 
|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
|
|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
142.197.196.110
In Reply to: RE: Here's the chart you ask for posted by sony6060 on July 18, 2023 at 17:50:56
BTW- per self acclaimed Tre, he knows more about rectifier tubes than the manufactures that made the tubes.
Follow Ups:
Tre is correctly describing how to interpret valve diode data sheets.
I will happily reiterate that any identification of a max capacitance is related to preset conditions. What dictates any capacitance value used is the continuous pk value of the current pulse occuring twice each mains cycle, as that pk value determines what the manufacturer rates the cathode as being able to sustain for a rated service life.
But, there are no charts stating maximum capacitance with less than full load current on tube or transformer resistance. Tre's info is useless to the audiophile here.
While not all rectifier data sheets have them, some include 3 charts that can be used to determine if a particular combination of PT, operating conditions and first cap value is permissible. Data sheets from GE often include them.
It's a 3 part "test" and if a particular combination passes all 3 tests then it is permissible. You must be able to input accurate data from a particular application in order to run the tests. There is not a single, simple, chart that covers all possibilities.
Here's an example of how they are used.
On one of my projects I used a 47uf cap with a 5Y3GT rectifier and someone responded saying that the first cap was too large. In response, I applied the actual conditions of use to the charts and showed that - under those specific conditions - that the use of a 47uf cap was permissible even though the value listed in the data sheet is 20uf.
Here's a link to that discussion. The charts are from the GE data sheet.
"Tre's info is useless to the audiophile here."
I'm an audiophile. It's not useless to me.
Does Tre's info tell you maximum size of capacitor to use on the 5U4GB rectifier with certain current draw and transformer secondary resistance?
"Does Tre's info tell you maximum size of capacitor to use "
YES. Tre pointed out that A) the information in the datasheets prescribes maximum allowable current, and B) the maximum value capacitor within that limit can be determined using PSUD.
No, but if you are smart enough you can calculate it. Or the PSUD2 will show you.I was only trying to dispel the notion that the value in the data sheet is a overall max. Many people make that mistake but it is only the max value for the condition shown. Change the condition and the max allowable value changes.
BTW I don't use capacitor input power supply filters. Mine are all choke input filters. With a critical inductance input choke between the rectifier tube and the first cap, the cap can be as big as you want because the choke causes each diode (in turn) to be on for it's full half cycle and there are no current spikes. The charging current (the current that flows into the filter) is only a tiny bit higher than the load current (the current that is drawn by the audio circuit) and it is sinusoidal. Any ripple that isn't filtered out is also sinusoidal. A cap input power supply filter will have saw tooth ripple that is nasty to listen to vs. a sine wave.
I typed all of that just in case you wanted to know.
Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"
Edits: 07/19/23 07/19/23 07/19/23
Pages 1174, 1175 and 1176 of the forth edition of the Radiotron Designer's Handbook gives the math to calculate what the peak diode current will be for a given set of conditions. The section is titled "(iii) To determine peak and average diode currents"The load current, the load resistance, the voltage across load, the total effective plate supply impedance per plate (Rs), the value of the first cap.
All of that and more are part of the equation.It is not for the faint of heart.
Page 1163 says that when designing a cap input filter power supply you need to use resistors (or a secondary winding of the power transformer that had a good amount of DCR) and to make sure that the peak plate current rating is not exceeded. The cap value is just one part of that process. Basically this book is saying the same thing I have been saying all along. Increase the Rs and you can increase the value of the first cap. As long as the peak plate current rating is not being exceeded your good. "in any case where the peak plate current is too high it may be reduced by adding resistance in series with each plate or by increasing the effective impedance of the transformer to give the same result"
The charts in the data sheets are just there so we don't have to calculate or measure it. Nothing more.
Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"
Edits: 07/19/23 07/19/23
There's no reason to make this personal. Tre is correct - the maximum capacitance that can be used is purely a function of peak current. That, in turn, is determined by the transformer voltage and series DC resistance ahead of the capacitor. 40uF is utilized in the GE 5AR4 datasheet as an approximate value that would have been in common use for B+ filtering at the time. To accommodate this, the curves for RMS voltage VS output current VS series resistance hinge on this value. Also note that the Amperex datasheet for this tube uses 60uF as the reference value for the same type charts. Why didn't they use 2uF or 470uF? Because in the era in which these sheets were published, such values wouldn't have resulted in curves that would be immediately useful for equipment designers. The designers could still extrapolate the maximum current capability from the curves, then applied it to their specific Rs, voltage and current, but thats a lot more work.Just to be clear on this point, do you really believe 40uF would be the maximum that can be safely used if we insert a 100KΩ resistor in series between the rectifier cathode and the first cap? Under those conditions, the 5AR4 would barely even know the cap is there.
Edits: 07/18/23
First, there is no chart showing various maximum capacitance with different transformer winding resistance or reduced current from rectifier maximum. I use the manufactures maximum recommended capacitance.
Now who would use a 100K resistor in-between a rectifier and input capacitor- nobody.
"Now who would use a 100K resistor in-between a rectifier and input capacitor-"
But do you believe doing so would still mandate a maximum value of 40uF? That's the question.
One could use 220uF with 100K resistor, but what is the point. AGAIN, there is no chart showing maximum capacitance with regard to transformer winding resistance or less than maximum rectifier current draw.
This forum is for practical applications, not useless design discussions such as using a 100K resistor in-between the rectifier output and first capacitor.
I am glad Tre is your friend and you want to defend him as that is what your replies are about, not practical design.
I am defending TRE on all of this, and
we don't always agree! (for a given use).
Give TRE a break! He's right on this.
-Dennis-
Problem as Tre is correct most important spec is the repetitive peak current, there is no recommended capacitor spec with regard to peak current. No one can use his suggestion. However, there is a maximum capacitor value of 40uF from the tube manufacture for the 5U4GB. Everyone can use that.
I was commenting on the correctness of TRE's
reasoning. It is perfectly sound, provable and reasonable.
It's all academic if you're building for a
definite set of purposes to get a better overall product.
Then, you adjust the rules only as needed, but
you don't deny the accuracy of TRE's work.
-Dennis-
Dennis, I did not adjust any rules. Show me a maximum capacitance chart for a 5U4GB based upon peak current and I will stand corrected. Problem is there is no chart. So, again use 40uF as recommended by the manufacture.
If you find charts to be confusing or not as explicit as you would like, just use PSUD to model a supply. If you design a supply that violates the limits of a particular rectifier tube type it will tell you.Earlier, Tre posted two PSUD sims of power supplies using a 5AR4 whose data sheet lists 60uf as an input cap.
The first PS used a 20uf cap and PSUD warned that the 20uf was too large. This was due to the particular specs of that design - specifically the combination of very low PT resistance and a relatively large current load. The second PS he posted had different design specs and used a 220uf cap and PSUD indicated that this was OK - when used under those conditions.
So the cap value listed in the data sheet is neither the maximum you can use under all conditions nor is it safe to use under all conditions.
If you set PSUD up to simulate a particular PS you can easily play with cap values and find out what the maximum cap value is for that particular supply. You may not want to use the maximum value, for a variety of reasons, but it will tell you exactly how far you can go without violating the data sheet specs and damaging the tube.
Edits: 07/23/23
I've seldom seen posts from anyone so dedicated to remaining uninformed.
Wrong, I am very informed and especially so about tubes. You refuse to admit no manufacture recommends a value capacitor based upon peak repetitive current. However, manufactures do recommend maximum capacitor value.
State whatever you want, but you are wrong about this subject and capacitor value.
No they don't. Here's a GE 5U4GB data sheet. Cap value is not listed under Maximum Ratings. It's listed under "Typical Operation".
A 40uf cap isn't the maximum that can be used just as a 10H Choke isn't the maximum value that can be used in a choke input filter.
Note that it does list Transient Peak Plate Current Per Plate under Maximum Ratings. That is determined by a combination of operating factors, only one of which is the cap value.
The factors involved in the combination interact with each other and NO SINGLE FACTOR DETERMINES SAFE USE.
That's why one of the two PSUD sims Tre posted warns that one combination (with a 20uf cap) is unacceptable and the other (with a 220uf cap) is safe.
Here's the RCA data sheet for the 5U4GB. The note at the bottom clearly indicates that 40uf in NOT the maximum.
It says "WHEN capacitance values higher than 40uf ARE USED" . . . one of the other factors (plate supply impedance) must be adjusted. If you do so you can use a cap value larger than 40uf. This is about as plain as it gets.
Exactly as I been stating with many posts. 40uF. No chart with repetitive peak current to capacitor value as I been stating all along.
Frankly, I do not completely trust PSU2. However, I did state Cary Audio used 470uF on a 5U4GB and no one has complained of reduced rectifier life. If reduced from 50K hours to 25K hours, who would care.
All of you defending Tre's comment. I am glad you are Tre's friend, but it has absolutely zero to do with what I have been posting. And, I easily admitted I do not have Tre's talents. What more do all of you want?
"Frankly, I do not completely trust PSU2."
Apparently, you don't trust RCA either. The footnote on the data sheet very clearly indicates that 40uf is not the maximum.
There are ways to calculate what the peak plate current will be (the Radiotron Desinger's Handbook has the math) and ways to reduce that current to keep it below the max rating without lowering the cap value but as you say, there is no chart.
Such is life.
Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"
Calculating is beyond most of us. However, Cary Audio uses 470uF and his amp been out or 15 years without any complaints.
they must have a large amount of DCR in the secondary of their power transformer and/or the current supplied must be comparatively small.
"Calculating is beyond most of us."
As I said, designing power supplies for audio circuits should be left to the engineers.
Question, which Cary Audio amp uses 470uf as an input cap in the power supply? Thanks.
Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"
I forget the model Cary Amp, just remember the advertising bragging about the large cap 15 years ago. Iy is not a high resistance transformer. 1940 thru 1960 transformers hv secondaries were wound with high resistance. New transformers are much lower secondary resistance. I bet the Cary Audio transformer was less than 30 ohms HV secondary.
Using PSU2, higher capacitance capacitors actually add little additional peak to average current.
Just because the ad said there was 470uf of capacitance in the power supply that doesn't mean it was in the C1 (input cap) position.In a CLC or a CRC filter, the second capacitor can be much larger without causing problems because it is "hidden" from the rectifier by the L or the R.
It is just the first C (the cap directly following the rectifier tube, C1) that we are concerned about.
To know for sure you would have to look at the schematic.
Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"
Edits: 07/24/23
The schematic shows 470uF directly connected to the 5U4GB. Tre, why are you constantly trying to discredit all of my posts?
Wait just a minute here. You said "I forget the model Cary Amp, just remember the advertising bragging about the large cap 15 years ago." and now you are saying you have seen the schematic.
I'm not sure I believe you. are you just talking through your teeth?
Maybe your posts need to be discredited. Show me the schematic or I call BS.
Don't be making up shit.
Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"
You are a jerk. I saw schematic in a review 15 yeasr ago. Who knows where the review is. Go away.
Here it is Cary Audio SLI-80 as t came up on a search Cary Audio 5U4GB. It has two 5U4GB. Now, you will still say I am a liar without the schematic, but spec is below with PS caps.
Edits: 07/24/23
Found this, shows 1200uF @ 450 volt capacitors in power supply as stated below. The original has two 1000uF @ 450 volt caps in series. See below:
SLI-80 INTEGRATED AMPLIFIER
Next
Previous
SLI-80 Integrated Amplifier
Previous
Next
ABOUT THE SLI-80 INTEGRATED AMPLIFIER
This item is replaced by SLI-80 Heritage Series (HS) tube amp and is no longer in production.The following section describes the SLI-80 basic specifications. The specifications are subject to change without notice or obligation.
Circuit Type Push-Pull Ultra-Linear Pure Class AB-1
Power Output 40 Watts - Triode
80 Watts - Ultra-Linear
Input Sensitivity .45 Volt for full output
Noise and Hum -82 dB below full output
Frequency Response 19 Hz to 23 KHz +/- .5 dB
Inputs CD, AUX1, AUX2, Headphone jack
Tubes 2 ea - 6922 Input Buffer Preamp
2 ea - 6SN7 Pre Driver/Phase Inverter
4 ea - KT88 Output Tubes
2 ea - 5U4 Rectifiers
Speaker Posts 5 way, gold-plated
Tube Sockets Ceramic w/ silver pins
Power Transformer 1 ea. - EI Laminate, Silicon Impregnated
Output Transformer(s) 2 ea. - EI Laminate, Silicon Impregnated
Resistors 1% Metal Film
Capacitors Oil filled coupling, (copper optional)
Power Supply Capacitors 2 ea - 1200µF @ 450 VDC
6 ea - 10µF @ 400 VDC Film & Foil
AC Cord 3 Conductor Shielded, Detachable
AC Power Requirements 117/234 VAC @ 50/60 Hz
Consumption 166 watts - Operation
Warm-Up Time 3 Minutes
Break-In Time 100 hours of playing time
Finish Textured black steel chassis with machined, black anodized aluminum front panel
Weight 42 lbs
Dimensions 7″ H x 17″ W x 16″ DNow, stop calling me a liar Tre. I am sick of it.
Edits: 07/24/23 07/24/23
Your you next BS towards me, the input choke. No Tre, there is no input choke on chassis as seen in picture. Where is your apology?
Edits: 07/24/23 07/24/23
I can only imagine how badly those rectifier tubes are being tortured.
I would still love to see a schematic, there must be some kind a compensating measures being taken.
Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"
To all friends of Tre stating I was full of BS, no apology from Tre. He still is defending the high capacitance with as he states a compensation network. There is no compensating network in this amp and Dennis Had the designer did make mistakes.
Go away Tre.
Sorry, but I believe you might need to go away. Clearly, you can't supply any schematic or any other data to support this over the top back and forth.
-Rod
Perhaps best not to dictate what this forum is for.
From the posting rules - What you can post:
You can post anything relevant to the world of audio, music, video and home theater. Remember to place your message in the correct forum area; that way it is more likely to attract the attention of someone knowledgeable in the subject.
To be fair I include this. Cary Audio designed an audio amplifier with 470uF connected to I believe a GZ-34 rectifier rated for 50uF about 20 years ago. No one has complained of tube rectifier issues and tube is operating at full current capacity. Perhaps the rectifier is reduced from a 50 year to 25 year life.
And, Hallicrafters Ham receivers connected 40uF to 5Y3GT 30uF rated rectifier tubes. 5Y3GT tube is operating at 80ma to 110ma in various models. No issues.
Here is a rectifier tube data sheet that says under the conditions of 325vac per plate, 60ma. current to the load and 65 ohms Rs that if you want to use a input cap with a value greater than 40uf you might need to add some more series resistance (Rs).Using something greater than the 65 ohms shown in the example will keep the peak repetitive current from exceeding the tubes rating even though the input cap value is bigger than 40uf.
I hope you can get your head around the implications of this.
I know you don't understand the math and how the voltage and load current and series resistance affects the permissible value of the input cap but they do. I don't understand it in total but I understand the concept.
To me it is clear (and it should be clear to you) that the max value, for the input cap of a power supply filter, is conditional. Not fixed.
Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"
Edits: 07/19/23
Now, who would use a 275ma 5U4GB at 60ma? Please.
That data sheet is for a 84/6Z4.
Now you are just looking to argue, you are not looking to learn.
Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"
No Tre, a 5U4GB is 275 milliamp DC current out as I stated. Spec here:
Vh Ih VaMax IkMax
5 3.0 PIV=1550V Tube drop = 50V @ 275mA DC
the screen shot I posted showing that you need to increase the series resistance to use a larger cap is a 84/6Z4 data sheet. I never said it was a 5u4 data sheet. I just said "Here is a rectifier tube data sheet..."Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"
Edits: 07/19/23 07/19/23
I am done.
"Now, who would use a 275ma 5U4GB at 60ma? Please."
Why do you come to a forum populated by more knowledgable members and then argue with them? You could at least try to understand the concept being presented here. Your original assertion regarding this issue is simply incorrect. If you don't want to believe Tre or myself, maybe consider sending an email to Max Robinson, Tim Robbins, Jim McShane or someone else with a name you might trust and ask about this. Alternately, you could post this question to a different forum, perhaps DIYAudio or AudioKarma. Feel free to come back here and rub my nose in it if I'm wrong.
Argue? How about common sense. Use manufactures recommended maximum capacitance on a rectifier tube. Tre states using only 60ma from a 275ma 5U4GB can exceed 40uF on a 5U4GB. Who would derate the tube to less than 25% current rating? Hammond does not make a 100 ma transformer with more than 2 amps at 5 volts rectifier current. 5U4GB takes 3 amps.
Common sense.
"Hammond does not make a 100 ma transformer with more than 2 amps at 5 volts rectifier current. 5U4GB takes 3 amps."
Except for the 363CX, 369KX, 370EX JX LX MX etc.
They make a lot of transformers that have 5V windings sufficient to run any rectifier.
kyle, playing games. I stated there is not a transformer at 100ma that has a 3 amp @ 5 volt winding. Tre stated for 60 ma load and that is what I quoted. Who would buy a 200 ma transformer for 60ma DC load. Hammond transformers start at 200ma with a 3 amp @ 5 volt winding. Nice try kyle.
As long as we're trying to be close to accurate, Hammond transformer 374AX is 138mA - 720 ct with a 5V 3amp winding. 274X is 126mA - 750 ct with a 5V 3amp winding.
I don't know if Hammond actually makes a 100mA transformer with heater windings.
Also, why would it need to be Hammond? There are other manufacturers including some that will custom wind whatever you need for not much more than off the shelf iron.
So as you say, nice try.
Pay Heyboer $250 for a custom transformer. I pay Hammond $125 for same item. Argue with me as much as you please, the 5U4GB maximum capacitor value from the tube manufacture stays vs your idea of calculating peak current with no capacitor value chart that means nothing to other audiophiles.
"the 5U4GB maximum capacitor value from the tube manufacture stays vs your idea of calculating peak current with no capacitor value chart that means nothing to other audiophiles"
Although not a 5U4GB, my post above explains how the 3 charts found in some data sheets are used to determine whether or not a particular cap value is safe to use in a particular situation. The formulas used are right on the charts. I'm not very technically oriented at all and I was able to work though the steps.
In that amp the 5Y3GT, whose data sheet lists a 20uf cap value, safely uses a 47uf cap according to the charts in its data sheet.
The link I provided goes through the process - step by step - so it shouldn't be too hard to follow if you are actually interested in understanding it.
Did you read the post I linked and work through the process using the charts?
"that means nothing to other audiophiles"I agree with you.
Designing power supplies for audio amplifiers needs to be done by engineers who do understand these things and most audiophiles are not engineers.
Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"
Edits: 07/20/23
"Who would buy a 200 ma transformer for 60ma DC load. "
I would. Overkill sounds nice.
This power transformer (back left) with the 5u4 and a LCLC filter is dedicated to that single 300B operating at 64ma, 385 volts plate to cathode into a 4k load impedance. Electric Print Partial Silver, Single Secondary output transformer (back right).
By the way, the driver stages are on their own chassis with a tube rectified power supply (on it's own chassis) that is LCLC with a constant current feed VR tube shunt voltage regulation.
So I have a 4 chassis amplifier. 2 output stages (one shown) each on it's own chassis with it's own power supply, 1 stereo driver chassis and 1 power supply chassis for the driver stages.
Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"
Over pay as you please. I'll pass myself.
Those power transformers and chokes came from 2 RCA series voltage regulated power supplies I picked up for a song.I built a pair of OTL amplifiers with those two chassis. A series voltage regulator capable of high current uses many pass tubes and all those tubes sockets were perfect for an OTL amplifier that uses many low plate resistance output tube in parallel. Think in terms of the Atma-Sphere M-60 amps with 8 6as7g twin triode tubes per mono block chassis.
What is funny is that the 6as7g was designed as a pass tube for series voltage regulators so I am sure those tubes feel right at home in those tube sockets. :-)
Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"
Edits: 07/20/23 07/20/23 07/20/23
Tre your talents are incredible, exceeds my talents.
.
Have Fun and Enjoy the Music
"Still Working the Problem"
Load current, Rs (series resistance per plate) and voltage makes all the difference.
Thank you TK.
Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"
This post is made possible by the generous support of people like you and our sponsors:
Top of Page ]
[ Contact Us ]
[ Support/Wish List ]
[ Copyright © 2024, Audio Asylum, all rights reserved ]
