anatech said:
Chris,
I'm not sure if you're comment was directed at JC or me. But, fast soft recovery or zero recovery diodes make a huge difference. I'm comparing to standard types that produce a lot of high frequency hash. In a simple power supply like I have they are very, very helpful. My preamp is quiet, much quieter than the commercial ones we have, so I'm happy.
I suspect that slow solid state diodes contaminate the ground. What I like about vacuum tube rectifiers is the cathode in many of them. I think that this reduces the shock of turn-on and turn-off. I have tried it both ways. However I have found that high speed, soft recovery (this can be important in a secondary way) solid state diodes are a great compromise. Not all high speed diodes are optimum. Some are actually too fast, and ring for another reason than what we were hoping to fix in the first place, so just don't get a bunch of surplus high speed diodes and EXPECT them to sound wonderful.
I have used Harris RHR series diodes for many years, but we have used good units from other companies. Fairchild make the Harris devices at this time. You don't need super-super fast, but soft recovery is important. It is a refinement to keep the diode from generating its own RF grit. Tubes don't appear to do this, and they usually sound OK at low listening levels.
I really had high hopes to easily replace GZ34 rectifier tubes with high voltage low speed diodes, but I was disappointed in the sound. It to me years to know why.
I really had high hopes to easily replace GZ34 rectifier tubes with high voltage low speed diodes, but I was disappointed in the sound. It to me years to know why.
Let me go on about faster diodes for AC power supply stages. I am asked not to recite my history, so let us say that a few decades ago, I used standard diodes in all 50-60Hz power supplies, AND that I couldn't see anything wrong with it, especially if the power supplies and their regulators were in a separate box. They were rugged, cheap, in popular bridge configurations, high voltage, easy to get, etc. Once, when a technician friend of mine offered high speed diode bridges, I laughed it off.
Then, about 1991, another friend of mine, actually measured different diodes with an RF spectrum analyzer, even though they were in an actual circuit with reasonable AC filtering. Wow! And as it usually happens, nobody believed him, except Walt Jung and me, and we had to get 'The Audio Amateur' to take his measurements and publish them in the magazine.
Later, when researching, you might say, in reverse, I realized that EVERYBODY knew about the problems in rectifier diodes, and that I had just ignored the research. Then, another colleague gave me an HP current probe. I knew about current probes, but I hadn't owned one up to that time. SO, I ran my own experiment on diodes, looking at the diode CURRENT rather than the DC voltage and 'Boy, Howdy'! It was there big time, yet it could not be easily measured with the normal oscilloscope voltage probe. A complete oversight.
Later, in reading up, I was also warned against getting JUST FAST diodes. They may work, but they are so abrupt, that they can also shock the inductors in the circuit path. That is where the 'soft recovery' comes in.
Now this is a hassle, and has obsoleted 100's of power bridges in my inventory. If you think that I look at this lightly, guess again. Also, it is next to impossible to get really big diode bridges, like we like to use in power amps, in high speed, soft recovery designs, without special order. And people why hi end costs so much.
Then, about 1991, another friend of mine, actually measured different diodes with an RF spectrum analyzer, even though they were in an actual circuit with reasonable AC filtering. Wow! And as it usually happens, nobody believed him, except Walt Jung and me, and we had to get 'The Audio Amateur' to take his measurements and publish them in the magazine.
Later, when researching, you might say, in reverse, I realized that EVERYBODY knew about the problems in rectifier diodes, and that I had just ignored the research. Then, another colleague gave me an HP current probe. I knew about current probes, but I hadn't owned one up to that time. SO, I ran my own experiment on diodes, looking at the diode CURRENT rather than the DC voltage and 'Boy, Howdy'! It was there big time, yet it could not be easily measured with the normal oscilloscope voltage probe. A complete oversight.
Later, in reading up, I was also warned against getting JUST FAST diodes. They may work, but they are so abrupt, that they can also shock the inductors in the circuit path. That is where the 'soft recovery' comes in.
Now this is a hassle, and has obsoleted 100's of power bridges in my inventory. If you think that I look at this lightly, guess again. Also, it is next to impossible to get really big diode bridges, like we like to use in power amps, in high speed, soft recovery designs, without special order. And people why hi end costs so much.
Speaking of diodes and history, 40 years ago I liked to listen to American radio. We had a radio receiver, with 2-gang variable capacitor, one tube mixer-oscillator, one tube IF amplifier, one tube double diode and AF pre-amplifier, one tube AF audio amplifier, and one tube rectifier. 5 tubes total. Similar to All American 5.
Antenna was hung above the roof, about 30 feet long.
You can understand what that meant; it was a Cold War time, citizen of Soviet Union had no right to listen to foreign propaganda, we had rights to listen to Soviet propaganda only. Voice of America broadcasting on Russian language was jammed by powerful military transmitters, but anyway it was possible sometimes during nights to hear voices and even music!
I had a dream, to build a receiver with sharp selectivity, but could not afford then...
Many years later, here in USA, I bought Hallicrafters shortwave receiver. It had 3 gang variable capacitor, with RF stage, and 2 AF stages. I did not need to listen to foreign stations, it was an old dream... But after fixing and tuning the set I found all short waves polluted by 120 Hz hum!
I sold it, and bought BC-348, used on long distance bombers. It had 4 gang variable capacitor. I restored it, tuned perfectly, but again the hum was loud.
I bought a super-duper shortwave receiver, Collins R392/URR. Fixed it. Tuned. But still was not satisfied.
I sold it, and built an external pre-selector with plug-in magnetic frame antenna made of thick copper tube, for my BC-348 I still keep. Reception was good, but nasty hum was still there. I lived in apartments then.
Once, listening to the radio, I went to bathroom and unplugged a nightlight... Wow! Cleaner louder reception, less loud hum... Thyristor controller, no filtering at all!
However, I could not enter every apartment in our building to unplug all thyristor-controlled dimmers, nightlights, energy-saving lamps, SMPSes. Power filtering and thorough grounding did not help: it was a receiver by definition!
I never had such problems with prototypes of audio equipment designed by me. I know what to do without super-duper diodes, and it is under my control. What is not under control, switched mode power supplies in another gear that may give much more of harm than that diode bridges I use inside of my equipment.
Antenna was hung above the roof, about 30 feet long.
You can understand what that meant; it was a Cold War time, citizen of Soviet Union had no right to listen to foreign propaganda, we had rights to listen to Soviet propaganda only. Voice of America broadcasting on Russian language was jammed by powerful military transmitters, but anyway it was possible sometimes during nights to hear voices and even music!
I had a dream, to build a receiver with sharp selectivity, but could not afford then...
Many years later, here in USA, I bought Hallicrafters shortwave receiver. It had 3 gang variable capacitor, with RF stage, and 2 AF stages. I did not need to listen to foreign stations, it was an old dream... But after fixing and tuning the set I found all short waves polluted by 120 Hz hum!
I sold it, and bought BC-348, used on long distance bombers. It had 4 gang variable capacitor. I restored it, tuned perfectly, but again the hum was loud.
I bought a super-duper shortwave receiver, Collins R392/URR. Fixed it. Tuned. But still was not satisfied.
I sold it, and built an external pre-selector with plug-in magnetic frame antenna made of thick copper tube, for my BC-348 I still keep. Reception was good, but nasty hum was still there. I lived in apartments then.
Once, listening to the radio, I went to bathroom and unplugged a nightlight... Wow! Cleaner louder reception, less loud hum... Thyristor controller, no filtering at all!
However, I could not enter every apartment in our building to unplug all thyristor-controlled dimmers, nightlights, energy-saving lamps, SMPSes. Power filtering and thorough grounding did not help: it was a receiver by definition!
I never had such problems with prototypes of audio equipment designed by me. I know what to do without super-duper diodes, and it is under my control. What is not under control, switched mode power supplies in another gear that may give much more of harm than that diode bridges I use inside of my equipment.
Hi John,
Here we go again. You read far too much into what I post, remember that I try to be very general, given that your designs are a state secret. I know only what you have posted on your designs and your expriences. I agree that our observations are different many times.
So ....
You don't. 'nough said.
I have stated that what I know about rectifier diodes comes from what I was taught and confirmed through testing in many different brands of equipment and my own designs. You do state your findings as if they apply to everything, but instantly take the position that your results are valid in your equipment if you feel you are challenged. Your argument is much like the defense used by people caught in error in listening tests.
Tiresome.
One thing that has bothered me about how you go about things is that you select individual components to achieve the results you want. You can get to the same place in different ways once you understand why these parts behave differently. However, the ad copy isn't nearly as good. Tube rectifiers have been popular for a while lately without good reason.
Are you saying that the choice of rectifier has something to do with the application? Holy bleep! I think most people will agree on this except for the last point you are making.
BTW, what point are you making? Let's see, surplus diodes can't be trusted? Their characteristics change as the box they are sitting in are written down (by accounting) and stock sold off? I don't think so! However, I buy my rectifying devices new. They even have numbers on them and everything John. My take on your statement is that you can't simply use a fast recovery diode and stick it in to any circuit and expect it will cure all your ills. In other words, you didn't say anything different - but it sounded profound.
You know John, if you stop going after people, your audience will be more accepting. It would also help if you were clear about special circumstances that you have findings for on components that differ from what most people see.
I responded to you like this for two reasons. Firstly, you went after me in a nasty way. You tried to put me down and treated me as if I am an idiot. Not pretty coming from someone who wants to be seen as superior, possibly better educated. This wasn't the first time and I'm tired of the way you treat other people some times.
Secondly. You have a habit of making claims with no supporting evidence. When challenged, you retreat to a situation that few but yourself can observe. A great deal of talk, some of which sounds good, but without actually saying anything of substance.
I'm too damned sore to play games right now John. In pain, tired and crabby to care much about your preamp or it's dark secrets. Your preamp was taken to the ground because a tweak power cord can not be created again. God, that's sad John. Think about what that says about your design a moment. Heaven forbid the electric company should change pole transformer brands, because that should be audible as well. Heaven knows how different wall outlets may sound on top of that. Do you have an approved breaker type, or must the customer use the old fuse box type?
What you are suggesting is that a micro change caused by a power cord is audible, but these other items do not have an effect on the sound. That is really difficult to fathom if it's not anything but a sales pitch, or ad copy.
-Chris
Here we go again. You read far too much into what I post, remember that I try to be very general, given that your designs are a state secret. I know only what you have posted on your designs and your expriences. I agree that our observations are different many times.
So ....
Stop insulting me. As I have stated many times, I know little of what you do under the hood. I'm interested, but not so much that I intend to go producing JC clones. What I conclude about things I state are observed behavior. If I am not sure, I ask a question. When I ask, it is not to show someone up, it's because I'm not sure about the answer. Occasionally I am mistaken in some things I have accepted as a fact. Ever notice that I state I'm wrong and apologize?
You don't. 'nough said.
I haven't and probably will not.
I have stated that what I know about rectifier diodes comes from what I was taught and confirmed through testing in many different brands of equipment and my own designs. You do state your findings as if they apply to everything, but instantly take the position that your results are valid in your equipment if you feel you are challenged. Your argument is much like the defense used by people caught in error in listening tests.
Tiresome.
Gee, that's fair and not up to debate. It does not mean you are right though. Ever consider that others do similar things, myself included. Wow.
Ditto.
... And some things that are easy to measure. What's your point? We all run up against these things.
You aren't alone there. I happen to disagree, having understood what differences there are. A vacuum tube has high series resistance. I have advocated adding resistance many times in low current supplies because that is what reduces the peak charging current in your filter caps. I will use rectifier tubes when they make life easier.
One thing that has bothered me about how you go about things is that you select individual components to achieve the results you want. You can get to the same place in different ways once you understand why these parts behave differently. However, the ad copy isn't nearly as good. Tube rectifiers have been popular for a while lately without good reason.
Why would that be John? There is nothing that these rectifiers do that other types will not to some extent. Besides, what is your definition of "slow solid state diodes"? Do you mean standard recovery like 1N400x series parts, or soft recovery? This should be defined so we are all on the same page.
Again you have not defined anything except that you like the cathode type as opposed to the filament types. What is getting shocked at turn on, the plate supplies? "Turn off shock" is identical no matter what diode type you are using. When the current is interrupted in the transformer primary, the current also stops in all secondaries, especially if that secondary winding is loaded. Whether the cathode is hot or not has exactly zero to do with anything.
Are you saying that the choice of rectifier has something to do with the application? Holy bleep! I think most people will agree on this except for the last point you are making.
BTW, what point are you making? Let's see, surplus diodes can't be trusted? Their characteristics change as the box they are sitting in are written down (by accounting) and stock sold off? I don't think so! However, I buy my rectifying devices new. They even have numbers on them and everything John. My take on your statement is that you can't simply use a fast recovery diode and stick it in to any circuit and expect it will cure all your ills. In other words, you didn't say anything different - but it sounded profound.
As opposed to the crappy ones from Harris and others?
Ahhhh, yeah. What's your point? How about comparing NOS tubes to new manufacture, Russian, Chinese or whatever? Does this mean you use tube rectifiers in your best equipment?
Anything you can put your finger on?
Why? Another secret? I have my own understanding as I'm sure most other fellow designers here do. Care to share?
Harder to see with a 'scope, but possible. I know I've looked at this issue in the early 80s as well. I'm more tha nsure many other people have as well. I used early HP 54600 series DSOs with the FFT cartridge to confirm what I thought was going on. They were just out then. My god!! John, we are both brilliant! <sorry, couldn't resist>
About the same time, I had to buy mine. Lost it when I sold my shop, but I did buy some more. Low and high frequency models. You should see what Agilent has available. Current probes that simply plug into your USB port on your 'puter. They also have RF power probes that plug into your USB port. Cool stuff!
Yup, that's what you do with Current probes.
No might about it. They even induce oscillations in straight wire if the conditions are right. They were probably designed for switching power supplies or something like that. You think?
Yes! That's what those "RC" or just "C" components are across those diodes. You know, from before 25 years ago. But, if we are determined to use a more expensive component that does the same thing, we can use a ...
Yes, the special soft recovery diode that we don't actually need.
Oops! How stupid of me! That's why you use them, they cost more and they are different. Now, that is why high end may cost more.
You know John, if you stop going after people, your audience will be more accepting. It would also help if you were clear about special circumstances that you have findings for on components that differ from what most people see.
I responded to you like this for two reasons. Firstly, you went after me in a nasty way. You tried to put me down and treated me as if I am an idiot. Not pretty coming from someone who wants to be seen as superior, possibly better educated. This wasn't the first time and I'm tired of the way you treat other people some times.
Secondly. You have a habit of making claims with no supporting evidence. When challenged, you retreat to a situation that few but yourself can observe. A great deal of talk, some of which sounds good, but without actually saying anything of substance.
I'm too damned sore to play games right now John. In pain, tired and crabby to care much about your preamp or it's dark secrets. Your preamp was taken to the ground because a tweak power cord can not be created again. God, that's sad John. Think about what that says about your design a moment. Heaven forbid the electric company should change pole transformer brands, because that should be audible as well. Heaven knows how different wall outlets may sound on top of that. Do you have an approved breaker type, or must the customer use the old fuse box type?
What you are suggesting is that a micro change caused by a power cord is audible, but these other items do not have an effect on the sound. That is really difficult to fathom if it's not anything but a sales pitch, or ad copy.
-Chris
Hi Johnloudb,
No, I was talking to John Curl.
You have hit the nail on the head though. In a simple circuit, components can and do make differences in performance. If the power supply is designed with some care you will find that effect of an individual component is lessened to some degree.
Analogy time. A single transistor amplifier with signal taken of the collector. Any changes in parts will change the sound, but for reasons that can be explained. Use a three transistor circuit, CFP with a buffer output. Individual components are not as effective in changing the sound. Lower distortion is a side benefit. You can get more complicated and improve the performance at the same time. You can also really mess things up if you don't design your circuitry with some thought based on factual knowledge.
-Chris
No, I was talking to John Curl.
You have hit the nail on the head though. In a simple circuit, components can and do make differences in performance. If the power supply is designed with some care you will find that effect of an individual component is lessened to some degree.
Analogy time. A single transistor amplifier with signal taken of the collector. Any changes in parts will change the sound, but for reasons that can be explained. Use a three transistor circuit, CFP with a buffer output. Individual components are not as effective in changing the sound. Lower distortion is a side benefit. You can get more complicated and improve the performance at the same time. You can also really mess things up if you don't design your circuitry with some thought based on factual knowledge.
-Chris
SY said:
Who likes them? Nobody. But they act as an obstacle in creating of some obvious problems. That's likable!
You saw one my "secret" in my barn, when I demonstrated you a blown fuse between first caps and the last one. Do you remember? Also, you probably saw a tall farm with high heatsink and 3 power toroids. "Poor Wavebourn can't afford one huge toroid, one huge bridge, and couple of huge low ESR reservoirs", somebody would say.
- Status
- Not open for further replies.