Jump to content

Rectifier rolling


michaelwjones

Recommended Posts

I was talking about when designing a new amplifier I am only going to use SS rectification from now on. I do not know what you problem is but carry on. It appears you do much the same as you do here on other threads as well. One way to avoid this is to just ignore all my postings. Make things much more pleasant on this forum. I just have a hard time understanding what your point is after saying above you have both SS and tube rectified amplifiers. Are you trying to say you prefer one over the other or is it just wanting someone to attack is what you are after. 

Link to comment
Share on other sites

19 hours ago, captainbeefheart said:

 

I just wanted to clarify that screen grid ratings are in watts and not current. The old Genelex datasheet says absolute max 8 watts where the design maximum is 6 watts. Which for 420v would be around 19mA for the 8 watts and 14mA for 6 watts.

 

Thanks for pointing this out. Then I should be fine without reducing the voltage, current in my g2 at 8ma. With 420V's and 8ma current if my math is correct I am running 3.36w and should be alright on Electro Harmonic KT-88's rated at 6.6w on g2. I cannot see the screen grids inside of these tubes to see if there is any glowing.  Typical ma on data sheet says 7ma which was what I was looking at. I have been experimenting with higher voltages on plates but being careful of g2 which is the main problem of higher voltages. Modjeski and Anderson both like higher voltages on plates but Modjeski took it to the extreme using 720V's on EL84's, those small tubes. Most plates can take the higher voltages . It is the screen one has to be careful of burning up. Tubes are not cheap anymore. A pair of Electro-Harmonix KT-88's is over $150 now. 

 

Modjeski did not say how he reduced the V on the screen grid but adding a regulated screen could be possible the way he did it. I may at some time use 700+v's but first I would like to see how using just a resistor looks like on scope and distortion software before abandoning using just a resistor for simplicity sake. I like to keep things simple a much as possible. Dealing with designing a high voltage regulated g2 supply is not something I relish doing. 

 

We are getting away from the original OP topic of rectifier rolling so perhaps we can start another thread and continue this discussion. I would like to see what you come up with for reducing the voltage on g2 using higher voltages on the plates. With higher voltages one has more voltage swing for the signal to operate in. Modjeski's point. He got 40 watts output from a PP EL-84 amplifier where 17.5w is more common in a PP EL-84 amp.  

Link to comment
Share on other sites

11 hours ago, henry4841 said:

Modjeski did not say how he reduced the V on the screen grid but adding a regulated screen could be possible the way he did it. I may at some time use 700+v's but first I would like to see how using just a resistor looks like on scope and distortion software before abandoning using just a resistor for simplicity sake. I like to keep things simple a much as possible. Dealing with designing a high voltage regulated g2 supply is not something I relish doing. 

 

Roger has the most simplest power supply setup. It's a voltage doubler circuit, 700v on the plates and the 350v at the half voltage point for the screens and front end.

 

As for the screen dissipation it's not the 8mA idle current that's a problem, it's when there is signal applied and as the signal gets larger on the plate it swings down close to ground potential. If you have a plate potential at 50v during peak negative cycle the electrons from the cathode will be much more attracted to the screen at 420v vs the plate at 50v. Screen current drastically increases during this portion of the signal cycle.

Link to comment
Share on other sites

15 hours ago, captainbeefheart said:

 

Roger has the most simplest power supply setup. It's a voltage doubler circuit, 700v on the plates and the 350v at the half voltage point for the screens and front end.

Voltage doubler, simple. 

 

I still would not advise using one of those SS rectifiers gadgets in place of rectifier tube without biasing your output tubes for the extra voltage which can be quite large. If your amplifier is conservatively biased it should not be a problem but if not you could damage some tubes or at least shorten their life expectancy. On my KT-88 amplifier I have been experimenting with I raised the cathode resistor value to lower the plate dissipation to a more safe value when I installed the $5 SS rectifier gadget which is as Nick said no more than two 1n4007 diodes. 

Link to comment
Share on other sites

On 11/2/2022 at 5:49 PM, captainbeefheart said:

 

Roger has the most simplest power supply setup. It's a voltage doubler circuit, 700v on the plates and the 350v at the half voltage point for the screens and front end.

 

I do not think I have ever worked on anything with a voltage doubler, at least I did not recognize it as such. After some reading, very little time spent, there is a question. Will there be enough current to operate all the tubes using a voltage doubler for a high voltage source. From what I have read a voltage doubler is limited on the current potential. 

Link to comment
Share on other sites

4 hours ago, henry4841 said:

I do not think I have ever worked on anything with a voltage doubler, at least I did not recognize it as such. After some reading, very little time spent, there is a question. Will there be enough current to operate all the tubes using a voltage doubler for a high voltage source. From what I have read a voltage doubler is limited on the current potential. 

 

You may have been reading about half wave Villard voltage multipliers which can be double, triple, quadruple, etc.... These voltage doubler circuits have poor regulation and not used in power amplifiers because of it. They are good for low current applications where the load is small and constant. I have seen them in audio only as bias circuits for electrostatic speakers.

 

What you want to look up is the Delon circuit which is a full wave voltage doubler that cannot be multiplied like the Villard circuit. This circuit is, essentially, two stacked peak detector circuits, each charging their respective capacitors during opposite halves of the incoming AC voltage signal. There are lots of tube amplifiers using the Delon full wave doubler circuit.

 

Here is the Delon full wave doubler circuit, I'm sure you have seen it before.

 

spacer.png

 

 

Link to comment
Share on other sites

3 hours ago, captainbeefheart said:

 

 

Here is the Delon full wave doubler circuit, I'm sure you have seen it before.

 

spacer.png

 

 

Yes I have but I had my doubts of having enough current to run tubes. Especially 4 of them. Modjeski claimed to run PP EL-84's at half dissipation creating 40 watts running 720v's on the plates. It is fascinating to me when most only expect a lot less from a PP EL-84 amplifier. Playing with over 700v's is not something I really feel confident doing. My field was always repair work and not designing. All SS until a few years ago.  

Link to comment
Share on other sites

49 minutes ago, henry4841 said:

Yes I have but I had my doubts of having enough current to run tubes. Especially 4 of them. Modjeski claimed to run PP EL-84's at half dissipation creating 40 watts running 720v's on the plates. It is fascinating to me when most only expect a lot less from a PP EL-84 amplifier. Playing with over 700v's is not something I really feel confident doing. My field was always repair work and not designing. All SS until a few years ago.  

 

Voltage multiplier circuits like the Villard are for low current applications but the Delon full wave doubler is fine.

 

There is so much controversy over that amp when it's really a very simple design. The reason he gets so much power and has such low bias current is because although it's technically a Class AB amp, it's right at the edge of being pure Class B. It only runs in Class A mode for a little over 1 watt then switches to Class B. Look at any power tube datasheets for Class B operation and you'll see much higher plate voltages with much lower idle bias currents.

 

Look at it this way, it's a Class B amplifier with it's bias current slid up just enough to technically make it a Class AB amplifier.

Link to comment
Share on other sites

Data

 

Test case, one low-cost Chinese 5AR4 and a 2x1N5408 rectifier plugin.  Amp is a 299C scratch clone I made last year.  I laid the amp out on PC boards an assembled the entire unit on a CNC machined plate.

 

image.thumb.jpeg.de002e7e910fc53d026877f35c9520ed.jpeg

 

image.thumb.jpeg.dcc0e7a357559d5cbb36c0082bc39b4f.jpeg

 

First test was to measure distortion at equivalent anode voltage, in this case 425 VDC.

 

With a Variac, the PS transformer primary VAC is adjusted to provide anode voltage at 425VDC for both rectifiers.   %THD to full output power at 20Hz is then measured and compared.  Load is a 50W non-inductive power resistor across the 8 Ohm output tube transformer secondary tap.  5 measurements for each rectifier were taken and compared.  If the result was consistent between each, it's reported here.

 

At 117.1 VAC at the PS transformer primary the 5AR4 produces 425VDC with 1243mA at idle.

At 111.6 VAC at the PS transformer primary the 2x1N5408 plug-in produces 425VDC with 1245mA at idle.

 

Blue is 5AR4

Green is 2x1N5408 rectifier

 

A measurable difference below about 2W RMS, above that it's a wash.  There are other tests that are worth performing but this was the quickest.  

 

image.thumb.jpeg.d1017d5635ccd12b8b0e508dcc157e25.jpeg

 

The next test was to fix the PS transformer primary voltage to 120 VAC for rectifier condition, turn the amp on and measure the PS transformer instantaneous (i.e., "turn on") current thru the AC power entry module.  The peak VDC at the anode is also measured.  

 

The 5AR4@120VAC peak current 1448mA AC and peak voltage 437.8 VDC

 

image.thumb.jpeg.868ae48cb95cd7d1829435919da998ce.jpeg

 

image.thumb.jpeg.7ac6a78fac305ff55d18511cd32080f3.jpeg

 

The 2x1N5408@120VAC peak current 2586mA AC and peak voltage 488.6VDC

 

image.thumb.jpeg.71a309433526cf4db1fd8a0a28123944.jpeg

 

image.thumb.jpeg.db9a15d12d2d2e80610b1ae7165ac388.jpeg

 

The instantaneous current draw is 1.2A or about 2X greater using the solid-state rectifier.  Peak anode voltage at turn-on is about 51VDC larger using the solid-state rectifier as well.

 

Note that wall power can approach 125VAC which would create larger differences in magnitude than what's being measured here. 

 

Next will be filament current measurements at turn on and the effect of various in-rush current limiters on the solid-state rectifier turn on characteristics.

 

  • Like 2
  • Thanks 4
Link to comment
Share on other sites

On 11/5/2022 at 9:10 PM, henry4841 said:

CL-90 I hope is one of them since that is what I use.

I haven't gone thru my inventory but maybe.  I'm noodling a good way to record the current v. time thru the power supply primary, tube heaters, etc.  I have plenty of meters and can capture max values but would prefer to record the response differences in a plot.  The sampling rate has to be high given trying to measure a transient and the resolution of the capture has to be a few mA both AC and DC.

  • Like 1
Link to comment
Share on other sites

1 hour ago, John Warren said:

I haven't gone thru my inventory but maybe.  I'm noodling a good way to record the current v. time thru the power supply primary, tube heaters, etc.  I have plenty of meters and can capture max values but would prefer to record the response differences in a plot.  The sampling rate has to be high given trying to measure a transient and the resolution of the capture has to be a few mA both AC and DC.

You are to be commended for doing real world test and not just simulations on some software which is what one mostly sees on forums these days. 

  • Like 1
Link to comment
Share on other sites

There are different types of rectifiers and independent tests and reviews are needed to make a decision! Otherwise, there may be security problems! Working in a plastic surgery clinic here cosmetic practice is not easy for me and takes a lot of time!

 
 
 
 
 
 
 
 
Edited by DerrickOlley
Link to comment
Share on other sites

I ran across this video from a guy I respect, along with many others, on youtube who in the video declares that SS rectification is the way to go. There are some other videos with more details on what he found to be the differences in SS vs tube rectification. One thing I picked up from this video is he measured a significant voltage drop on a tube rectifier when the amplifier produces more power which in turns lowers the voltage to the power tubes reducing the power. Not something one particularly desires in a hifi amplifier.  Personally I favor SS today, might change opinion later, because of the extra voltage one has to work with. This is if one considers higher voltages a plus. 

 

https://www.youtube.com/watch?v=ietxiWXVNNo

 

  • Like 1
Link to comment
Share on other sites

6 minutes ago, 82 Cornwalls said:

This was supposed to be a thread about tube rectifiers, can you understand that?

Start you own thread about the (debatable) superiority of solid state rectification. 

 

Actually I started this thread on rectifier rolling and as I acknowledged in a subsequent post plug-in SS rectifiers had been considered. After reading the posts, it affirmed my lay conclusion that due to the nearly instantaneous availability of higher voltages. a SS rectifier is best in a circuit designed for SS rectification.

  • Like 1
Link to comment
Share on other sites

On 11/5/2022 at 7:54 PM, John Warren said:

At 117.1 VAC at the PS transformer primary the 5AR4 produces 425VDC with 1243mA at idle.

At 111.6 VAC at the PS transformer primary the 2x1N5408 plug-in produces 425VDC with 1245mA at idle.

 

I applaud you for doing these real world tests. Many people are wondering about these types of situations where they have an amp and want to play with it by changing rectifiers. I have a few thoughts though I would like to add.

 

First is these people doing these changes aren't going to step the voltage down on the primary. I see what you are trying to do which is create the same operating point for the circuit in order to compare just the rectifier and not the changes made to the operating conditions around the tube. E.g. higher plate voltage and current if bias isn't adjusted.

 

This leads to a completely new problem created by reducing the primary voltage so low. The tube heaters are now running low, for 117v to 6.3v secondary at 111v the heater will be now ~5.8v. This is most likely lowering emissions and subsequent transconductance of the devices.

 

*I know the Scott uses the negative bias rail for 4 of the tubes series filament string, so they should be unaffected but the entire power amp uses AC filaments.

 

 

 

  • Like 2
Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...