Heater regulator for guitar amp? - Page 2

darkskeptic · 5th July 2012, 03:51 PM

Thanks everyone, for your help. Think I'm just going to pull the resistors out (for now).

Simon B · 5th July 2012, 04:10 PM

Data at dawn then Enzo?

Here's mine:

On an EMC 213 emission type tester, with correct settings as per the manual, testing for 'quality', and testing valves from cold.

The meter stays at the left hand stop for a while, then rises reasonably quickly towards it's final reading, slowing towards the end. I timed how long it took from plug-in to the meter needle appearing to be within a needle's breadth of its final position. Not high precision, I know, but a start, at least.

I tested five different valves, results as follows, times in seconds:

JJ ECC83S #1 16 and 13

JJ ECC83S #2 15 and 13

Sovtek 12AX7 25 and 25

RCA 6L6GC 26

MESA 6L6GC 40

The double triodes were new, the 6L6GCs were used but still reading good for quality.

Still undetermined is resistance, cold and hot. We didn't lay a wager on the times, I'm certainly willing to lay one on the resistances changing substantially during warm up.

Connecting any valves heaters in series without current regulation is poor practice, let alone uncontrolled warm up types.

Simon B · 5th July 2012, 04:18 PM

Quote:

Originally Posted by darkskeptic

Thanks everyone, for your help. Think I'm just going to pull the resistors out (for now).

Not a bad plan. Suitably rated resistors will have more or less constant resistance, so peak heater dissipation during warm up will be less than with a voltage regulator. Dissipation should be about a dozen watts running, higher than this during start-up, 20 to 25W rating should be about right.

Enzo · 5th July 2012, 06:03 PM

Oh, no argument about the tubes taking a moment to heat, and that their resistance changes as they do so. If you want to replace a tube heater with a resistor, you need to determine the hot resistance, not the resistance you measure with a meter. (Hot resistance would be current through and voltage across, then Ohm's Law.)

So if you want to wager on the resistance change, well you won;t lose. I'd buy you a pint anyway.

MY only beef in the debate is whether that is a concern for the amp owner. My position is that any brief skewing of the currents and voltage drops really does no damage to the tubes. As someone who works on these amps for a living, I can count on my fingers the number of tubes I have had to replace for failed heaters in the last decade. They are just more robust than that.

SOme people freak out when certain brands of 12AX7 flare bright white for a second at power up. Plain old 6.3,not in some series arrangement. I don't remember which, but after a time or two you get used to seeing it.

The Peavey total series heater string like in this ValveKing is fairly recent, so not having seen a heater related failure there doesn;t tell me much. The Peavey Classic 30 has four little EL84s with heaters in series. They have been making them that way the last 18 years. Very popular amp. EL84s fail or wear out quickly enough in ANY amp, but I find darn few heater failures in them. Peavey also runs 12AX7s in their preamps in series across DC power supplies. Like four of them strung between +24 and -24, or three across +/-18vDC. Never seen one burnt out... so far. Keeps hum down, running them that way.

One design idea behind these series heaters is that a common weakness in tube amps is the heater connections. Talking printed circuit boards now. COnnector pins carrying the heater currents heat up and even burn out. Heaters stop or get intermittant. Bugera even sent out bulletins saying just hardwire the transformer molex connector. By running these heaters in series, they only need a 0.9A current supply. The starting surge will be greater than that, but it lasts only seconds and is not going to heat the connectors.

That 0.9A will need to be at a higher voltage, but the power transformer can be wound with thinner wire, and the connectors won;t burn up like they do on high current supplies.

darkskeptic · 5th July 2012, 06:43 PM

My initial thought was that in this instance that a resistor in series with a heater string would still be inferior to a current source, as there would be an initial (limited by the linear resistor instead of heater) current spike on startup. Eventually, I may try to put in a current source, so tubes could be pulled or added in pairs with little thought to the state of the heaters or screen resistors. However, if the evidence points towards tube heaters being more rugged than suspected in this application, then it may not matter a whole lot other than the convenience of automatically delivering the right currents. If I'm horribly wrong here, hey I'm learning, and working through this...

Enzo · 5th July 2012, 08:23 PM

COnsider this. If I am right and it doesn't much matter, then adding current control circuitry certainly won't hurt anything or even get in the way. If it does make a difference, then your idea sounds like it would work. SO doing it doesn't really present a lot of downside.

Tube heaters have a low resistance cold, and it rises as the filament gets hot. Without me thinking it through, it would seem to me that a plain resistor would be better than a tube heater in that regard. The resistor is already at the hot resistance of the tube. It is as if there is already a hot tube in the socket. If you just had four resistors in series, there would be the target current from the first moment, and it would remain steady. NO cold surge. Now if you replaced one resistor with a tube, then you would start to see a cold surge. The more resistors you replace with tubes, the greater the cold surge, because the total resistance cold goes down. SO it seems to me that two tubes and two resistors would be easier on the tubes than it would be with two more tubes in place of the resistors.

That is an opinion, I could be wrong.

darkskeptic · 5th July 2012, 09:20 PM

You are right, resistors start out at a resistance and stay there. heaters would start low resistance and normalize once warm. I'm just wondering how much extra heating time a ballast resistor would add to startup... Probably would take less time to heat up than using a current source but more time than if it was all heaters in series. For the time being, I'm just gonna pull the screen resistors for 2 of the tubes, and when I find a suitable current source to put in, I may go with that. The current source would allow me to simply jumper the heater pins in the sockets to remove half the tubes. If I need full power, no disassembly would be needed to go back.

darkskeptic · 5th July 2012, 10:23 PM

With 2 tubes out, will the bias for the remaining tubes stay within expected range? Also, the 16/8/4 ohm switch will need to be changed to half the actual speaker impedance right?

Simon B · 11:59 PM

Quote:

Originally Posted by darkskeptic

With 2 tubes out, will the bias for the remaining tubes stay within expected range?

Possibly, but definitely check it!

Quote:

Originally Posted by darkskeptic

Also, the 16/8/4 ohm switch will need to be changed to half the actual speaker impedance right?

Sounds right to me

Quote:

Originally Posted by Enzo

.........The more resistors you replace with tubes, the greater the cold surge, because the total resistance cold goes down. SO it seems to me that two tubes and two resistors would be easier on the tubes than it would be with two more tubes in place of the resistors.

That is an opinion, I could be wrong.

Again, sounds right.

I know that actual heater failure is normally quite rare, but I believe that thermal cycling of the whole valve has a big impact on tube life. This is based on some reading and reasoning, rather than any testing on my part.

One thing that particularly stuck in my mind was an account of much extended tube life being achieved in the WW2 Colossus cryptanalysis computer at Bletchley Park, partly by never turning the thing off.

Obviously heater dissipation is a fairly small fraction of the total in an output valve, but as far as I know, the mechanical tolerances in valves are tightest close in to the cathode, relaxing somewhat further out. Also, possible effects on the coating of the cathode might be worth considering.

Good point about not putting darkskeptic off, Enzo. I hope I haven't done that, if I have, then apologies and yes, I'm probably over-doing it, these amps clearly do work.

I don't suppose there's a big rush on to research this stuff, so we'll probably have to make do with discussing it over a few beers. My round I think.

darkskeptic · 6th July 2012, 10:15 AM

All your help was awesome in this, and even the more heated discussions bring other knowledge and viewpoints out. What can and can't be done in audio is not always obvious to a beginner like myself. Ive read Morgan Jones' Valve Amplifiers book, and several other books on audio, but I learn just as much from this forum... The amp works and sounds great now at 50W, and in the future I may experiment further with it.

5th July 2012, 04:10 PM	#12
Simon B diyAudio Member Join Date: Dec 2011 Location: North-East England	Data at dawn then Enzo? Here's mine: On an EMC 213 emission type tester, with correct settings as per the manual, testing for 'quality', and testing valves from cold. The meter stays at the left hand stop for a while, then rises reasonably quickly towards it's final reading, slowing towards the end. I timed how long it took from plug-in to the meter needle appearing to be within a needle's breadth of its final position. Not high precision, I know, but a start, at least. I tested five different valves, results as follows, times in seconds: JJ ECC83S #1 16 and 13 JJ ECC83S #2 15 and 13 Sovtek 12AX7 25 and 25 RCA 6L6GC 26 MESA 6L6GC 40 The double triodes were new, the 6L6GCs were used but still reading good for quality. Still undetermined is resistance, cold and hot. We didn't lay a wager on the times, I'm certainly willing to lay one on the resistances changing substantially during warm up. Connecting any valves heaters in series without current regulation is poor practice, let alone uncontrolled warm up types. __________________ Engineer: One who can do for 10 shillings what any fool can do for 10 pounds.

Thread Tools	Search this Thread
Show Printable Version Email this Page	Search this Thread: Advanced Search

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
how to wire 6.3VDC heater on a heater with center tap (12.6V series, 6.3V parallel)?	jarthel	Tubes / Valves	9	14th April 2012 05:36 PM
Zener heater regulator	arold19	Tubes / Valves	18	28th July 2010 06:56 AM
Regulated Heater on the 6SN7 Aikido Plate to Heater question	john65b	Tubes / Valves	6	24th November 2009 05:44 AM
A better LDO heater regulator	astouffer	Tubes / Valves	6	10th November 2006 06:02 AM
6.3v Heater Regulator Schematic	cantskienuf	Tubes / Valves	3	29th April 2005 12:20 PM

New To Site?	Need Help?
Register to Participate Search Privacy Statement Contact Us	Frequently Asked Questions Did you forget your password? Mark Forums Read


	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

5th July 2012, 03:51 PM	#11
darkskeptic diyAudio Member Join Date: Apr 2012	Thanks everyone, for your help. Think I'm just going to pull the resistors out (for now).

5th July 2012, 06:03 PM	#14
Enzo diyAudio Member Join Date: Sep 2004 Location: Lansing, Michigan	Oh, no argument about the tubes taking a moment to heat, and that their resistance changes as they do so. If you want to replace a tube heater with a resistor, you need to determine the hot resistance, not the resistance you measure with a meter. (Hot resistance would be current through and voltage across, then Ohm's Law.) So if you want to wager on the resistance change, well you won;t lose. I'd buy you a pint anyway. MY only beef in the debate is whether that is a concern for the amp owner. My position is that any brief skewing of the currents and voltage drops really does no damage to the tubes. As someone who works on these amps for a living, I can count on my fingers the number of tubes I have had to replace for failed heaters in the last decade. They are just more robust than that. SOme people freak out when certain brands of 12AX7 flare bright white for a second at power up. Plain old 6.3,not in some series arrangement. I don't remember which, but after a time or two you get used to seeing it. The Peavey total series heater string like in this ValveKing is fairly recent, so not having seen a heater related failure there doesn;t tell me much. The Peavey Classic 30 has four little EL84s with heaters in series. They have been making them that way the last 18 years. Very popular amp. EL84s fail or wear out quickly enough in ANY amp, but I find darn few heater failures in them. Peavey also runs 12AX7s in their preamps in series across DC power supplies. Like four of them strung between +24 and -24, or three across +/-18vDC. Never seen one burnt out... so far. Keeps hum down, running them that way. One design idea behind these series heaters is that a common weakness in tube amps is the heater connections. Talking printed circuit boards now. COnnector pins carrying the heater currents heat up and even burn out. Heaters stop or get intermittant. Bugera even sent out bulletins saying just hardwire the transformer molex connector. By running these heaters in series, they only need a 0.9A current supply. The starting surge will be greater than that, but it lasts only seconds and is not going to heat the connectors. That 0.9A will need to be at a higher voltage, but the power transformer can be wound with thinner wire, and the connectors won;t burn up like they do on high current supplies.

5th July 2012, 06:43 PM	#15
darkskeptic diyAudio Member Join Date: Apr 2012	My initial thought was that in this instance that a resistor in series with a heater string would still be inferior to a current source, as there would be an initial (limited by the linear resistor instead of heater) current spike on startup. Eventually, I may try to put in a current source, so tubes could be pulled or added in pairs with little thought to the state of the heaters or screen resistors. However, if the evidence points towards tube heaters being more rugged than suspected in this application, then it may not matter a whole lot other than the convenience of automatically delivering the right currents. If I'm horribly wrong here, hey I'm learning, and working through this...

5th July 2012, 08:23 PM	#16
Enzo diyAudio Member Join Date: Sep 2004 Location: Lansing, Michigan	COnsider this. If I am right and it doesn't much matter, then adding current control circuitry certainly won't hurt anything or even get in the way. If it does make a difference, then your idea sounds like it would work. SO doing it doesn't really present a lot of downside. Tube heaters have a low resistance cold, and it rises as the filament gets hot. Without me thinking it through, it would seem to me that a plain resistor would be better than a tube heater in that regard. The resistor is already at the hot resistance of the tube. It is as if there is already a hot tube in the socket. If you just had four resistors in series, there would be the target current from the first moment, and it would remain steady. NO cold surge. Now if you replaced one resistor with a tube, then you would start to see a cold surge. The more resistors you replace with tubes, the greater the cold surge, because the total resistance cold goes down. SO it seems to me that two tubes and two resistors would be easier on the tubes than it would be with two more tubes in place of the resistors. That is an opinion, I could be wrong.

5th July 2012, 09:20 PM	#17
darkskeptic diyAudio Member Join Date: Apr 2012	You are right, resistors start out at a resistance and stay there. heaters would start low resistance and normalize once warm. I'm just wondering how much extra heating time a ballast resistor would add to startup... Probably would take less time to heat up than using a current source but more time than if it was all heaters in series. For the time being, I'm just gonna pull the screen resistors for 2 of the tubes, and when I find a suitable current source to put in, I may go with that. The current source would allow me to simply jumper the heater pins in the sockets to remove half the tubes. If I need full power, no disassembly would be needed to go back.

5th July 2012, 10:23 PM	#18
darkskeptic diyAudio Member Join Date: Apr 2012	With 2 tubes out, will the bias for the remaining tubes stay within expected range? Also, the 16/8/4 ohm switch will need to be changed to half the actual speaker impedance right?

6th July 2012, 10:15 AM	#20
darkskeptic diyAudio Member Join Date: Apr 2012	All your help was awesome in this, and even the more heated discussions bring other knowledge and viewpoints out. What can and can't be done in audio is not always obvious to a beginner like myself. Ive read Morgan Jones' Valve Amplifiers book, and several other books on audio, but I learn just as much from this forum... The amp works and sounds great now at 50W, and in the future I may experiment further with it.