i always use 12AX7EH, i think theyre pretty good, ive never really used 6v6, only EL84, as for output impedance of transformers, it should be stated on the datasheet, the cathode resistor is irrelevant, if not i think there is a way of working it out that i think was mentioned on the valve wizards page. are you still thinking about voltages and gain stages? ive had a few ideas i was planning to do eventually, but ive just run out all my funds after building a high gain 2 channel 50watt amp... and IMO i think itd be better to use an SS rectifier to start with, alot easier, and takes up less space
I am, but I have a rough idea and I figure i can try to work that stuff out when i get testable prototype. I am definitely open to suggestions.
I am having trouble finding a suitable power trafo at edcor. I figure i want at least 300V and enough current to run 6 12AX7's, and 2 6v6's (roughly 200-250 ish mA...i think) and the heater needs to be 6.3V at atleast 2.7mA (more if i go with tube rectification). I think for right now I'll go with silicon rectifiers, but i might and probably will want to try out tubes at some point just to see the difference, and I dont want to have to buy a new trafo at that point. i think this is the closest i can find, but it will count out tube rectification for the future...so i dont know if I want that. Is there any other supplier that may carry more selection/does edcor or another supplier allow you to give them your own required specs for a reasonable price?
agghhh.. the death cap
If you do you the schematic you posted earlier, make sure to omit the .05uF 600 v cap that is located between the PT primary and the chassis ground. Use a three wire power cord instead and setup a proper star ground on your chassis. This cap is often referred to as "the death cap", they were used in the days before grounded outlets, when they get old and leaky, you could wind up with line voltage on your chassis, which would then be present on your strings. If you touched anything else that was grounded, like a microphone, things would get a little scary.
If you do you the schematic you posted earlier, make sure to omit the .05uF 600 v cap that is located between the PT primary and the chassis ground. Use a three wire power cord instead and setup a proper star ground on your chassis. This cap is often referred to as "the death cap", they were used in the days before grounded outlets, when they get old and leaky, you could wind up with line voltage on your chassis, which would then be present on your strings. If you touched anything else that was grounded, like a microphone, things would get a little scary.
300v sounds about right for PP 6v6, im not sure how much current that will need, but i know im running 2 EL34's and 5 12ax7s at 250mA, but why 6 12AX7s? even high gain guitar amps with PI and valve effects loops generally use 5? id say go with 3, 2 for preamp, and a PI or 4 if you want an effect loop. the 12AX7s i wouldnt run on anything less than 200v. (BTW i think 2.7A is a better current for a heater winding) and bear in mind if you want something with more power in future, with that transformer, 2 EL34s would take up all of that heater current. (and with that transformer your probably going to get an HT of about 220 volts unles you dont use the CT in which case you would get about 470 ish volts...)
sorry for the confusion, I'm using 3 (with 2 sections each)...the datasheet says the anode current is around 9 mA for this tube, so i took that to mean i would be using 9*6 = 54mA for the 12AX7's and the only number i could find for the 6v6's was "between 72 and 80 mA" so taking the safe bet, that gives me about 214 mA correct?
also, about the "death cap" I was planning on using a 3-prong power cord and not using that cap (i am very aware of issues that could go wrong resulting in me getting fried). I am also going to use a different rectifier tube and get rid of the 5V supply they have, so the schematic I posted wont be copied EXACTLY...but it will be very close. Thank you guys for the info! now to figure out this transformer issue...
12AX7 with 9mA going thru it is going to have a lifetime of minutes with any real life voltages accross it!
In most cases the 12AX7 uses up say 0.8 ... 1.5 mA of current. At most 2mA. The highest I've seen in a guitar amp was 1.8mA on a cathode follower.
6V6's sound just fine with currents over 15mA. Maybe 30 ... 35 mA per tube would be my optimum.
In most cases the 12AX7 uses up say 0.8 ... 1.5 mA of current. At most 2mA. The highest I've seen in a guitar amp was 1.8mA on a cathode follower.
6V6's sound just fine with currents over 15mA. Maybe 30 ... 35 mA per tube would be my optimum.
ok good!...i dont remember where i got that info, but its possible i confused it with something else, ive been looking at a lot of voltage and current values for different tubes over the past few days. in that case, do you think one of these transformers would suit my purposes for a plexi 6v6?
EDCOR Electronics Corporation. XPWR147-120/240
EDCOR Electronics Corporation. XPWR164-120
I realize the voltages are a little higher than i will need, but i figure i can attenuate them as needed. I think im looking at the 164-120. It has all the heater voltages necessary with enough current. the real benefit to the other one is that 12V that i could pretty easily bring down to 9V for other effects if i wanted. what do you think?
by artificial center tap, do you mean 2 equal value resistors from either side of the winding to ground? This would guarentee that I cant run a tube rectifier off of this winding, so i would use the 12V one to get that 6.3 right?
I would start with a champ design.....
Lynchy: As you've probably already figured out, the champ and the plexi schemo's are quite different, and will require different power and output iron.
The champ is a simple tube rectified cathode bias single ended design, and the plexi is a push pull SS rectified fixed bias design.
So, the champ requires a single ended output transformer and the plexi requires a PP OT. The power transformer requirements are different as well due to the tube count and SS vs. tube rectification.
When specifying a power transformer for a capacitor input design (CRC or CLC topology) the resulting B+ will be sqrt2*transformer secondary V less any series resistance drops for each B+ stage. For SS rectification, your B+ will be a bit less than sqrt2*transformer secondary V due to the series resistance of the R's and L's you use for smoothing the AC to DC. For tube rectification, the rectifier tube's voltage drop will drop additional volts in addition to any series R. For FWCT rectification SS diodes drop less than a volt and tube rectifiers can drop 20-70 volts or more.
NOS 5Y3's are a dime a dozen and easy to get so I wouldn't worry about spec'ing a 6V rectifier tube. Edcor's power transformer offerings should have choices with both 5V and 6V windings for this type design. There are lots of other online vendors that sell power transformers like Angela Instruments, Triode electronics, Antique Electronics supply, etc for guitar amps. I imagine some have "champ" clone transformers......
Also remember that a tube rectified power supply will have higher output impedance than SS diodes which will result in more B+ sag (a good thing for guitar amps/distortion and a bad thing for hi-fi amps) than a SS rectified PS. For a given B+, tube rectified PS designs require a higher secondary voltage transformer due to the tube rectifier's voltage drop.
PSUDII is your friend here, as you can model your power supply to determine the resulting B+ voltages. You can easily switch between SS and tube rectified FWCT in PSUD with a given transformer to see the changes in B+ voltages.
Finally, if you are looking for tube distortion, resist the temptation to "over-specify" the power transformer, as the distortion in amps like the champ, etc was partly due to the fact that Leo Fender was very frugal, and the original parts spec'd were just enough to get the job done, and that's where some of the tone magic comes from.
Lynchy: As you've probably already figured out, the champ and the plexi schemo's are quite different, and will require different power and output iron.
The champ is a simple tube rectified cathode bias single ended design, and the plexi is a push pull SS rectified fixed bias design.
So, the champ requires a single ended output transformer and the plexi requires a PP OT. The power transformer requirements are different as well due to the tube count and SS vs. tube rectification.
When specifying a power transformer for a capacitor input design (CRC or CLC topology) the resulting B+ will be sqrt2*transformer secondary V less any series resistance drops for each B+ stage. For SS rectification, your B+ will be a bit less than sqrt2*transformer secondary V due to the series resistance of the R's and L's you use for smoothing the AC to DC. For tube rectification, the rectifier tube's voltage drop will drop additional volts in addition to any series R. For FWCT rectification SS diodes drop less than a volt and tube rectifiers can drop 20-70 volts or more.
NOS 5Y3's are a dime a dozen and easy to get so I wouldn't worry about spec'ing a 6V rectifier tube. Edcor's power transformer offerings should have choices with both 5V and 6V windings for this type design. There are lots of other online vendors that sell power transformers like Angela Instruments, Triode electronics, Antique Electronics supply, etc for guitar amps. I imagine some have "champ" clone transformers......
Also remember that a tube rectified power supply will have higher output impedance than SS diodes which will result in more B+ sag (a good thing for guitar amps/distortion and a bad thing for hi-fi amps) than a SS rectified PS. For a given B+, tube rectified PS designs require a higher secondary voltage transformer due to the tube rectifier's voltage drop.
PSUDII is your friend here, as you can model your power supply to determine the resulting B+ voltages. You can easily switch between SS and tube rectified FWCT in PSUD with a given transformer to see the changes in B+ voltages.
Finally, if you are looking for tube distortion, resist the temptation to "over-specify" the power transformer, as the distortion in amps like the champ, etc was partly due to the fact that Leo Fender was very frugal, and the original parts spec'd were just enough to get the job done, and that's where some of the tone magic comes from.
Thank you boywonder, I was actually able to find a fitting power trafo for the champ pretty easily, but the one with 5v that i found was about $10 more than one with just a 6.3 at a higher current, so i figured i would go with that. I didnt really look into the price of the rectifiers themselves, I'll look more into that aspect tonight.
I am buying parts for 2 different amps here, I feel like i can get a good champ trafo without too much issue, and I will check pricings from other suppliers tonight. My real issue is finding a good power trafo for the plexi-ish.
also, If i used diode rectification now with a higher trafo voltage and some sort of attenuator (to lower the B+ before it gets to the amp) would I be able try out a tube rectifier on the same amp later? (I assume PSUDII will help a lot with this) A huge part of this project is to give myself the chance to play around with things and HEAR the differences, rather than just read about what other people say online about them, so I would like to spec this amp to work either way (if possible). Thank you for all your help!
I am buying parts for 2 different amps here, I feel like i can get a good champ trafo without too much issue, and I will check pricings from other suppliers tonight. My real issue is finding a good power trafo for the plexi-ish.
also, If i used diode rectification now with a higher trafo voltage and some sort of attenuator (to lower the B+ before it gets to the amp) would I be able try out a tube rectifier on the same amp later? (I assume PSUDII will help a lot with this) A huge part of this project is to give myself the chance to play around with things and HEAR the differences, rather than just read about what other people say online about them, so I would like to spec this amp to work either way (if possible). Thank you for all your help!
You can drop excessive power transformer volts with series resistance (chokes and/or resistors), however sometimes dropping lots of volts particularly at high currents results in lots of heat watts generated and giant power resistors. Do some simple ohms law calcs and see.
Chokes can be more effective at smoothing AC than resistors but cost more and take up more space but they can also handle a bit of power dissipation just because of their size. Do a PSUDII model with a CRC (cap-resistor-cap) and CLC (cap-choke-cap) supply and compare B+ volts, ripple V, etc.
Another way of reducing B+ volts without having to use big honkin' power resistors is to use a CLC supply and make the value of the first cap small, say less than 10uf down to less than 1uf. Small first cap values mimic a choke input supply topology (LC), which provides about 90% of the transformer voltage as a B+ voltage as opposed to a cap input supply at 1.41*transformer voltage. As the first cap gets smaller, say below about 10uf or so, the supply begins to mimic an LC supply and the voltage drops.
So think of it this way; there is a spectrum of B+ voltages available from .9*transformer voltage to 1.41*transformer voltage depending on the value of the first cap.
Chokes can be more effective at smoothing AC than resistors but cost more and take up more space but they can also handle a bit of power dissipation just because of their size. Do a PSUDII model with a CRC (cap-resistor-cap) and CLC (cap-choke-cap) supply and compare B+ volts, ripple V, etc.
Another way of reducing B+ volts without having to use big honkin' power resistors is to use a CLC supply and make the value of the first cap small, say less than 10uf down to less than 1uf. Small first cap values mimic a choke input supply topology (LC), which provides about 90% of the transformer voltage as a B+ voltage as opposed to a cap input supply at 1.41*transformer voltage. As the first cap gets smaller, say below about 10uf or so, the supply begins to mimic an LC supply and the voltage drops.
So think of it this way; there is a spectrum of B+ voltages available from .9*transformer voltage to 1.41*transformer voltage depending on the value of the first cap.
thanks boywonder, I will see if i can figure out a good way to make this work. I'll try to keep all of this in mind. Ill run a few simulations tonight and see what i can get.
razorrick, you mentioned that you wouldnt run a 12AX7 on less than 200v, but a schematic of the fender Champ 5E1, which i found seems to be running them on 150v...is this bad design on their parts? or potentially an attempt to coerce more distortion from the tubes with less volume?
razorrick, you mentioned that you wouldnt run a 12AX7 on less than 200v, but a schematic of the fender Champ 5E1, which i found seems to be running them on 150v...is this bad design on their parts? or potentially an attempt to coerce more distortion from the tubes with less volume?
There's a big difference between n volts on the plate of 12AX7 or n volts supply for 12AX7. The latter is before the plate load resistor, and the first is after. These get mixed up quite a lot (at least in my head when reading stuff on the forums).
Generally as you said, more voltage means more clarity, possibly more chime, less voltage means a bit more muddy and dirty. It's a matter of taste. Anything over a 100 volts and under 250 on the plates is good. Generally the 150 volt range seems to be common.
Generally as you said, more voltage means more clarity, possibly more chime, less voltage means a bit more muddy and dirty. It's a matter of taste. Anything over a 100 volts and under 250 on the plates is good. Generally the 150 volt range seems to be common.
Yes, for a two wire filament winding, you can make a virtual filament center tap by tying each 6.3V leg to ground through a 100 ohm resistor. For additional noise reduction you can attach the center tap to an elevated voltage reference to reduce 60hz filament noise. The heater reference is usually around 50-75V or so. A simple resistive voltage divider from B+ to ground of say 220K and 100K R's does the trick. This can also serve as a cap bleeder, which is highly recommended for safety.
Elevating filament CT reduces noise by reverse biasing the psuedo-diode between the cathode and the filament in the tube.
ok, just to clarify,if i had a 6.3v winding at 3A and all of the heaters in my amp have a nominal total current value of 2.5A, will it deliver 3A without a pseudo center tap? If this is the case, I should choose a resistor value that will eat up that extra .5A correct?, if this is not correct, my understanding of transformers seems to be a little off...Also, should i measure the heater reference voltage when i have the thing hooked up and choose resistor values that way? does it matter? or is the heater just floating, meaning i should try to set the heater reference myself, I'm sorry my understanding of transformers is limited to the theory behind them, not how they actually work in production.
Thank you for the explanation McCurwen, that makes what you said make a lot more sense now. in a pre-built circuit like this, I can figure out almost exactly what I can expect the anode current to be, assuming the grid current is very small. which it usually is, right?
Thank you for the explanation McCurwen, that makes what you said make a lot more sense now. in a pre-built circuit like this, I can figure out almost exactly what I can expect the anode current to be, assuming the grid current is very small. which it usually is, right?
You don't need to eat up any extra heater current. You just connect all the heaters parallel to the heater supply and they take the current they need. Any extra current capability on the transformer simply means it won't be running as hot as maxed out. Generally it's good to have some current headroom just to prevent excessive heat.
You don't need to take possible grid current into account, not at all.
You don't need to take possible grid current into account, not at all.
The pseudo center tap on the heater supply is simply to provide a ground reference for the heaters, thus preventing hum. It has no other purpose and it won't affect the current capability of the heater supply. It's a good thing to do, and even better is to do what boywonder described and elevate the pseudo center tap to some positive voltage.
You can use for example Voltage Divider Calculator to calculate what resistor values to use. Add a 22µF cap from the middle point (where you connect the heater pseudo center tap) to ground. Use such resistors, that the total resistance from B+ to ground on this divider is at least 100k.
You can use for example Voltage Divider Calculator to calculate what resistor values to use. Add a 22µF cap from the middle point (where you connect the heater pseudo center tap) to ground. Use such resistors, that the total resistance from B+ to ground on this divider is at least 100k.
If your filament winding is rated for 3A and you are drawing 2.5A, the voltage may be slightly high since you are not fully loading the transformer to it's rated current (depends on the regulation of the transformer, ie unloaded voltage is higher than rated voltage). Open circuit/unloaded the transformer may read over 7V......
Tube filaments are typically voltage rated at +/- 5% so anything between about 6 and 6.6 volts or so is OK. If you want, after wiring up the heaters, if the voltage is a little high you can add series R's to each leg of the transformer to reduce it a bit. They usually end up being less than 1 ohm and rated at a couple of watts.....use ohm's law to calculate both R and power rating.
As MrCurwen mentioned above, if you want to elevate the filament CT, you want the total R of the B+ voltage divider to be high (>100K ohms) so that you are not burning much power through those R's. The CT does not draw any current, it's just a voltage reference. 1ma through the voltage divider is fine. It will also act as a bleeder for the PS caps (VERY IMPORTANT FOR SAFETY), and larger resistance will take longer to bleed the caps down and smaller R will bleed down faster. 100K-300K total would work fine.
Tube filaments are typically voltage rated at +/- 5% so anything between about 6 and 6.6 volts or so is OK. If you want, after wiring up the heaters, if the voltage is a little high you can add series R's to each leg of the transformer to reduce it a bit. They usually end up being less than 1 ohm and rated at a couple of watts.....use ohm's law to calculate both R and power rating.
As MrCurwen mentioned above, if you want to elevate the filament CT, you want the total R of the B+ voltage divider to be high (>100K ohms) so that you are not burning much power through those R's. The CT does not draw any current, it's just a voltage reference. 1ma through the voltage divider is fine. It will also act as a bleeder for the PS caps (VERY IMPORTANT FOR SAFETY), and larger resistance will take longer to bleed the caps down and smaller R will bleed down faster. 100K-300K total would work fine.
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.