tube filament specs are not written in stone...
varies from manufacturer to manufacturer...
tubes rated 6.3 volts can in fact function
anywhere from 5.9 to 6.9 volts or any value in between...
even power traffos with filament winding rated 3 amps
can supply 1 ampere more without problems...
so i am at a loss what your problem is Poty.....
varies from manufacturer to manufacturer...
tubes rated 6.3 volts can in fact function
anywhere from 5.9 to 6.9 volts or any value in between...
even power traffos with filament winding rated 3 amps
can supply 1 ampere more without problems...
so i am at a loss what your problem is Poty.....
It is not MY problem actually. If a person want to, Russian way to say, steps on a rake over and over again and not learn from that - it has all right to do this.
My main idea - the tube filament is designed, tested and specified by voltage, not current. Then - voltage supply is the best way of doing this without harm and danger. That is all. I can't continue to discuss it further here, because I don't want to make this thread one more dump place - about everything and not about anything.
Last edited:
I said that it took 30secs to reach 6.3V in first method and comparable time to fully settle at 1.55A in both methods while the voltage was constant all the time in second method.
It does not mean the same beyond start up. Its CC limit for 30secs then its CV. That would make a decisive difference if used for a mixed heater types system. If just CC series the current would be forced to be the same through all heaters all the time. When just a limit in voltage source max current its a normal shunt voltage source after reaching nominal consumption.
Bottom line, current limited voltage source works fine too.
When talking series CCS i.e. current drive, it should be per heater. To avoid any current hogging and imbalance between same spec heaters even, due to tolerances. Like in my 6V6 line preamp.
When talking voltage source with a current limit like in the bench PSU experiment or an SSLV reg for that matter, then multiple same nominal voltage heaters of different current load each can be paralleled. It will just take a little time to reach 6.3V while absorbing the inrush peak. As it did in the demonstration. Or as AJT does when applying NTC soft start.
When talking voltage source with a current limit like in the bench PSU experiment or an SSLV reg for that matter, then multiple same nominal voltage heaters of different current load each can be paralleled. It will just take a little time to reach 6.3V while absorbing the inrush peak. As it did in the demonstration. Or as AJT does when applying NTC soft start.
Not with such low spare current for universal usage first-hand, not with every tolerance-oriented tubes second hand and without any advantages of such complications, the third hand. A simple 3-leg serial regulator will outperform the current-limited version not having the artificial limitations and corresponding problems.
Sorry, not answered to your answer about your experiment and the difference between current limiting and current powering. I understand your initial phrase not correctly like in you experiment you used current limiting, while actually you used current powering.
Second addition: the experiment with 12AX7 was a real trying on the bench. In this case the heater current was stabilized in about 15 seconds with constant voltage, voltage was stabilized in more than a minute (greatly depends on room temperature) with constant current.
Last edited:
Wide margin would be much wasted heat so its better to start tight with SSLV. And set higher only if there is an issue.
A non limited voltage source is care free, but SSLV is what I was asked about.
Someone might want to exploit the unbreakable CC limited and soft start attributes or just want to listen to it VS a 3 pin reg in CV mode, whatever. Not my business.
A non limited voltage source is care free, but SSLV is what I was asked about.
Someone might want to exploit the unbreakable CC limited and soft start attributes or just want to listen to it VS a 3 pin reg in CV mode, whatever. Not my business.
Well, shunt-regulated power supplies (and all other too, excluding maybe buck regulators) are always wasted power for heat. In the low voltage area it is much more wasted, because the impedances of the power supplies and consumers are very low and to achieve comparable (to HV) regulation and efficiency you have to dump more current. Of course someone might want to bother tuning every time changing the tubes or lowering the mains voltage, but may I ask - for what? What is a single improvement of using the SSLV vs. ever simpler linear regulator? And what "unbreakable" is in the current limited version?
About SSLV VS 3pin chip regulator in heaters I can not give opinion, it will be down to individual build and subjective appreciation. king4joy will possibly report on that.
I know series CCS mode chip reg VS same chip reg in CV mode do sound different from my 6V6 line pre (one per tube). More relaxed and resolute.
"Unbreakable" is you can't burn it out no matter if you short it out. Because the CCS limit over the CV shunt part intervenes.
0.2A extra times 10V drop would bring 2W on the input CCS MOSFET sink (down to 5V drop still good if mains allowing) and 1.26W on the shunt MOSFET sink for 6.3V voltage source setting. Not too bad.
I know series CCS mode chip reg VS same chip reg in CV mode do sound different from my 6V6 line pre (one per tube). More relaxed and resolute.
"Unbreakable" is you can't burn it out no matter if you short it out. Because the CCS limit over the CV shunt part intervenes.
0.2A extra times 10V drop would bring 2W on the input CCS MOSFET sink (down to 5V drop still good if mains allowing) and 1.26W on the shunt MOSFET sink for 6.3V voltage source setting. Not too bad.
You mean in the heating circuit?
Even simplest linear regulators like LM317 have current protection. So it is also rather unbreakable sort to say.
Let me point out my phrase "comparable (to HV version) regulation".
And we already decided that 0.2A extra does not cover even tolerance margins.
Not knowing details - I'd assume something wrong with the build. If something in the heating circuit, considering all the voltages and currents are in datasheet margins, influence the signal - it means a big problem of the build.
The same words can be said about yours unbreakable ones. If something goes completely wrong in huge amount your current limiter will be broken too.
I still havent heard a thing that made the shunt regulator in the heating circuit worth trying.
-Yes its a huge problem in the build obviously.
-A 50W Mosfet in a CCS running as much current it runs regularly and is sinked for in an SSLV will run the same current & heat when the parallel part is shorted out.
Same as a 317's thermal protection, alright.
-You won't hear a thing, because I did not claim a thing. I have been asked how, not recommended it.
-A 50W Mosfet in a CCS running as much current it runs regularly and is sinked for in an SSLV will run the same current & heat when the parallel part is shorted out.
Same as a 317's thermal protection, alright.
-You won't hear a thing, because I did not claim a thing. I have been asked how, not recommended it.
I would not CCS drive into paralleled heaters.
If the heaters are in values at near the same voltage then series connect the heaters and CCS drive them.
The CCS drive removes the "shock" that voltage drive would deliver. This shock could reduce the life of the valve.
Earlier Poty said
Instead use a 290mA CCS and just current drive the heater/s
Agree. But you could still set an upper current limit to a voltage source driving parallel heaters for a soft start effect.
Let me suppose not short out, but open circuit for shunt-regulated version. Considering short circuited - let me assume the input voltage not properly chosen for SSLV and I wonder how long your 50W MOSFET will stay.
Really? And you will have different voltage across the heaters because of different tolerance for heating resistance. Very good thing for different channels, balanced design, push-pull, SRPP...
The tubes are designed for the shock.
Aha, and tune every time the current changes (I mean the current needed for proper heating of tube)!
Last edited:
I don't see why CCS/Shunt-reg would be BAD for tube heaters, i would rather say that it makes them live longer.
The heaters get way more stressed with constant voltage, because of the high inrush current - while with CCS / tight set shunt-reg it would rather just be a soft start as people have already said - which would rather make the tubes live longer. Myself i'm just running LM317 CCS w. common mode choke after for the heaters.
The heaters get way more stressed with constant voltage, because of the high inrush current - while with CCS / tight set shunt-reg it would rather just be a soft start as people have already said - which would rather make the tubes live longer. Myself i'm just running LM317 CCS w. common mode choke after for the heaters.
The only thing - after changing the tubes you should tune the current or your tubes will be stressed not only at start, but for the whole working cycle.
Why would they be stressed? tubes doesn't need 6.3v exactly, they are often rated 5.7-6.9v or at least 6-6.6v, and they are pretty much never exactly the same.
So if i set them at constant current that gives 6.1 or 6.3v between different tubes - it doesn't matter.
And the heater is pretty much the same as a light bulb - and when does those often break? - exactly, when you turn them on.
So if i set them at constant current that gives 6.1 or 6.3v between different tubes - it doesn't matter.
And the heater is pretty much the same as a light bulb - and when does those often break? - exactly, when you turn them on.