I was going to mention that, kind of defeats the purpose of twisting the others if the there is noise generated near the first tube in the filament chain.
I just wanted to say thank you for taking the time to put this together. Even though I already know most of the content, it is still a very nice refresher and you have such good technique to your work. Thanks Tony.
You can get away with chassis grounding here since you have a small one-piece stainless chassis. If you build any high gain audio circuit, chassis grounding can be very problematic. Randy Sloane talks about this in detail in his Power Amplifier book. Most chassis have multiple parts screwed together. Star grounding combined with board level ground bus lines is the preferred method. I got serious hum issues by re-using an old (multi part) chassis with a built-in terminal block as the main grounding point.
10:28 I have to love how, 2 "Other" countries are either still manufacturing (or have a working stockpile of) Tubes/Valves/"Vacuum Bulbs 😁", whereas no company in the US is finding these items in demand enough to keep making them. Also, at 15:50, I wonder, if anyone has experimented with reversing the "phase" of a heater string at certain points in audio equipment, or some tube equipped TV's, to see if it can increase or decrease any EMF induced hum. I don't believe anyone has proven, or disproven, (in a UA-cam video anyways) that doing so will have any effect. The argument might be made that, reversing the phase on 2 different tubes might help cancel out 60 cycle hum. I might have to get my hands on an old tube amplifier and play with it. BTW, very nice attention to detail in the twisted wire layout of the filament connections! Enjoyable view of the build, glad you are sharing it with us.
I can't help asking you why when you illustrate using twised heater wires you don't apply the same to the two wires coming from the transformer. They will radiate just as effectively as any ac wire and are right there by the components going to the B9A base.
Check twice solder once: When the 12-pin 12AE10 tube first came out it was twice or three times the price of a regular tube. Dad gave me the money to purchase one but warned me that my project better work. I wired the base wrong at first, but luckily didn't burn the tube out.
Hi Tony, Thank you for this series! I'm working my way through it (I did just quickly watch the one about the power transformer). I purchased a similar type pre-amp (pre assembled in China, purchased from ebay) and used a Variac to turn the 125VAC outlet voltage down to 110V to match the transformer rating (I was worried 110V meant just that) and it brought the filament voltage right down to 6.3V. Not sure it was necessary but I was freaking out about being out of spec and killing the filaments (lots of drama on line about this). Hopefully me doing this is not throwing off some other necessary voltage, guess I'll have to re-investigate (if I can find the schematics).
This "not shown" issue is also true with schematics that use Op-amps. Most of the time they DON'T show the PSU connections. They just numerically label the chip pins on the schematic. Proper power bus wiring is typically left as an "exercise to the builder".
I do know that there is an optimal twist per inch, but I don't remember what it is. This would be the angle at which the wires cross which varies by wire size. There is a Goldilocks zone for optimum noise rejection.
I use a terminal strip and run the filament wires from the transformer to the strip then run wires from the strip to each tube. I always try to keep all power “voltage wires away from signal connections.
I have built and repaired many amps, and heater wiring, if it is not for example floating freely without any connection to ground etc., has very seldom been causing any major hum problems. Plate supply ripple, and power supply filter cap ripple currents in unwisely done ground wiring, and signal ground loops usually cause all major hum problems, not heater wiring. BTW that lamp was still before the fuse in the schematic. Just nagging. 🤪
Excellent video! Thank you. You have probably said before, so forgive an old folgey's forgetfulness, but what temp do you set you soldering iron for 60/40 or the Radio Shack silver solder? 73, AA4EZ
You didn't mention the method I have used to locate the heater pins - ohming the tube out. On a cold tube there should be open circuit between any two pins (unless they are internally connected) except for the heater pins which will measure the cold resistance of the heater.
@@srtamplification Since the heater has a non zero resistance, you should be able to tell the difference between heater pins and shorted pins. (Actually, finding shorted pins with an ohmmeter sounds like a bonus).
Since you have the pcb you could use the VOM to find the continuity between the transformer secondary and the tube socket footprint to find the heater pins. Assuming the heater voltages aren't used elsewhere in the circiut.
That's true. And it works fine for example in old Fender amps but I prefer to use noval sockets equipped with center shield and use it as a ground pole for grid leak resistors etc. In that case it is much easier to lay heater wiring to go near chassis surface. To be honest, it seems not to matter much. I have lately built my amps using a very strict and thoroughly thought star grounding and those amps are totally humfree (without any need to use for example DC filament supply) causing a need to reduce tube and resistor hiss in every possible way instead 😆, because of that good grounding, and how heater wiring is located doesn't really make much difference. Star grounding needs separate ground wires to every amp stage which very easily looks like a mess compared with neat looking traditional ground rails but it IMO gives better result.
WOW! My dream series. A detailed Xraytonyb amp building series. Thank you sooo much for sharing your knowledge. At the risk of being that guy who asks the stupid question, what does JABOP stand for?
After I wired the heater string, I tested for continuity. I expected no continuity but measured that pins 4 & 5 were connected to each other. Did I make a wiring mistake (I don't see it)? Will I burn up the tubes?
It depends on the type of solder and type of iron I am using. For lead solder (60/40 type) start with about 600-650F (316-343C) and for lead-free 650-700F (343-371C). If your soldering iron is temperature controlled and has really fast thermal recovery, you can set it a bit lower. If the iron doesn't have fast recovery or good thermal mass, you may need to set it a bit higher. The idea is the get the solder to flow quickly and not dwell for a long time on the connection, so as not to overheat the component or the solder trace. Overheating the trace by dwelling on it too long or having too much heat could damage/lift the trace. You really have to experiment a little bit with your iron to get the correct "Goldie Locks" setting ;)
@@xraytonyb I have the Hakko FX-888D, I have it set to 680F, but it seems to take a little longer then yours to heat things up. Maybe it's because my tip is too small, I'm using a real pointy one. I have other tips, I'll have to experiment some. Thanks!.
Howdy. Neat. One thing I have never seen. According to The Electrical Code. All wiring should have overcurrent protection. The filament cascode should have fuses in both poles at the sourcing point. Why ? Are the power transformer primary fuses not enough ? No. The primary fuses will activate only to a short or partial short somewhere is the high voltage circuits. A partial short in the filament cicuits may increase the filament current to a level of melting the filament wiring insulations, without activating the primary fuses. Alternatively a separate filament transformer may be used. This transformer's primary fuses should be rated only large enough allow the higher start up currents of the filaments. Regards.
The problem with fusing filaments -- and it is causing chronic reliability problems in amps where it is used -- is that when cold, the tube filaments appear briefly as an almost direct short to the filament winding, so every time you turn the filaments on, there is a brief, but high, current surge. These turn-on surges that are far in excess of steady-state filament current eventually cause the fuses to blow even if there is no fault, leaving the amplifier disabled. So imagine if you are setting up for a show and you turn your guitar amp on, only to have the filament fuse blow for no particular reason. And this is not an imaginary scenario. It happens regularly in amps with fused filament secondaries.
@@southerner66 Howdy again. No intention to be a smart one. It looks to me that the filament wire is AWG 17 about 1 mm2 in European. AWG 16 is about 1,3 mm2 and AWG 18 about 0,8 mm2. If filament fuses are used they need not be rated according to the filament current running warm. They may rated according to the wire gage. AWG18 can carry about 7 A, AWG17 about 8,5 A and AWG 16 about 10 A for some period of time. Not indefinitely. The fuses may be rated one step smaller, slow. I should think this will allow for the start up rush of the filaments. In regards.
The center tap doesn't go to the filaments. It is connected to chassis ground to reduce hum by holding the filament near ground potential. The two ends of the heater winding go to the two ends of the filament at each tube, providing 6.3 volts.
Thanks so much for this video series
I watch them long or short , I learn a lot ! Thanks keep up the good work!
You are an excellent teacher. Thanks!!!
You can also twist the filament leads between the transformer and the first tube to increase noise rejection.
I was going to mention that, kind of defeats the purpose of twisting the others if the there is noise generated near the first tube in the filament chain.
I just wanted to say thank you for taking the time to put this together. Even though I already know most of the content, it is still a very nice refresher and you have such good technique to your work. Thanks Tony.
i always tin wires too. it really makes things alot easier
You can get away with chassis grounding here since you have a small one-piece stainless chassis. If you build any high gain audio circuit, chassis grounding can be very problematic. Randy Sloane talks about this in detail in his Power Amplifier book. Most chassis have multiple parts screwed together. Star grounding combined with board level ground bus lines is the preferred method. I got serious hum issues by re-using an old (multi part) chassis with a built-in terminal block as the main grounding point.
10:28 I have to love how, 2 "Other" countries are either still manufacturing (or have a working stockpile of) Tubes/Valves/"Vacuum Bulbs 😁", whereas no company in the US is finding these items in demand enough to keep making them. Also, at 15:50, I wonder, if anyone has experimented with reversing the "phase" of a heater string at certain points in
audio equipment, or some tube equipped TV's, to see if it can increase or decrease any EMF induced hum. I don't believe anyone has proven, or disproven, (in a UA-cam video anyways) that doing so will have any effect. The argument might be made that, reversing the phase on 2 different tubes might help cancel out 60 cycle hum. I might have to get my hands on an old tube amplifier and play with it. BTW, very nice attention to detail in the twisted wire layout of the filament connections! Enjoyable view of the build, glad you are sharing it with us.
Hang in there buddy, going good. Thanks for the look.
I can't help asking you why when you illustrate using twised heater wires you don't apply the same to the two wires coming from the transformer. They will radiate just as effectively as any ac wire and are right there by the components going to the B9A base.
Check twice solder once:
When the 12-pin 12AE10 tube first came out it was twice or three times the price of a regular tube. Dad gave me the money to purchase one but warned me that my project
better work. I wired the base wrong at first, but luckily didn't burn the tube out.
Thanks so much Tony.
Xraytonyb wiring the filament circuit this cool
Hi Tony, Thank you for this series! I'm working my way through it (I did just quickly watch the one about the power transformer). I purchased a similar type pre-amp (pre assembled in China, purchased from ebay) and used a Variac to turn the 125VAC outlet voltage down to 110V to match the transformer rating (I was worried 110V meant just that) and it brought the filament voltage right down to 6.3V. Not sure it was necessary but I was freaking out about being out of spec and killing the filaments (lots of drama on line about this). Hopefully me doing this is not throwing off some other necessary voltage, guess I'll have to re-investigate (if I can find the schematics).
most kits dont give you a heater lay out and how to wire your heaters
This "not shown" issue is also true with schematics that use Op-amps. Most of the time they DON'T show the PSU connections. They just numerically label the chip pins on the schematic. Proper power bus wiring is typically left as an "exercise to the builder".
And discrete logic gates.
I do know that there is an optimal twist per inch, but I don't remember what it is. This would be the angle at which the wires cross which varies by wire size. There is a Goldilocks zone for optimum noise rejection.
there is another method of finding the filament and that's to measure the pins for dc resistance.
21:00. Hakko is the expensive version of the $30 hand held iron. Metcal is your basic soldering iron for this century. The Hot Rod is JBC.
I use a terminal strip and run the filament wires from the transformer to the strip then run wires from the strip to each tube. I always try to keep all power “voltage wires away from signal connections.
An ohmmeter can be used to locate the heater too
I have built and repaired many amps, and heater wiring, if it is not for example floating freely without any connection to ground etc., has very seldom been causing any major hum problems. Plate supply ripple, and power supply filter cap ripple currents in unwisely done ground wiring, and signal ground loops usually cause all major hum problems, not heater wiring.
BTW that lamp was still before the fuse in the schematic. Just nagging. 🤪
Excellent video! Thank you. You have probably said before, so forgive an old folgey's forgetfulness, but what temp do you set you soldering iron for 60/40 or the Radio Shack silver solder? 73, AA4EZ
You didn't mention the method I have used to locate the heater pins - ohming the tube out. On a cold tube there should be open circuit between any two pins (unless they are internally connected) except for the heater pins which will measure the cold resistance of the heater.
Yes this works as long as you don't have a bad tube with shorted elements.
@@srtamplification Since the heater has a non zero resistance, you should be able to tell the difference between heater pins and shorted pins. (Actually, finding shorted pins with an ohmmeter sounds like a bonus).
Since you have the pcb you could use the VOM to find the continuity between the transformer secondary and the tube socket footprint to find the heater pins. Assuming the heater voltages aren't used elsewhere in the circiut.
Would you ever add a grounded shield to the twisted AC filament wires to help reduce noise?
21:46 - Interesting. Uncle Doug recommends keeping the filament wiring AWAY from the chassis!
That's true. And it works fine for example in old Fender amps but I prefer to use noval sockets equipped with center shield and use it as a ground pole for grid leak resistors etc. In that case it is much easier to lay heater wiring to go near chassis surface. To be honest, it seems not to matter much. I have lately built my amps using a very strict and thoroughly thought star grounding and those amps are totally humfree (without any need to use for example DC filament supply) causing a need to reduce tube and resistor hiss in every possible way instead 😆, because of that good grounding, and how heater wiring is located doesn't really make much difference.
Star grounding needs separate ground wires to every amp stage which very easily looks like a mess compared with neat looking traditional ground rails but it IMO gives better result.
WOW! My dream series. A detailed Xraytonyb amp building series. Thank you sooo much for sharing your knowledge. At the risk of being that guy who asks the stupid question, what does JABOP stand for?
Just a box of parts
@@xraytonyb Thank you.
15:04 The terms is called “bird caging”.
After I wired the heater string, I tested for continuity. I expected no continuity but measured that pins 4 & 5 were connected to each other. Did I make a wiring mistake (I don't see it)? Will I burn up the tubes?
You should read low resistance from pins 4 to 5 because you are reading across the filaments in parallel with the filament transformer winding.
You didn't wire it with 1930's cloth insulated wire. It won't have the voodoo magic lol
filament notching
Hi Tony, I was wondering what you set your soldering iron temperature to?
It depends on the type of solder and type of iron I am using. For lead solder (60/40 type) start with about 600-650F (316-343C) and for lead-free 650-700F (343-371C). If your soldering iron is temperature controlled and has really fast thermal recovery, you can set it a bit lower. If the iron doesn't have fast recovery or good thermal mass, you may need to set it a bit higher. The idea is the get the solder to flow quickly and not dwell for a long time on the connection, so as not to overheat the component or the solder trace. Overheating the trace by dwelling on it too long or having too much heat could damage/lift the trace. You really have to experiment a little bit with your iron to get the correct "Goldie Locks" setting ;)
@@xraytonyb I have the Hakko FX-888D, I have it set to 680F, but it seems to take a little longer then yours to heat things up. Maybe it's because my tip is too small, I'm using a real pointy one. I have other tips, I'll have to experiment some. Thanks!.
Howdy. Neat.
One thing I have never seen. According to The Electrical Code. All wiring should have overcurrent protection. The filament cascode should have fuses in both poles at the sourcing point. Why ? Are the power transformer primary fuses not enough ?
No. The primary fuses will activate only to a short or partial short somewhere is the high voltage circuits. A partial short in the filament cicuits may increase the filament current to a level of melting the filament wiring insulations, without activating the primary fuses.
Alternatively a separate filament transformer may be used. This transformer's primary fuses should be rated only large enough allow the higher start up currents of the filaments.
Regards.
The problem with fusing filaments -- and it is causing chronic reliability problems in amps where it is used -- is that when cold, the tube filaments appear briefly as an almost direct short to the filament winding, so every time you turn the filaments on, there is a brief, but high, current surge. These turn-on surges that are far in excess of steady-state filament current eventually cause the fuses to blow even if there is no fault, leaving the amplifier disabled. So imagine if you are setting up for a show and you turn your guitar amp on, only to have the filament fuse blow for no particular reason. And this is not an imaginary scenario. It happens regularly in amps with fused filament secondaries.
@@southerner66 Howdy.
Yes. You are 101% correct.
Regards.
@@southerner66 Howdy again. No intention to be a smart one.
It looks to me that the filament wire is AWG 17 about 1 mm2 in European.
AWG 16 is about 1,3 mm2 and AWG 18 about 0,8 mm2.
If filament fuses are used they need not be rated according to the filament current running warm. They may rated according to the wire gage.
AWG18 can carry about 7 A, AWG17 about 8,5 A and AWG 16 about 10 A for some period of time. Not indefinitely. The fuses may be rated one step smaller, slow. I should think this will allow for the start up rush of the filaments.
In regards.
Me gustaría tener ese esquema amplificador a valvulas
each pin gets 6.3 volts? Your transformer is center tapped at 3.15volts. So the filament gets 3.15v per pin?
The center tap doesn't go to the filaments. It is connected to chassis ground to reduce hum by holding the filament near ground potential. The two ends of the heater winding go to the two ends of the filament at each tube, providing 6.3 volts.
@@xraytonyb Thanks... I dived deeper into it, lol
What gauge wire are you using?
Use 18 or 20 gauge for filament connections.