Heater flash is a signature feature of many European tubes. I've seen it in Holland Amperex tubes, I've seen it in Blackburn Mullards, I've seen it in lots of European tubes, and since those tubes are often 50 years old and still work great, it's evident that this is NOT a "problem" but just a characteristic of their heater chemistry.
Today I built a vacuum tube based one. For large power tubes the filment needs to be soft started. So a big resistor and a bypass circuit, i didnt want transistors so I thought, how about a vacuum tube in series with the relay and the delay offered by the vacuum tubes indirect filament is a bit long 20s but still perfect for such application! Note that the voltage drop on the tube, i ended up with a voltage doubler on the tube to get close to 12v on the relay.
Hi Eric! I have been repairing a Tek 2467B and, a it has one big vacuum tube - the CRT - it also experiences a heavy draw on initial power up and uses two NTC thermistors as current inrush limiters. After initial power on, according to the theory of operation they "cease to have any effect on the circuit". Also, i have an old Marconi RC oscillator that blew its 2080 SQ tube because the NTC thermistor that it used just after the transformer had completely burned to a crisp. I couldn't even tell what it had been until I tracked down the schematics and BOM, but it was definitely a recognized solution to cold heater current.... I've just thought that maybe heater-flash caused the thermistor to burn up....the value was obviously calculated to limit the cold heater inrush, but maybe the filament flash on one or more tube pushed it over the edge?
Nice! Not sure if i remember seeing one of those before or not. With 40 years experience, I'm going to show one of my favorite, so Simple and so overlooked components Very soon on my Channel!
With a relay you can also use the old school RC delay using 2 resistors and a capacitor fed from DC, smoothed or not, using the resistors so you get a delay using the build up of current through the coil to turn it on. With the remainig tubes drive them heavily at 12V filament current and try to get the anodes to cherry red, connecting all the electrodes to the anode either direct or via 100r, so you can get max plate current through. Take the popped one and apply high voltage to do the same, might work well on a NST, especially as there should be some free gas in there now from the cathode.
+SeanBZA I do have quite a bit of cheap junk tubes. Will have to torture them sometime like I did for the bad rectifier tube in my Atwater Kent 55C radio.
Nice demo, thanks. Would you say that filament flash is a problem? I bought two 12at7 NOS tubes and did not like that they had filament flash when using them with my guitar amp (none of the new/modern tubes have filament flash) and the vendor said it is normal and that they are high quality, I am thinking of returning them ... your thoughts?
I just got an old analog light control timer. Thought it had a tube amplifier but it was actually an Amperite time delay. The delay looks like a good buffer for the tube amp. But in a motor wouldn't that be called in rush current so would an easier option be a capacitor or some other discrete component?
Here is another idea that some engineers can work out. I know that there are purist old radio folks that do not want to add anything to alter their radio in any way shape or form. Is it possible for a separate delay timer to be installed in its own enclosure outside of the tube radio or amp? I was thinking that a timer delay relay with a heater rated at 120 VAC might be used. The device would have a cord to plug into the wall outlet and a receptacle for the radio or amp to plug in. The device would actually have the entire 120 VAC reduced down to 60 VAC through an appropriate sized resistor. When the radio or amp is not on no current is flowing. A current sensing device or circuit might be installed so that when the radio or amp is turned on the timer relay starts its count down. Then when it is at the end of its time period it would supply the full 120 VAC to the radio. It might have to have a latching feature as long as current is flowing to the radio and will unlatch when the current flow stops. I have seen some time delay relays that work on 120 VAC. The relay is a double pole double throw with contacts rated at 10 Amps.
I am working on a project whereby I only want to light up the tube filament with DC current. I am putting 6.3 through them and they light up. Not very bright though. When I put 7 to 10 volts they glow real pretty!! I only did it for a few seconds. How much current can I safely use? What happens if too much? Do they just burn up or is it dangerous?
I have this 6n045t tube but is this practical what if I wanted to run 3 6dj8s in parallel filaments does this mean I have to triple up or down the resistance on the resistor and can the time delay tube handle the heavy current load I'm assuming that resistor cuts that down.
Why is the filament flash problem? There are many of ECC82 and ECC83 made by RFT (DDR) flashing every time when power up. These tubes can work for decades.
+brys555 Most tubes will not have filament flash, but at the beginning of his video, Mr. Carlson explains how he experienced a significant lot of specific types of tubes that were prematurely failing due to filament flash.
Always liked the sound of tube amps, but dont think they would last long in any car :) no longer an xm owner moved to an 806 which is not such a bad drive.
+lez briddon The have it because the filament end that sticks out at the bottom have very low thermal contact to the rest of the tube, so it's the part that heats up first and takes all the energy.
+lez briddon It's partially what +zaprodk said. And also partially that these filaments are PTC's almost. They start off at a fairly low resistance untill they heat up. The same goes for the filament in lightbulbs
Great addition Eric! Thanks for the kind mention as well.
+Mr Carlson's Lab Glad to help!
Heater flash is a signature feature of many European tubes. I've seen it in Holland Amperex tubes, I've seen it in Blackburn Mullards, I've seen it in lots of European tubes, and since those tubes are often 50 years old and still work great, it's evident that this is NOT a "problem" but just a characteristic of their heater chemistry.
Saw your comment on Mr. Carlson's Lab, nice follow-up.
Cheers,
Mark
***************
Today I built a vacuum tube based one.
For large power tubes the filment needs to be soft started.
So a big resistor and a bypass circuit, i didnt want transistors so I thought, how about a vacuum tube in series with the relay and the delay offered by the vacuum tubes indirect filament is a bit long 20s but still perfect for such application!
Note that the voltage drop on the tube, i ended up with a voltage doubler on the tube to get close to 12v on the relay.
Hi Eric! I have been repairing a Tek 2467B and, a it has one big vacuum tube - the CRT - it also experiences a heavy draw on initial power up and uses two NTC thermistors as current inrush limiters. After initial power on, according to the theory of operation they "cease to have any effect on the circuit". Also, i have an old Marconi RC oscillator that blew its 2080 SQ tube because the NTC thermistor that it used just after the transformer had completely burned to a crisp. I couldn't even tell what it had been until I tracked down the schematics and BOM, but it was definitely a recognized solution to cold heater current.... I've just thought that maybe heater-flash caused the thermistor to burn up....the value was obviously calculated to limit the cold heater inrush, but maybe the filament flash on one or more tube pushed it over the edge?
Nice! Not sure if i remember seeing one of those before or not.
With 40 years experience, I'm going to show one of my favorite, so Simple and so overlooked components Very soon on my Channel!
With a relay you can also use the old school RC delay using 2 resistors and a capacitor fed from DC, smoothed or not, using the resistors so you get a delay using the build up of current through the coil to turn it on.
With the remainig tubes drive them heavily at 12V filament current and try to get the anodes to cherry red, connecting all the electrodes to the anode either direct or via 100r, so you can get max plate current through. Take the popped one and apply high voltage to do the same, might work well on a NST, especially as there should be some free gas in there now from the cathode.
+SeanBZA I do have quite a bit of cheap junk tubes. Will have to torture them sometime like I did for the bad rectifier tube in my Atwater Kent 55C radio.
Nice demo, thanks. Would you say that filament flash is a problem? I bought two 12at7 NOS tubes and did not like that they had filament flash when using them with my guitar amp (none of the new/modern tubes have filament flash) and the vendor said it is normal and that they are high quality, I am thinking of returning them ... your thoughts?
Dear Eric! Hope you still alive, we need an update on you new lab =)
I just got an old analog light control timer. Thought it had a tube amplifier but it was actually an Amperite time delay. The delay looks like a good buffer for the tube amp. But in a motor wouldn't that be called in rush current so would an easier option be a capacitor or some other discrete component?
Here is another idea that some engineers can work out. I know that there are purist old radio folks that do not want to add anything to alter their radio in any way shape or form. Is it possible for a separate delay timer to be installed in its own enclosure outside of the tube radio or amp? I was thinking that a timer delay relay with a heater rated at 120 VAC might be used. The device would have a cord to plug into the wall outlet and a receptacle for the radio or amp to plug in. The device would actually have the entire 120 VAC reduced down to 60 VAC through an appropriate sized resistor. When the radio or amp is not on no current is flowing. A current sensing device or circuit might be installed so that when the radio or amp is turned on the timer relay starts its count down. Then when it is at the end of its time period it would supply the full 120 VAC to the radio. It might have to have a latching feature as long as current is flowing to the radio and will unlatch when the current flow stops. I have seen some time delay relays that work on 120 VAC. The relay is a double pole double throw with contacts rated at 10 Amps.
i think the "flash heater" is for fast heat up feature.. like the datasheet said 10second warm up time.. they are design for turn on TV faster.
I can't get Jurassic Park back on-line without Eric Wasatonic
I am working on a project whereby I only want to light up the tube filament with DC current. I am putting 6.3 through them and they light up. Not very bright though. When I put 7 to 10 volts they glow real pretty!! I only did it for a few seconds. How much current can I safely use? What happens if too much? Do they just burn up or is it dangerous?
I have this 6n045t tube but is this practical what if I wanted to run 3 6dj8s in parallel filaments does this mean I have to triple up or down the resistance on the resistor and can the time delay tube handle the heavy current load I'm assuming that resistor cuts that down.
Nice vid, I miss Photon though :(
You could just use an NTC thermistor to limit the inrush current.
+rocketman221projects I like that idea. I had not thought of using an inrush current limiter. It may be the simplest of all solutions.
Why is the filament flash problem? There are many of ECC82 and ECC83 made by RFT (DDR) flashing every time when power up. These tubes can work for decades.
+brys555 Most tubes will not have filament flash, but at the beginning of his video, Mr. Carlson explains how he experienced a significant lot of specific types of tubes that were prematurely failing due to filament flash.
Pretty much every ECC83/12XA7 tube I've used does this.
00:51 And I thought that the +EEVBlog had a ton of scopes and other measuring equipment
Nice, maybe a NTC thermistor would work ? Thumbs up :)
Alex
+High Voltage Yes. Others have suggested that. I think it may the simplest solution.
Eric Wasatonic Ok :)
why do tube filaments have this flash?
+lez briddon I didn't expect you are interested to vacuum tubes and XM's :D
Always liked the sound of tube amps, but dont think they would last long in any car :) no longer an xm owner moved to an 806 which is not such a bad drive.
+lez briddon The have it because the filament end that sticks out at the bottom have very low thermal contact to the rest of the tube, so it's the part that heats up first and takes all the energy.
+lez briddon It's partially what +zaprodk said. And also partially that these filaments are PTC's almost. They start off at a fairly low resistance untill they heat up. The same goes for the filament in lightbulbs
What if your thermal time delay has filament flash ? :D
Elegant
ah....using a CL90 works to prevent that.
+Amy Marie Thanks for the suggestion of a CL-90 Thermistor. Very helpful.
Your Welcome....
I'm unconvinced that the flash matters, but the simplest solution would be just to use USA valves, if you can. I don't believe any of those flash.
NTC???
just saying its not pronounced am-pure-rite or am-pure-ex its am-per-ex