Jump to content

Rectifier rolling


michaelwjones

Recommended Posts

5 hours ago, henry4841 said:

I say better being SS does the job as good if not better at a much cheaper cost.

 

So it boils down to efficiency like I said. 

 

5 hours ago, henry4841 said:

 The problems associated with SS over tubes for rectification can be easily worked around with working solutions.

 

The opposite is also true; The problems associated with tube over SS for rectification can be easily worked around with working solutions.

 

5 hours ago, henry4841 said:

Face it guys tube rectification is old school. This is not to criticize the diehard tube heads who still favor tube rectification. I respect and admire them for stating there opinions and have to admit an extra tube sure looks nice in an amplifier. I would say half of my dozen or so tube amplifiers use tube rectification but I would not design a new amplifier using a tube for rectification. Why bother. Sure a tube might bring some flavor into the mix but it was not designed to do so.

 

Have you tried a choke input filter with tube rectification in a Class AB amp? The tube does not add any "flavor" and the supply is very well regulated. It also meets your "simple" criteria.

 

A tube rectifier with active regulation will outperform any "simple" linear power supply with SS diodes. No "flavor" here either.

 

The only time you will get "flavor" from a tube rectifier is when you have changing current demands like with Class B operation. Class A circuits the argument is moot.

 

 

 

 

 

 

 

Link to comment
Share on other sites

3 hours ago, michaelwjones said:

The article I took the "chart" from also tested a SS rectifier. They noted its voltage sag/recovery was .01 sec while the 5Y3 was almost .5 sec. Their point was in a guitar amp that difference can be audible. Such a replacement also could place a strain on the transformers and tubes. To my non-technical/engineering mind, that makes sense just looking at the numbers. One is bringing your hands together, the other snapping your fingers.

 

Guitar amps are a completely different beast and many players prefer the added compression "sag" from a tube rectifier with a Class AB amp. A vintage amplifier that wasn't designed for SS rectification has no place using SS rectifiers unless an engineer or someone smart enough can assess if the amp will handle it without issues or make necessary changes to allow the use of SS rectifiers.

 

With hifi amps it's moot because if you want the best performance your going to go with a regulated supply anyway so at that point the rectifier impedance is of no concern. It's the cheap consumer grade amplifiers with tube rectifiers that have sag/compression issues from the tube rectifier. If you look at many of the lab grade or high quality amplifiers from RCA (50 watt high fidelity amplifier circuit) or HH Scott Lab grade mono amps etc...   they all have a regulated B+.  Even if you don't want the complexity of active regulation go with a choke input filter instead of a cap input filter. Great regulation and simplicity from a tube rectifier.

Link to comment
Share on other sites

2 hours ago, tube fanatic said:

Aside from having an extra tube filament to enjoy looking at I don’t see any need to add the expense of another tube which will eventually need replacement, and another socket.  Modern SS diodes can typically handle a 100-200 amp surge so that is a non-issue, and I find that the power xfmr impedance is sufficiently high to serve as a peak current limiter.  I have never had a power xfmr fail due to peak current issues.  The component which takes the hit is the fuse and often needs to be a slow blow type.  Another advantage of SS rectifiers is that they allow the use of very large value input filter caps so that ripple can be brought down to negligible levels very inexpensively.  Two caps and a resistor can create ripple values which can only be duplicated by using a more costly choke with a tube rectifier (the value of the input filter cap is generally very low to protect the tube).  When designing new equipment it comes down to a matter of taste.  

 

All I keep hearing is; "expense", "cost", "efficiency", and "simplicity" from everyone. Yes I get that and agree. There are plenty of times I go SS for rectification but it's always due to cost and space. If I'm needing to keep the cost down ultimately SS is a great advantage.

 

 

My only point I am trying to get across to people is that using a tube rectifier does not ultimately equate to worse performance, you just need to design around them.

 

Those datasheets that give state a maximum cap input filter value is for a specific scenario which they will state; load current,  ac plate voltage, winding resistance, etc..... But how did they calculate this value of capacitance? The answer is from the more important spec in the datasheet given as maximum repetitive peak forward current. For a quick design reference sure it's probably close but in some instances you may have to even use less capacitance than what they state, or on the flip side you can use more capacitance. I've just seen too many "gurus" tell me as fact that you can only use "xx" amount of capacitance with "xx" rectifier and it's just not true. Granted, yes you can use more capacitance with SS rectifiers but you reach a point of there is no benefit of going any higher, typically 47uF-100uF is all that is required.

 

For Class A amplifiers I agree with your last sentence, it's a matter of taste but also a matter of cost and space. Single ended your most likely going to need a choke anyway with a power amp unless it's a very low output power amp.

 

For Class AB amps I don't see it as a matter of taste so much. It's efficiency and cost. Ultimately the application will dictate everything. I mean if you are trying to get the most bang for your buck with as simple an amplifier as possible yes SS is your best choice. It does not mean that you cannot get just as good a result with a tube rectifier, it's just going to take up more space and cost more.

  • Like 1
Link to comment
Share on other sites

The RCA 50 watt amplifier uses a 5R4 rectifier which has lots of sag if used with a C input filter. Looking at the schematic you'll see it's a choke input filter. The screens are regulated. Local and global feedback loops.

 

I made a pair of mono amps with this circuit except I swapped the 7199 for 6BL8 and I used 6L6GC for the finals. This amp has amazing performance. Although it uses the lossy 5R4 rectifier it's of no issue due to the choke input filter, the regulation is stiff under full power. I was getting 50 watts output with .07% THD. That's pretty darn impressive if you ask me. This design blows many of the more popular cloned circuits out of the water. I am always curious to why more people do not clone this circuit more often since it's so impressive. I know the golden ears don't care about "specs" like low distortion but the amplifier sounds better than McIntosh amps, at least I think so. I compare against McIntosh because they too have very clean and powerful amplifier designs.

 

spacer.png

 

 

Link to comment
Share on other sites

8 hours ago, 82 Cornwalls said:

 

So I can assume you only have class D amplifiers at your place? 🙄

I do not understand how favoring SS rectification in a tube amplifier has anything to do with class D  and yes I do have many class D amplifiers but seldom listen to them. For me it is mostly class A either tube or SS amplifiers of which I have many along with some class A/B throwed into the mix. I do not understand how your post has anything to do with favoring SS rectification in a tube design. Sorry no offense meant but your statement makes no sense to me. 

  • Like 1
Link to comment
Share on other sites

33 minutes ago, 82 Cornwalls said:

 

You said tube rectifiers are old fashion, TUBE AMPS are old fashion no matter how they are rectified.

There are many current production amplifiers that have tube rectifiers.

 

Your statement was like saying wooden wheels wheels on a horse drawn buggy are old fashion.

It's not like solid state rectification is something new.

 

You like tube amps, many people like tube rectifiers in their tube amps.

Both are old fashion, so what?

 

WRT cost, a new mini class d amp cost less than a pair of tubes.

Why spend money on tubes? Old fashion indeed.

A little secret. Tubes still rule when it comes to sound. How you rectify AC voltage tmaking DC really should not be a serious discussion but that is what happens on a social forum. 

 

If I were like Seti, having many rectifier tubes in stock, I would be using rectifier tubes as well. Moot point actually, whether tube or SS rectification. Both do the job and that is all that counts. I only SS rectification in my tube builds now. Why not. Most say quicker and better. At least most of the new designers of tube circuits. 

  • Like 2
Link to comment
Share on other sites

46 minutes ago, 82 Cornwalls said:

 

You said tube rectifiers are old fashion, TUBE AMPS are old fashion no matter how they are rectified.

There are many current production amplifiers that have tube rectifiers.

 

Your statement was like saying wooden wheels wheels on a horse drawn buggy are old fashion.

It's not like solid state rectification is something new.

 

You like tube amps, many people like tube rectifiers in their tube amps.

Both are old fashion, so what?

 

WRT cost, a new mini class d amp cost less than a pair of tubes.

Why spend money on tubes? Old fashion indeed.

I am curious on what your point is. Do you design your tube builds with tube rectification and are against SS?

  • Like 1
Link to comment
Share on other sites

This is my last build which I chose to have the option of using either SS or tube for a rectifier. One thing I like with SS is the increase in voltage. There is a considerable voltage drop across a rectifier tube. You can lower the extra voltage if desired with a resistor or you can use the extra voltage to increase the current through the output tube for more plate dissipation or lower the current, if for a cathode biased circuit, by increasing the resistance. I am running the EL-34's on this build around 20 watts for tubes rated at 25 watts. In other words 80% which most consider the best way. With a few value changes I can move a few wires and use a rectifier tube if desired. I may or may not do some bias adjusting for the audio tubes depending on what the numbers are. 

 

The SE EL-34 amplifier below is currently wired for SS rectification. 

 

 

P1040108.JPG

Link to comment
Share on other sites

37 minutes ago, 82 Cornwalls said:

Sorry, I didn't realize this thread is all about you and your opinions (that was not indicated in the tittle).

I will move on now.

 

Friend this is not my thread. I just posted what little I know about rectification when it comes to tube circuits. You seem to be the one who has nothing to contribute to the thread other than negative comments. This is what you said that you have since deleted. It is obvious that you are not a tube guy but just wanted to express your opinions of tubes in general which is fine. Let's just get over it and move one. 

 

  2 hours ago, 82 Cornwalls said:

 

You said tube rectifiers are old fashion, TUBE AMPS are old fashion no matter how they are rectified.

There are many current production amplifiers that have tube rectifiers.

 

Your statement was like saying wooden wheels wheels on a horse drawn buggy are old fashion.

It's not like solid state rectification is something new.

 

You like tube amps, many people like tube rectifiers in their tube amps.

Both are old fashion, so what?

 

WRT cost, a new mini class d amp cost less than a pair of tubes.

Why spend money on tubes? Old fashion indeed.

Link to comment
Share on other sites

Let's stay on topic since this is a great conversation and I'm sure there are lots of lurkers interested in this topic and following the discussion.

 

So moving forward

 

Some say ditch the tube rectifier and use SS so you can use larger filter caps and not need a choke. With SE amps and no common mode rejection at the output stage I think if you are going to want good performance a choke is necessary and you won't get as good ripple reduction with just larger caps in RC networks. Well unless it's a flea watt amp then there might be low enough current to not need a choke. For the uninitiated it's the capacitor directly after the rectifier that needs to be lowered with a tube rectifier, not so much the capacitance further down the power supply networks.

 

Let's assume a single ended Class A amplifier 2 channel amplifier. Total DC load current of 150mA.

 

Typically minimum I feel for each filter stage is at least -20db but the lower the better obviously. With -20db we need each filter stage to have 1/10 impedance ratio; i.e. the resistor needs to be at least 10x the impedance of the capacitor at 120Hz.

 

For a full wave rectifier we need to filter 120Hz. There isn't much of a point going higher than 470uF because it's impedance is around 2 ohms. In the divider network of the filter going any lower in impedance has more and more diminished returns.

 

With 150mA you are most likely going to be maxed out at around 150 ohms for the resistor value. That gives us around 22v drop across it and over 3 watts dissipation. But since this is just hypothetical let's go crazy and make the series resistor 330 ohms. With 150mA that's a 50v drop and 7.5 watts of dissipation!!! You'll most likely want a minimum of 20 watt resistor rating. Big and hot!!

 

With the 470uF cap the 330 ohm resistor we get roughly -44db ripple attenuation.

 

Now with a 10H choke - which is 7.5k impedance to 120Hz. And we will use a 100uF cap. This filter roughly gets us -55db. In many builds I'm below the repetitive peak forward current with 100uF after the choke and can safely use used 220uF in the past. This will get you down to -62db. Not too bad considering a SS rectifier with the choke and 470uF you would be at -71db.

 

-62db vs -44db, the choke more than makes up for the slightly reduced ripple you would get from a larger value reservoir cap and SS rectification. If you are going to build single ended amps with more than flea power a choke is pretty much necessary if you want the best performance. And if you are going with LC filters than you might as well add the cool factor of a tube rectifier since it's not going to make a difference in performance going SS.

  • Like 1
Link to comment
Share on other sites

Here are two simulations between the two. One is SS rectified with large value capacitors. The other is tube rectified with a choke. I had to increase the voltage of the secondary winding to make the comparison more accurate for the same output voltage of 340v. The 5AR4 is within it's peak repetitive forward current rating and peak surge current rating.

 

The SS rectifier circuit has double the ripple compared to the tube rectifier circuit.

 

What I really like about chokes and tube rectifiers is capacitor life is extended for all filter capacitors after the choke. Along with the plots for ripple voltage at C2 I'm also posting the ripple current through C2. Yellow is the SS ripple current through C2 and the blue is tube+choke ripple current through C2. Have a look at the ugly and large current waveform. C2 will certainly have a shorter service life in the SS no choke circuit compared to the tube rectifier and choke circuit. Electrolytic capacitor life is determined by heat and the current passing through it.

 

Another bonus I like about tube rectifiers is I don't need to work out a snubber for rectifier induced ringing which doesn't typically occur with tube rectifiers but almost always does with SS rectifiers.

 

Yes chokes and tube rectifiers are more expensive and take up more space. But if we are after efficiency then we might as well just ditch tubes all together and go Class D.

 

 

Screenshot from 2022-10-28 12-45-12.png

Screenshot from 2022-10-28 12-56-47.png

Screenshot from 2022-10-28 13-13-47.png

Screenshot from 2022-10-28 13-14-27.png

  • Like 2
Link to comment
Share on other sites

On another tube thread I mentioned one of those SS replacements for a rectifier tube wanting to know what members thought of them. I decided to do some test and see just how much the voltage is increased when installing one of those SS tube gadgets in place of a rectifier tube. I would not recommended it because there is a considerable voltage increase. At least in my SE KT-88 amplifier. The K to P voltage in this amplifier increased from 388V to 421V which means the current is going to increase as well raising the wattage of the tube. Chances are the tube can handle this increase if you do not see any red platting but it will decrease the life of the tube. The part that concerned me though was the screen grid current. The max screen grid current of my tubes ,Electro Harmonix KT-88's increased to a little over 8ma when the tube is rated at 7ma max. I lowered this to 6.5ma on my amplifier before I listened to my amplifier to see what if any the improvement of using SS rectification in place of a tube. Chances are one would not have a tube failure installing one of these replacement SS rectifier gadgets but tubes are too expensive these days for me to take a chance. What all this amounts to is in my opinion I would use what the amplifier is designed for and not replace a tube rectifier with one of the SS gadgets without making some adjustments in the bias and screen grid circuits. Just to be safe without having someone doing some adjustments for the increase in voltages. Stick with what the designer intended your amplifier to rectify with. 

Link to comment
Share on other sites

3 hours ago, henry4841 said:

The part that concerned me though was the screen grid current. The max screen grid current of my tubes ,Electro Harmonix KT-88's increased to a little over 8ma when the tube is rated at 7ma max.

 

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.

 

This isn't quiescent (idle) measurements either, when signal is applied and the plate gets pulled down closer to ground the screen grid will have a large increase in current. Try and use a resistor to limit the current, which it will but how it limits current is via a voltage drop which also lowers the screen voltage. This squishes the plate curves together, you can see this via different plate characteristics graphs for different screen grid voltages in datasheets. The less voltage on the screen the more squished the plate curves are, the more voltage on the screen the more the curves open up and separate. This is why regulated screens improve the performance, a simple resistor changes the tube characteristics.

  • Like 1
Link to comment
Share on other sites

On 10/28/2022 at 10:07 AM, henry4841 said:

It is obvious that you are not a tube guy but just wanted to express your opinions of tubes in general which is fine. 

 

Err, what?

I currently have 3 tube amplifiers* (2 are tube rectified), my ss class ab amps are only used for a couple (hot) months. Came to this thread looking for some discussion on tube rectifiers (thanks @captainbeefheart), not someone to tell me how solid state is better and cheaper.

 

*have owned many, many more

  • Like 1
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...