment, two example`s, a dynaco 70 it did not seem to make one bit of difference or with a Scott 222c. Now fast forward to today, with my LM 518IA 845 amp, any tube change be it the the rectifier tubes, driver tubes or the pre amp tubes one does not have to search for a change in the music for the good or bad it`s there in front of you. Replacing the 5AR4 tube with the Philips 5R4GYS in the LM 518 has made a remarkable change and for me it`s for the better, to be honest i do not know what i was expecting but the tube has put me on the edge of my seat to the point where i`m looking for my Brimar 5R4GY.
Regarding the SS rectifier i guess i`ll have to find out for myself.
battradio
Electron trainer
The power supply of an amp is like a foundation of a building , the better it is built the less likely minor changes will any difference in the sound of the amp .
I have heard claims that they can hear differences between brands of 5U4's not just between a 5AR4 and a 5R4 , if both 5U4's are good i don't know anyone could tell the difference by listening and is total BS as for as i'm concerned .The voltage Difference of lets say two volts can't be heard , now the difference of 70 volts between a 5R4 and a 5AR4 can be heard easily if the tube amp has fixed bias and the bias wasn't reset when the rectifier tubes where changed ..
Amps with cathode bias it wouldn't be as noticeable .
The impedance of the power supply is set by the filtering , and the current capability at at different frequencies , a SS rectifier will provide a lower impedance over the audio band , a UF4007 would be less noisy than a 1N4007 , A choke filter supply and snubber caps as well as low ESR caps would make the noise they generate a moot point because it couldn't be heard . A tube rectifier generates less noise , but can't supply the current as well at lower frequencies so it doesn't need to be filtered as well for noise , but need a better choke and more filter caps to proved ample current for lower frequencies .
On a scratch build i will use a power transformer rated for a 50 watt amp on a 10 watt amp and a transformer rated for 100 watts for a 25 watt amp . They stay cool to the touch and the voltage doesn't sag under load , also i use at least two filter chokes and three if they will fit in the chassis along with the lowest ESR caps that i can afford so the ripple and impedance of the power supply is as low as it can be .I also always use cathode bias so any change of rectifier tubes or output tubes can't be heard at all
A dynaco ST-70 has a barely large enough power transformer , a too small of rectifier tube at full power out , minimal filter capacitance and too small a choke , its power supply is equivalent to the leaning tower of pisa's foundation , make any change in type of the rectifier tube and it is certain to be heard , especially if the bias isn't reset .
I have heard claims that they can hear differences between brands of 5U4's not just between a 5AR4 and a 5R4 , if both 5U4's are good i don't know anyone could tell the difference by listening and is total BS as for as i'm concerned .The voltage Difference of lets say two volts can't be heard , now the difference of 70 volts between a 5R4 and a 5AR4 can be heard easily if the tube amp has fixed bias and the bias wasn't reset when the rectifier tubes where changed ..
Amps with cathode bias it wouldn't be as noticeable .
The impedance of the power supply is set by the filtering , and the current capability at at different frequencies , a SS rectifier will provide a lower impedance over the audio band , a UF4007 would be less noisy than a 1N4007 , A choke filter supply and snubber caps as well as low ESR caps would make the noise they generate a moot point because it couldn't be heard . A tube rectifier generates less noise , but can't supply the current as well at lower frequencies so it doesn't need to be filtered as well for noise , but need a better choke and more filter caps to proved ample current for lower frequencies .
On a scratch build i will use a power transformer rated for a 50 watt amp on a 10 watt amp and a transformer rated for 100 watts for a 25 watt amp . They stay cool to the touch and the voltage doesn't sag under load , also i use at least two filter chokes and three if they will fit in the chassis along with the lowest ESR caps that i can afford so the ripple and impedance of the power supply is as low as it can be .I also always use cathode bias so any change of rectifier tubes or output tubes can't be heard at all
A dynaco ST-70 has a barely large enough power transformer , a too small of rectifier tube at full power out , minimal filter capacitance and too small a choke , its power supply is equivalent to the leaning tower of pisa's foundation , make any change in type of the rectifier tube and it is certain to be heard , especially if the bias isn't reset .
battradio
Electron trainer
One point in favor of tube rectification - they will survive (within reason) an EMP whereas most solid-state devices will pop. Granted, this is very much an edge case, but if you want to be spinning tunes after the apocalypse...
-D
The tube might survive but the power transformer will fry itself from the induced magnetic field :tears:
+1, this is well noted by Dave Gillespie in article regarding the ST-70. In dealing with manufactured equipment, one must deal with the fact that the electronics is built to a price and thus any extemporaneous expenses would have been done away with, you can improve somewhat those deficiencies, most of these engineers knew what they were doing but were under cost constraints...
whoaru99
Epic Member
When multiple changes are made it's really difficult to pin down the exact cause of any given effect.
In my most recent build I used the same exact tubes, same exact OPTs, same exact power transformer, but with SS rectifier instead of GZ34. No choke for the plate supply, only 500uf low ESR capacitance right from the UF4007s. Bass is impressively low and tight.
Problem is, there are other changes like a completely different driver stage and fixed bias vs. cathode bias.
+1. That is the reason i was such a fan of TAS when they first came on the scene. Get rid of your white papers and all of the fancy specs. all that matters is what you hear.
Being an amp and speaker manufacture at the time nothing rang soooo true.
never made any claims about being anything but a cheapwad. I just can't make myself spend more on a pair of used rectifiers than I paid for a sleeve of NOS output tubes. It sounds quite nice as it is anyway.
My Pilot is that way to some extent. Its got plenty of rectifier (dual 5U4GB) and seemingly enough transformer, but the choke is laughable. Think the ST-70 one is small? The one in a Pilot 260 is actually smaller. Unfortunately theres just no room to upgrade it. I made up the difference with bigger caps and basically I hope for the best.
battradio
Electron trainer
Theres just no chassis or under-hood space on an SA-260 to do much beyond up the existing cap values. I struggled to re-cap it even with a pair of top mount cans. There are two I think under the hood, and they barely fit. The choke is on top, basically wedged between the rectifier, the power transformer, and where the power jacks (now power switch) lived. I might be able to squeeze a little bit bigger, but probably not much more than the size of the ST70 choke in there. If anyone made a tall one with a small footprint, thats about all that I could fit. Ive got vertical room. No way to pull off dual chokes either, there isn't the space to fit the extra caps required to split the supply.
gullaz
Active Member
I have a collection of new old stock gz34 or ez81... I started gathering them when I realized that they are the most important tube if you want your amp to work properly and found out that quality of new production rectifiers is generally very low.
Of course you can hear a difference between a good used Mullard gz34 and a new Sovtek one when the latter does not put out enough current resulting in low B+ or worse it fails (and to my experience it will, sooner than later).
If I try and compare good old gz34's, let's say a well respected (and very expensive) Philips metal base and a cheaper but good Mullard from the late sixties then I have to admit that I can't hear any difference (maybe I'm deaf?).
Of course you can hear a difference between a good used Mullard gz34 and a new Sovtek one when the latter does not put out enough current resulting in low B+ or worse it fails (and to my experience it will, sooner than later).
If I try and compare good old gz34's, let's say a well respected (and very expensive) Philips metal base and a cheaper but good Mullard from the late sixties then I have to admit that I can't hear any difference (maybe I'm deaf?).
Warfcreek,
Switching spikes from ss diodes can be alleviated by capacitors across them: Some 10n - 47nF as is found suitable. One can also use Schottky or other fast-switching diodes. Then I got a useful tip from a well-known member Eli D. on another forum: He mounts a r.f. choke succeeded by a 1nF capacitor to ground right after the rectifier(s). This takes care of some r.f. hash induced by the diode switching transients. But the 'commutating' caps are fairly widely used (use a high enough voltage rating to cope with peak voltages across them!)
Regarding your remark about tube cathode stripping: This has pretty much been found not to occur with voltages common to audio amplifiers. One seems to need KVs before this can become a factor.
Switching spikes from ss diodes can be alleviated by capacitors across them: Some 10n - 47nF as is found suitable. One can also use Schottky or other fast-switching diodes. Then I got a useful tip from a well-known member Eli D. on another forum: He mounts a r.f. choke succeeded by a 1nF capacitor to ground right after the rectifier(s). This takes care of some r.f. hash induced by the diode switching transients. But the 'commutating' caps are fairly widely used (use a high enough voltage rating to cope with peak voltages across them!)
Regarding your remark about tube cathode stripping: This has pretty much been found not to occur with voltages common to audio amplifiers. One seems to need KVs before this can become a factor.
jaymanaa
RIP 1961-2018
Now don't forget, for it to work properly, the amp has to be plugged in and turned on. It doesn't hurt to make sure you have at least a 30 amp fuse and breaker too.
(just kidding Craig)! Oh, I forgot to mention that the operator should be shoe-less standing in a mud puddle too.
NOSValves
Super Member
Now don't forget, for it to work properly, the amp has to be plugged in and turned on. It doesn't hurt to make sure you have at least a 30 amp fuse and breaker too.
Hey no worries I'm changed my mind and I'm going to use my wife's dishwasher I just have to install an electrical outlet inside. I have plenty of 30 amp fuses I've taken out of gear calling for 2 amp....
I guess I'm ready for business......promotional thread to follow...