Under rating of Vertical Sweep Tubes - Page 3 - diyAudio
Go Back   Home > Forums > Amplifiers > Tubes / Valves

Tubes / Valves All about our sweet vacuum tubes :) Threads about Musical Instrument Amps of all kinds should be in the Instruments & Amps forum

Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 18th October 2010, 11:41 PM   #21
diyAudio Member
 
Join Date: Jun 2004
Location: Indiana
Wow, that got really heavy quite fast. I will have to save this thread for further study and try to get hold of some of those texts that you guys have referenced. Let me get your take on some fooling around I did over lunch today.

Using these 6LR8 curves (same as 6LU8):

Click the image to open in full size.

Ignore the load line on that graph as it is from another project. I used the methods described here to check out a couple of bias points with 8K a-a load.

loadmatch4-pp-beamtetrodes

I ran the numbers at a bias point of 300Vp and just under 50mA Ip. I came up with about 19W full power with about 7W being class A. I ran some other points as well and it appeared that around 25W was possible at 350V.

The interesting thing is that, if I am understanding the method correctly, the idle current does not really seem to effect the maximum output power as the class AB power calculation uses only the class B load line which is determined only by the load impedance (1/4 Zp-p) and the maximum plate voltage. The non-intuitive result seems to be that running the tube "hotter" (at the same plate voltage) does not increase power output but only moves the class A load line up and down which will effect the total amount of class A power available before transitioning into class B. Presumably this would also effect the smoothness of the transition in some way but I am unclear on exactly how.

So by the looks of it I can play plate voltage against idle current to trade total power for greater "undistorted" power or trade "undistorted" power for tube life. Sort of like a "Low bass, high efficiency and small size... pick any two" type situation.

Does it sound like I am using the method correctly? Am I interpreting the results right?

Seems like I would be wise to use fixed bias so I can diddle with it to get the best combination for my situation.
Attached Images
File Type: gif 6LR8_pentode_load.gif (62.1 KB, 463 views)
__________________
If We The People refuse to hear the truth we will be ruled by liars.
  Reply With Quote
Old 19th October 2010, 12:57 AM   #22
diyAudio Member
 
Tom Bavis's Avatar
 
Join Date: Jun 2002
Location: Macedon NY
You pretty much HAVE to use fixed bias, as the increase in current with drive in AB would increase cathode bias, shifting the operating point...
  Reply With Quote
Old 19th October 2010, 02:32 AM   #23
diyAudio Member
 
tubelab.com's Avatar
 
Join Date: Jul 2005
Location: West Virginia panhandle
Quote:
Been waiting for George to chime in. I think he has gotten well over +50% on some exceptional tubes. But then you have to wonder how long they will last when they start glowing!
NOT LONG! Operating any tube in the glow region in an invitation to a short unhappy life. Some tubes can be run hotter than others. If all of the materials used in the construction were nearly pure (Bendix Red Bank) then a little glow wouldn't hurt. If the plates were made out of melted down Corollas (early Chinese tubes) then the tubes wouldn't last long.

When the plate starts glowing (or even before glow) impurities in the materials will be outgassed contaminating the vacuum. This will lead to grid current and possible tube runaway. Some impurities can poison the cathode reducing emission.

When the screen starts glowing two possible things can happen, both real bad. When the grid glows it can deform or melt since the wires are small. The obvious consequence is a short. Yes I have seen it happen. Sweep tubes usually have G1 and G2 close together, making shorts a real possibility. Due to the close proximity the screen usually has a low voltage rating on sweep tubes. It is not uncommon to see the screen run at 150 volts or so while the plate sees 500+ volts. This leads to another failure mechanism. When the screen starts glowing it can emit electrons. When the plate and screen are at the same potential, this is of no consequence. When the plate is several hundred volts more positive than the plate, a strong uncontrolled current flow from a glowing screen to the plate will occur. This will destroy the tube and screen supply components in short order.

OK, so we don't want glow, at least not for long. How is it than we can bend the ratings?

Quote:
Some other issues to consider. Even within the same tube type, you will find different size plates, presence or absence of grid cooler fins, different plate construction/material/surface finish.
There are vast differences between two tubes of the same number. I have seen two tubes with the same numbers on them with at least a 2 to 1 difference in power handling capability. Why?

First off look at what the sweep tube does. It moves an electron beam across the face of a CRT. It does this at full power every minute that the TV set is on. There is no volume control on the sweep circuit. When many of these tubes were designed most houses had only one TV and it got used a lot. Even in Miami most houses did not have AC. We didn't get AC until the mid 60's. So the tube ran full throttle in a hot box, inside a hot house for several hours a day. It is no surprize that the sweep circuits were responsible for most TV failures. Since excess heat kills tubes, sweep tubes get lower ratings than an identical tube used in a less severe service. The thread linked in post # 18 has a bit more information.

Quote:
I guess they had to linearly sweep both ways, where as the horiz. channel only had to linearly sweep one way? ......Some TV experts around?
Been there done that too......Both sweep sections sweep only one way linearly. Then the CRT beam is switched off (blanking) while the beam is returned to the starting point (retrace). As mentioned the vertical sweep tube does indeed run class A. In fact it is a SET class A audio amp. Some bigger TV's used pentodes (SEP). It needs to run at only one frequency (60 Hz in NTSC), but many vertical sweep amps did indeed cover the entire audio spectrum. I made guitar amps by copying the vertical sweep section from a TV set way back in the 1960's. In fact the original Bottlehead SEX amp is indeed a vertical sweep amp, including the OPT!

The horizontal sweep amp has another function. It sweeps the beam AND it generates the high voltage. A 25 inch color TV runs about 25 KV at 1 mA. This alone is 25 watts of output. Well you don't get 25 watts out of a Class A SET easilly, so the Horizontal Output Tube is NOT linear. It is a switch. The HOT is either cutoff, or operating as a CCS. This is why the HOT is always a pentode. The HOT feeds its constant current into a flyback transformer, which functions like an automotive ignition coil on steroids. The current charges the coil, and then the current is switched off generating a big voltage spike that is stepped up and rectified to make HV.

Both types of sweep tubes saw a hard life and died young. The tube manufacturers worked to improve their tubes to reduce failures. This involved better materials, thicker and bigger plates, and heat radiating fins welded to the plates and grid rods. This means that there may be several versions of the same type number from the same manufacturer that have different power handling capabilities even though the specs are the same.

How do you know hard you can push a given tube? You must test. In most cases the 20% number mentioned is a given, since that is about the derating factor for sweep tubes. This is concerning plate dissipation only. The screen grid voltage rating is a wild card with sweep tubes. Some tubes will go unstable when the rating is mildly abused and some others of the same type number may be violated by a hundred volts or more.

Quote:
Using these 6LR8 curves (same as 6LU8):
I beat that one rather heavilly and found some strong ones and some wimps here:

Spud anyone?

Quote:
So by the looks of it I can play plate voltage against idle current to trade total power for greater "undistorted" power or trade "undistorted" power for tube life.
In a class A amp the tube dissipation is worst case at idle and goes down as the output power goes up. You set the bias as a trade off between tube life and distortion, damping factor and harmonic make up. The bias current has only a minor effect on power output. Due to its effect on damping factor higher tube current can clean up sloppy bass. This is why I craank the KT88's in my Simple SE to 100 mA!

In an AB or B amp the maximum dissipation is never at idle and may or may not be at maximum power output. This depends on the tubes and the load impedance. It can be measured. It is not advisable to exceed the rated dissipation at idle, since things will only get hotter as the volume gets cranked.

Quote:
"So it can take advantage of averaging and the crest factor to handle above rated power on peaks."
Consider my extreme cranking of the red board. See post # 425 in this thread:

Posted new P-P power amp design

The 35LR6 is specced at 30 watts. Red glow on these Sylvanias sets in at about 55 watts. The idle dissipation is set to 22 watts per tube. The tube dissipation goes up as the power output increases and tube dissipation is maximum at max output power. The board makes 250 WPC! Tube dissipation is 62 watts per tube at this power level. This is clearly way over spec. If this amp was being used to make continuous sine waves clearly a meltdown would be imminent. So, we don't listen to sine waves do we? Even highly compressed music has a crest factor (or peak to average power ratio to us cell phone designers) of at least 10 db. This means that if I crank this amp to the edge of clipping the AVERAGE power output will be less than 25 watts. Its the average power that melts tubes, so for all musical applications (even a guitar amp) this 250 WPC amp will not melt and the tubes will live long and prosper!
__________________
Too much power is almost enough! Turn it up till it explodes - then back up just a little.
  Reply With Quote
Old 19th October 2010, 04:35 PM   #24
diyAudio Member
 
smoking-amp's Avatar
 
Join Date: Dec 2001
Location: Hickory, NC
"and try to get hold of some of those texts that you guys have referenced. "

I've attached the diagram from Terman's book on projected cutoff biasing. The Terman book I quoted earlier has just a brief synopsis of this method. It refers back to his earlier book: "Radio Engineers' Handbook" p. 384 (1943) for a complete workup. But its simple enough to get from the diagram. The second tube would have its curve flipped and reversed below the bottom graph axis. The curves then get slid around on the axis until the projected cutoffs line up. The symmetry reached then sums the curves to give a near straight line composite curve for both. So you really just have to select a screen grid V specific plate current versus grid1 V curve off the data sheet, and then do the line projection to get the grid bias. 6LU8 pentode curves attached also.
Obviously, some room for interpretation of the projected cutoff line position due to the curvature of the tube curves.

Good to hear from you George. A refreshing perspective on when to expect 250 Watts of sound and when to expect smoke and flying parts as usual.
Attached Images
File Type: gif Projected_cutoff.gif (4.2 KB, 188 views)
File Type: gif 6LU8_bias_selector.gif (29.0 KB, 189 views)
__________________
I want a Huvr-Board!!

Last edited by smoking-amp; 19th October 2010 at 04:40 PM.
  Reply With Quote
Old 19th October 2010, 06:42 PM   #25
diyAudio Member
 
tubelab.com's Avatar
 
Join Date: Jul 2005
Location: West Virginia panhandle
Quote:
Good to hear from you George. A refreshing perspective on when to expect 250 Watts of sound and when to expect smoke and flying parts as usual.
I sort of dissappeared from the audio world for almost a month since Sherri was in town. I don't see her much lately. My workbench has been dark and any post I sent except for last nights dissertation came from my phone. we completely cleaned out one rental warehouse to save money. Some of the "stuff" will go on Ebay soon.

If no more wild cards come my way I should be able to resume my usual antics this weekend. I have 3 Simple P-P amps partially completed, and Pete sent me a shiny new red board so I can build a bigger one than his and document it in the red board thread. Two different people have directed me to a web site of an amp builder that claims to be running EL84's on 700 volts. I find it hard to believe so I am going to find out just how much I can squeeze out of a Simple P-P! Smoke and flying parts are a given.
__________________
Too much power is almost enough! Turn it up till it explodes - then back up just a little.
  Reply With Quote
Old 20th October 2010, 01:56 AM   #26
diyAudio Member
 
Join Date: Jun 2004
Location: Indiana
Thank you for the discussion which continues to be very informative. Was I correct that given a specific plate voltage in a class AB amp the bias current will not affect total power output but only the portion that is class A?

George, in your experience would a 6LU8/6LR8 be up to running 400V on the plate (120V screen) with 25mA plate current as a class AB bias point? Should get about 30W there if my quick calculations are correct. A less ambitious 350V looks good for 25W.
__________________
If We The People refuse to hear the truth we will be ruled by liars.
  Reply With Quote
Old 20th October 2010, 02:07 AM   #27
diyAudio Member
 
Join Date: Jun 2004
Location: Indiana
Sorry, just realized that I had been thinking in terms of the 6EM7 max plate voltage which is 330V. The 6LU8 is 400V so this should be no problem. (insert sheepish grin) So we ought to be able to push it even harder. I may be starting to get the disease.
__________________
If We The People refuse to hear the truth we will be ruled by liars.

Last edited by mashaffer; 20th October 2010 at 02:19 AM.
  Reply With Quote
Old 20th October 2010, 02:56 AM   #28
diyAudio Member
 
tubelab.com's Avatar
 
Join Date: Jul 2005
Location: West Virginia panhandle
The 6LU8 is happy with 400 volts and probably quite a bit more. The screen voltage may need to be higher than 120 volts to get the grid bias more negative for a given current. This allows more drive voltage to be applied which results in more power. I have only experimented with the 6LR8 in SE mode. I was running them triode wired and got to at least 450 volts without blasting the screen grid. TV sets ran them in pentode mode with 250 to 300 volts on the screen. I fixed a lot of Philco TV's in the early 70's that used the 6LU8 so I played with them then, but those memories have long since faded.

I have seen at least one P-P amp posted somewhere on this forum using these tubes. Search it up.

The bias current in a class A amp has almost no effect on power output, maybe 5%. I have seen some AB amps where cranking the current raises the power a bit more, maybe 20%, and some where raising the current actually reduces the power output. I suspect it has something to do with the load impedance, and the ability of the driver to push the grid positive. Remember if you turn up the current you reduced the negative grid voltage, now you have less room for drive voltage without going into AB2 territory. If your driver can't cope, you lose overall.

Increasing the bias current will allow the output tubes to remain in class A longer, but use up the dissipation headroom in the tubes. It also tends to reduce the higher order harmonics.
__________________
Too much power is almost enough! Turn it up till it explodes - then back up just a little.
  Reply With Quote
Old 20th October 2010, 03:31 AM   #29
diyAudio Member
 
Join Date: Jun 2004
Location: Indiana
Ah, thanks. I did find a couple of mentions and they seem to be in the same territory as I was exploring but at lower B+. Idle currents around 25mA with grid voltages around 16 to 18. Looks like we have a starting point.

Since I was planning on using voltage divider screen supply I could make it adjustable with a pot on the upper side of the divider so that I could easily experiment with different screen voltages. My logic in using the voltage divider is that I can get somewhat more stable screen voltage than a single dropping resistor would provide but without the complexity of a regulated supply for both screen and plate. The screen voltage should vary proportionally (or close to it) with variations in B+.
__________________
If We The People refuse to hear the truth we will be ruled by liars.
  Reply With Quote
Old 20th October 2010, 05:12 AM   #30
diyAudio Member
 
Join Date: Jun 2009
Quote:
Originally Posted by mashaffer View Post
I could make it adjustable with a pot on the upper side of the divider so that I could easily experiment with different screen voltages.
You will be dissipating several watts here - it will require a rather robust pot.

A voltage divider screen supply would have significant "sag" under load - often considered desirable in a guitar amplifier, but probably not appropriate in your intended application.
  Reply With Quote

Reply


Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
Sweep tubes and G3 rsumperl Tubes / Valves 3 5th October 2010 03:54 AM
Do Sweep tubes Glow Blue? wrenchone Tubes / Valves 26 20th July 2009 10:32 PM
Oddball Sweep tubes in PP mashaffer Tubes / Valves 8 4th June 2009 01:17 AM
Some interesting reading about Sweep tubes Vargas Tubes / Valves 2 28th April 2009 04:48 AM
Thoughts on horizontal sweep tubes? DaveCan Tubes / Valves 14 28th August 2008 09:24 PM


New To Site? Need Help?

All times are GMT. The time now is 03:02 PM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2