EEVblog Electronics Community Forum
Electronics => Projects, Designs, and Technical Stuff => Topic started by: 001 on October 13, 2019, 08:44:14 am
-
Hi
I`m chocked with fact what Voltage regulator tubes has BETTER temperature stability than silicon zeners at same voltage :o
But it can float a little
What the way to replace rare VR tube with zeners without stability missing?
-
Not surprising, though you can get much better stability using an amplified low TC zener, most commonly 6V8, or using a TL431 and appropriate cascoded transistors to get the shunt voltage across the actual TL431 down to somewhere in it's operation range, probably by using a 6V8 zener and a high voltage transistor to do that.
-
The TL431 is probably less noisy. :-DD
(Bandgap references, like TL431, are notoriously noisy, as solid state signals go; but glow discharges are so awful, they've been used as noise sources (http://www.mif.pg.gda.pl/homepages/frank/sheets/137/6/6D4.pdf) (with some tweaks, namely, magnetic bias).)
Also, you seem to be under some illusion as to the value of VR tube stability -- few tube circuits needed it, indeed few circuits in general need regulation. It's just done as a matter of course because, if we're starting with a 90-250VAC line spec, we can't very well work directly with that, and the output is usually variable anyway (e.g. flyback power supply) so the output must be stabilized with feedback regardless, and we might as well use a puny cheap TL431 or the like, as an error amplifier, to provide that feedback. The fact that a TL431 happens to give a <= 5% precision (and typically ~0.1% stability) is practically accidental, but we aren't going to argue with results that are both good and cheap.
The average tube circuit consists of a few amplifier stages, which will amplify equally well over perhaps a +/- 20% range of supply, give or take reduced gain or maximum power output at the low end, and reduced lifetime at the high end (excessive plate or heater dissipation; cathode wear). Unless you have 3rd-world mains (in which case the bigger problem is usually just that it's terribly unreliable..?), an unregulated, mains-derived supply will do fine. :-+
Tim
P.S. A motor vehicle gets "chocked" to prevent it from rolling away. Short, harder 'ch'. You might be shocked (soft 'sh') to discover something though. English, right?...
-
thank you for your wisdom!
How would you recommend replacing the VR tube with TL481?
Any practical schematics. lincks. etc?
-
VR tubes also have a voltage spike when starting. This is because the gas must first be ionized at a higher voltage before the tube starts regulating. If memory serves a 105 V VR tube needs 130 V to strike and start conduction.
-
VR tubes were routinely used in old communications receivers dating from WWII in to the 1960's, supplying stabilised voltage just for the local oscillator and perhaps the BFO. The HRO receivers, which actually date from the 1930's, were not originally fitted with VR tubes, but their use for ham-radio SSB reception greatly benefited from the addition of a VR tube.
It's not accurate to say the VR tubes need a spike; they need an overvoltage to light them up, that tends to occur naturally as the HT rises before the valves have fully warmed up and so are initially passing little current.
Are VR tubes rare now? They are certainly no longer in production, but I suspect that they are still fairly easy to find; I've got several in my junk box.
-
Look at the cascode circuit in https://en.wikipedia.org/wiki/Cascode Wikipededia, and contemplate replacing the bottom transistor with a TL431, and having the upper one biased so the base is at say 15V, using the resistive divider for driving both the TL431 feedback input and bias the upper transistor.
-
Yes, adding a cascode to the TL431 works quite well. I've done it before. See the thread linked below.
(https://www.eevblog.com/forum/projects/tl431-common-base-amplifier-stability/?action=dlattach;attach=133669;image)
https://www.eevblog.com/forum/projects/tl431-common-base-amplifier-stability/msg599580/#msg599580 (https://www.eevblog.com/forum/projects/tl431-common-base-amplifier-stability/msg599580/#msg599580)
-
Yes, adding a cascode to the TL431 works quite well.
Thanx
What is maximum voltage for this circuit? I`m looking for schematic to substitute 115V VR tube (supply voltage is about 280V)
-
Yes, adding a cascode to the TL431 works quite well.
Thanx
What is maximum voltage for this circuit? I`m looking for schematic to substitute 115V VR tube (supply voltage is about 280V)
It's limited by the transistor's maximum collector-emitter voltage and safe operating area.
Higher voltage ratings can be achieved by connecting more than one transistor in series.
(https://www.eevblog.com/forum/projects/ccfl-inverter-as-basis-for-easy-low-current-500v-dc-supply/?action=dlattach;attach=221726;image)
https://www.eevblog.com/forum/projects/ccfl-inverter-as-basis-for-easy-low-current-500v-dc-supply/msg931968/#msg931968 (https://www.eevblog.com/forum/projects/ccfl-inverter-as-basis-for-easy-low-current-500v-dc-supply/msg931968/#msg931968)
-
marvelous :-+
is it way to use current sink to avoid direct current?
-
Mr Carlson's Zener Diode vs Tube Regulation video was interesting because I had never considered the tempco of gas regulator tubes. Temperature coefficient depends on the gas mixture. Neon with Helium can have positive or negative temperature coefficient. Reference "British Special Quality Valves and Electron Tube Devices Data Annual 1964–65" page 810 onwards, Google Books link below.
LM723 used as a shunt regulator with an external transistor might be an option to replace high voltage zeners. LM723 has a buried zener reference so it's nice and quiet.
https://books.google.co.uk/books?id=8K83BQAAQBAJ&printsec=frontcover&dq=%22British+Special+Quality+Valves+and+Electron+Tube+Devices+Data+Annual+1964%E2%80%9365%22&hl=en&sa=X&ved=0ahUKEwiPwovMlZrlAhVYUxUIHb5-BaUQ6AEIKDAA#v=onepage&q=%22British%20Special%20Quality%20Valves%20and%20Electron%20Tube%20Devices%20Data%20Annual%201964%E2%80%9365%22&f=truehttps://books.google.co.uk/books?id=8K83BQAAQBAJ&printsec=frontcover&dq=%22British+Special+Quality+Valves+and+Electron+Tube+Devices+Data+Annual+1964%E2%80%9365%22&hl=en&sa=X&ved=0ahUKEwiPwovMlZrlAhVYUxUIHb5-BaUQ6AEIKDAA#v=onepage&q=%22British%20Special%20Quality%20Valves%20and%20Electron%20Tube%20Devices%20Data%20Annual%201964%E2%80%9365%22&f=true (https://books.google.co.uk/books?id=8K83BQAAQBAJ&printsec=frontcover&dq=%22British+Special+Quality+Valves+and+Electron+Tube+Devices+Data+Annual+1964%E2%80%9365%22&hl=en&sa=X&ved=0ahUKEwiPwovMlZrlAhVYUxUIHb5-BaUQ6AEIKDAA#v=onepage&q=%22British%20Special%20Quality%20Valves%20and%20Electron%20Tube%20Devices%20Data%20Annual%201964%E2%80%9365%22&f=truehttps://books.google.co.uk/books?id=8K83BQAAQBAJ&printsec=frontcover&dq=%22British+Special+Quality+Valves+and+Electron+Tube+Devices+Data+Annual+1964%E2%80%9365%22&hl=en&sa=X&ved=0ahUKEwiPwovMlZrlAhVYUxUIHb5-BaUQ6AEIKDAA#v=onepage&q=%22British%20Special%20Quality%20Valves%20and%20Electron%20Tube%20Devices%20Data%20Annual%201964%E2%80%9365%22&f=true)
-
marvelous :-+
is it way to use current sink to avoid direct current?
What do you mean? Are you talking about using the TL431 at higher currents than 100mA? It's possible to add a PNP pass transistor to divert current around the TL431, once it exceeds a certain threshold. Refer to page 27, figure 31, on the TL431 data sheet. When the current through the TL431 exceeds R3*VBE the BJT starts to conduct, diverting current, around the TL431. It can be combined with the cascode circuit I showed you above.
[attachimg=1]
http://www.ti.com/lit/ds/symlink/tl431.pdf (http://www.ti.com/lit/ds/symlink/tl431.pdf)
-
marvelous :-+
is it way to use current sink to avoid direct current?
What do you mean? Are you talking about using the TL431 at higher currents than 100mA? It's possible to add a PNP pass transistor to divert current around the TL431, once it exceeds a certain threshold. Refer to page 27, figure 31, on the TL431 data sheet. When the current through the TL431 exceeds R3*VBE the BJT starts to conduct, diverting current, around the TL431. It can be combined with the cascode circuit I showed you above.
(Attachment Link)
http://www.ti.com/lit/ds/symlink/tl431.pdf (http://www.ti.com/lit/ds/symlink/tl431.pdf)
No. Not for current reason but High quiet and stable voltage to use it as reference in serial regulator
-
In fakt TL431 is not good choise for HV circuit
it oscillate :horse:
-
VR tubes are not good choice, it oscillate!
Depends on what's connected to it, and in the case of the TL431, how it's connected too. Which circuit did you follow?
Tim
-
In fakt TL431 is not good choise for HV circuit
it oscillate :horse:
Was it connected in the cascode configuration I suggested above?
I fount it was stable, but larger, slower transistors can cause additional phase shift and oscillation. Try adding an emitter resistor.
(https://www.eevblog.com/forum/index.php?action=dlattach;topic=42367.0;attach=855200;image)
https://www.eevblog.com/forum/projects/tl431-common-base-amplifier-stability/msg606605/#msg606605 (https://www.eevblog.com/forum/projects/tl431-common-base-amplifier-stability/msg606605/#msg606605)
-
LM723 has a buried zener reference so it's nice and quiet.
Nope, no burried zener there. And pretty noisy.
-
Here is the Zero999's cascode with "most accurate ED TL431 model".