I briefly skimmed through this thread, over time I will finish reading through the whole thing. BUT, in the meantime I would like to continue the discussion because I constantly keep seeing this pop up in audio circles and forum groups I am a member of.
An example quote is as such; "I just changed my Sovtek 5AR4 with a NOS Mullard GZ34 and the difference is night and day!! The bass is much tighter and the midrange is much smoother with much more 3 dimensional sound."
I personally have never experienced this. The most I have noticed which I am still not certain is when switching out a tube rectifier that has slightly high internal resistance with one that is lower the amp is stiffer. I have tested this by watching the B+ during testing, I'll see slightly more sag in the voltage. Some rectifiers like the 5Y3 show this more than say a 5AR4. I have never seen distortion measurements improve with swapping a similarly tested one for another. Example I have two 5AR4 from different manufacturers and they both test the same for voltage drop across them with a test shunt load. Maybe I can't measure low enough but then again I don't know why I would get less/more distortion from a rectifier swap. My problem is even if it is possible can people hear that?
The problem I have with some of these people if not all of them is that they don't test the rectifiers at all, they buy them on ebay or wherever and just roll them in and out and claim huge sonic attributes. I can completely see having a tired rectifier or ones with drastically different rp have an impact on transients being able to keep the voltage more steady.
No joke I have heard people make these claims in Class A preamps with tube rectifiers where there isn't any large current variations in the supply at all and a 5AR4 just loafs along passing 20mA.
Maybe I'm crazy, or maybe I can't hear that good but to me unless there is something wrong, or drastically different performance difference from two of the same rectifiers I can't hear or measure any difference.
Reliability is what I look for in certain brands. Like why Mullard GZ34's got popular is they are bullet proof and last a long time.
EDIT: I am mainly want to talk about swapping one type. i.e. swapping a 5AR4 for another 5AR4 and claim huge sonic gains.
An example quote is as such; "I just changed my Sovtek 5AR4 with a NOS Mullard GZ34 and the difference is night and day!! The bass is much tighter and the midrange is much smoother with much more 3 dimensional sound."
I personally have never experienced this. The most I have noticed which I am still not certain is when switching out a tube rectifier that has slightly high internal resistance with one that is lower the amp is stiffer. I have tested this by watching the B+ during testing, I'll see slightly more sag in the voltage. Some rectifiers like the 5Y3 show this more than say a 5AR4. I have never seen distortion measurements improve with swapping a similarly tested one for another. Example I have two 5AR4 from different manufacturers and they both test the same for voltage drop across them with a test shunt load. Maybe I can't measure low enough but then again I don't know why I would get less/more distortion from a rectifier swap. My problem is even if it is possible can people hear that?
The problem I have with some of these people if not all of them is that they don't test the rectifiers at all, they buy them on ebay or wherever and just roll them in and out and claim huge sonic attributes. I can completely see having a tired rectifier or ones with drastically different rp have an impact on transients being able to keep the voltage more steady.
No joke I have heard people make these claims in Class A preamps with tube rectifiers where there isn't any large current variations in the supply at all and a 5AR4 just loafs along passing 20mA.
Maybe I'm crazy, or maybe I can't hear that good but to me unless there is something wrong, or drastically different performance difference from two of the same rectifiers I can't hear or measure any difference.
Reliability is what I look for in certain brands. Like why Mullard GZ34's got popular is they are bullet proof and last a long time.
EDIT: I am mainly want to talk about swapping one type. i.e. swapping a 5AR4 for another 5AR4 and claim huge sonic gains.
Last edited:
I take this as you also haven't experienced a drastic sonic difference when using like for like rectifiers. Are you curious at all if these people are actually hearing anything and we just can't hear it? Do you think it's a form of expectation bias on there part where it's pretty much all in their head?
Funny side note. I heard the dreaded quote yesterday "My wife and kids even heard a drastic difference". Which I told him if they did hear such a drastic difference there must be something wrong with one of them. He swapped a JJ 5AR4 with a Mullard 5AR4. He disagreed and said that different manufacturer tubes sound different and that "when you mess with the power supply you change the sound of the amp".
as a builder of amps, my goal has always been to build them as best as i could from get go....
i will get depressed if by changing rectifiers my amp sounded better, that meant i did a lousy job building my amp...but this is just me.....
i have no problem with others stating their opinions, it doesn't bother me...
i go by what i have...
i will get depressed if by changing rectifiers my amp sounded better, that meant i did a lousy job building my amp...but this is just me.....
i have no problem with others stating their opinions, it doesn't bother me...
i go by what i have...
I admit that I haven't read this whole thread in its entirety. Guilty as charged, but from a theoretical standpoint I could see how voltage sag under load could affect things. I don't think it would be drastic though. None of my amps have tube rectifiers. I'm too lazy, and I use transformers I dig out of the trash so I never have the 5V winding or a center tap, etc. I'm a cheapskate, I know. As such, I don't "tube roll" with rectifiers.
That said, it doesn't surprise me that people hear a difference. If the rectifier is under heavy load, two different tubes may have slightly different characteristics. I would never agree with "night and day", however, unless one rectifier was in seriously sad condition. And people "hear" a difference in speaker wire.
I would expect it would be more prevalent in a guitar amp where it is frequently driven into distortion (more load on the power supply).
The real snake oil comes when someone says that changing out 1N4007 diodes for a UF4007 "drastically improved the tone". Noise, maybe. Tone? Um... it's a silicon diode.
Every time I hear someone say that replacing X-unassuming-component-that-isn't-in-the-audio-path with a premium one "drastically improved the sound" I have to wonder what traumatic event that component underwent, or what esoteric solder flux they're using without ventilation. Speaking of which... I heard an argument about what kind of solder flux has the best "tone" the other day...
To make a long story short, yes, it could make a difference but I doubt it's going to be noticeable in 90% of instances. Let's face it, the chair you're sitting in will have a bigger impact than a Mullard vs. GE 5AR4.
That said, it doesn't surprise me that people hear a difference. If the rectifier is under heavy load, two different tubes may have slightly different characteristics. I would never agree with "night and day", however, unless one rectifier was in seriously sad condition. And people "hear" a difference in speaker wire.
I would expect it would be more prevalent in a guitar amp where it is frequently driven into distortion (more load on the power supply).
The real snake oil comes when someone says that changing out 1N4007 diodes for a UF4007 "drastically improved the tone". Noise, maybe. Tone? Um... it's a silicon diode.
Every time I hear someone say that replacing X-unassuming-component-that-isn't-in-the-audio-path with a premium one "drastically improved the sound" I have to wonder what traumatic event that component underwent, or what esoteric solder flux they're using without ventilation. Speaking of which... I heard an argument about what kind of solder flux has the best "tone" the other day...
To make a long story short, yes, it could make a difference but I doubt it's going to be noticeable in 90% of instances. Let's face it, the chair you're sitting in will have a bigger impact than a Mullard vs. GE 5AR4.
A thread so old, come on!
Rectifier swapping can be very enjoyable! Sometimes we would gather in groups, swap, listen and comment different rectifiers, ranging from Soviet 5C3S, to American 5U4Gs, GZ34, damper diodes, Mercury vapor, RGNs or AZ, etc.. Just make sure you don't exceed their electrical limitations. Have fun!
Rectifier swapping can be very enjoyable! Sometimes we would gather in groups, swap, listen and comment different rectifiers, ranging from Soviet 5C3S, to American 5U4Gs, GZ34, damper diodes, Mercury vapor, RGNs or AZ, etc.. Just make sure you don't exceed their electrical limitations. Have fun!
I use the search function and since there is already a discussion going I would like to continue it instead of starting a new thread on the subject. There is already pretty good information already in there I just haven't had time to read it all.
You are talking about swapping different types of rectifiers and yes they are changing two conditions; one is raising or lowering the voltage so you are changing the operating points of the signal tubes. Two, different sag conditions. This is can understand sonic changes. But I have seen a lot of people change like for like tubes, e.g. a Mullard 5AR4 from a GE 5AR4 and they claim all kinds of improvements including clarity in the midrange etc..... I personally don't buy that and I think they are imagining it. Or there is something wrong with one of their valves because they do no other form of testing besides listening. As stated I have seen like for like valves perform differently with the same load, I don't attribute that to the manufacturer though, just tolerance or age.
I feel if you have two different manufactured valves and they test similarly in a double blind test these people would NEVER be able to tell the difference between brands.
Some hints:
-some rectifiers have almost the same internal resistance, hence almost no variation, less than 1% in the rectified voltage, at least in the condition we tested them lately
-we test them in a class A amplifier, so the sag theory doesn't apply.
-different brands, same tube, no measured voltage variations, but different tembral signatures.
I don't fully understand it yet as well. But it works.
-some rectifiers have almost the same internal resistance, hence almost no variation, less than 1% in the rectified voltage, at least in the condition we tested them lately
-we test them in a class A amplifier, so the sag theory doesn't apply.
-different brands, same tube, no measured voltage variations, but different tembral signatures.
I don't fully understand it yet as well. But it works.
Last edited:
A few years back I was making IMD measurements on an SE amp, swapping between a new Bendix 6106 and a well used GE 6087/5Y3WGTB ( I think all 6087's are GE, regardless of the markings ). These are both sort of premium 5Y3 variants. They have very different construction.
The loaded dc voltage output between the two rectifiers was only a couple of volts, and, IIRC, the GE had the higher voltage output. The Bendix repeatedly measured with much lower IMD. All other conditions were the same between the two tubes. At that time, I could not test for THD. I don't believe a couple of volts could make such a difference, but I did not test for that. EDIT: I was actually testing for something else, and ran across this curiosity by accident.
I did not go to my warehouse to get a new 6087 to test against the 6106. The 6087 being well used could explain the result. It tested like new, however, on my TV-7/D. Most well used U.S. rectifiers will test as new, or close to it, in my experience.
I don't hear distortion well, apparently not at all unless it is pretty gross, but if someone claimed to hear a difference between those two tubes I would be in no position to argue the point.
I have tubs of NOS and well used U.S. made rectifiers of all types so at some point I might see if there is any distortion difference between new, and used rectifiers that test out as new. My time is limited unfortunately, and that would be a really tedious undertaking.
I no longer dismiss claims that rectifiers sound different to some folks.
Win W5JAG
The loaded dc voltage output between the two rectifiers was only a couple of volts, and, IIRC, the GE had the higher voltage output. The Bendix repeatedly measured with much lower IMD. All other conditions were the same between the two tubes. At that time, I could not test for THD. I don't believe a couple of volts could make such a difference, but I did not test for that. EDIT: I was actually testing for something else, and ran across this curiosity by accident.
I did not go to my warehouse to get a new 6087 to test against the 6106. The 6087 being well used could explain the result. It tested like new, however, on my TV-7/D. Most well used U.S. rectifiers will test as new, or close to it, in my experience.
I don't hear distortion well, apparently not at all unless it is pretty gross, but if someone claimed to hear a difference between those two tubes I would be in no position to argue the point.
I have tubs of NOS and well used U.S. made rectifiers of all types so at some point I might see if there is any distortion difference between new, and used rectifiers that test out as new. My time is limited unfortunately, and that would be a really tedious undertaking.
I no longer dismiss claims that rectifiers sound different to some folks.
Win W5JAG
Last edited:
SE amps with vacuum rectifiers are likely to also have the following features:
- little or no global feedback, so gain depends on supply rail voltage
- CLC PSU filters, which can ring subsonically and may depend on rectifier resistance for damping
It is not too hard to imagine that in such a circuit the usual SE problems will vary with rectifier characteristics, and this may be audible.
- little or no global feedback, so gain depends on supply rail voltage
- CLC PSU filters, which can ring subsonically and may depend on rectifier resistance for damping
It is not too hard to imagine that in such a circuit the usual SE problems will vary with rectifier characteristics, and this may be audible.
On the other hand, I've swapped various tubes through my Dyna ST-70. Going from the original 6CA7 outputs to KT-66, 5AR4 to 5U4GB to 5R4GYB rectifiers, and 7199 to 6GH8A to 6EA8 and 6U8A of various brands of drivers, and I would not say that any of those swaps produce a variation in SQ worth any attention. Did I drive it to maximum power output and run precision test equipment on it for ??%% changes? Nope. Just listening tests like all the other anecdotes. But while those swaps were in place I measured B+ and biasing which all changed very much of course. Amp sounds great in many configurations. That's just my amp. One of millions....
Last edited:
I like to keep the discussion to rectifier swaps within the same family, so specs are the same and operating parameters stay constant. 5V4 for 5V4, etc. That eliminates the argument that one is simply changing the B+ voltage and thus operating points. People do claim to hear differences in output transformers, input transformers, chokes, plate chokes, made of different materials (M6, nickel,cobalt). Note I am not including the physical improvements in laminations such as amorphous or nanocrystalline as those are still basically the same metals.
People do claim to hear differences in rectifier,driver and output tubes having different materials (mercury vapor,thoriated tungsten). Should not be a surprise that rectifiers of the same type with the same specs (so voltages, B+ is the same) can sound different in a SE amp where the power supply is in the audio signal loop, and one component of it might be audible.
The old classic tube manufacturers kept their metal compositions for plates, anodes, cathodes secret, and their coatings for them secret as they considered them proprietary. Their best tubes had some very expensive combination of materials and coatings in them. It is hard for us to really judge what the difference is between tubes when the exact composition of their materials and coatings is unknown and cannot be compared. And even tubes in the same family have different physical construction: long plates short plates black plates, different filament constructions, coke bottle/ST/globe glass, tall tubes short tubes. Some constructions may sound better than others, there are very many possible variables to consider. So I can well understand that cheap generic tubes could be of lesser quality and sound worse than tubes made in the glory days of old.
People do claim to hear differences in rectifier,driver and output tubes having different materials (mercury vapor,thoriated tungsten). Should not be a surprise that rectifiers of the same type with the same specs (so voltages, B+ is the same) can sound different in a SE amp where the power supply is in the audio signal loop, and one component of it might be audible.
The old classic tube manufacturers kept their metal compositions for plates, anodes, cathodes secret, and their coatings for them secret as they considered them proprietary. Their best tubes had some very expensive combination of materials and coatings in them. It is hard for us to really judge what the difference is between tubes when the exact composition of their materials and coatings is unknown and cannot be compared. And even tubes in the same family have different physical construction: long plates short plates black plates, different filament constructions, coke bottle/ST/globe glass, tall tubes short tubes. Some constructions may sound better than others, there are very many possible variables to consider. So I can well understand that cheap generic tubes could be of lesser quality and sound worse than tubes made in the glory days of old.