4 is pretty close to the cutest age, no doubt!
With a 6V heater winding, the EL84 heaters are all in parallel. The ECC81 heaters are 12V, but they have a center tap at pin 9. So they can also be driven by the 6V winding by connecting together pins 4 and 5 and attaching one end of the 6V winding to pins 4/5 and the other end to pin 1. You still need to bias up the winding by the voltage divider (the 200k/51k pair). If your 6V winding has a center tap (i.e. 3.15-0-3.15), that's the spot to do it. If it doesn't, you can create an artificial CT by connecting two 47R-100R in series, connect them across the winding, then attach the bias voltage to their junction.
With a 6V heater winding, the EL84 heaters are all in parallel. The ECC81 heaters are 12V, but they have a center tap at pin 9. So they can also be driven by the 6V winding by connecting together pins 4 and 5 and attaching one end of the 6V winding to pins 4/5 and the other end to pin 1. You still need to bias up the winding by the voltage divider (the 200k/51k pair). If your 6V winding has a center tap (i.e. 3.15-0-3.15), that's the spot to do it. If it doesn't, you can create an artificial CT by connecting two 47R-100R in series, connect them across the winding, then attach the bias voltage to their junction.
Thank you for a nice to read article and for sharing your interesting design!!
Quoted from http://syclotron.com/?page_id=33 :
Has there been interest? At least I am very interested, if one will be availiable.
I also have a couple of additional questions. Sorry if they have been asked before but this is a long thread...
1. For simplicity I would be tempted to use a lower voltage for the input tubes. If I remember correctly you have said that the higher voltage is needed for gain reasons. Would this in practice mean that only a part of the output power can be used? Could the issue be solved with a preamp with gain or would that overdrive the input stage?
2. Could a toroidal power transformer be a good candidate for OT? At least the DC currents need to be matched very well. Is separate screen supply regulators for each tube the best solution for accurate matching? Is there other possible solutions? Are there specific reasons why I shouldn't use a toroid as OT?
Quoted from http://syclotron.com/?page_id=33 :
Has there been interest? At least I am very interested, if one will be availiable.
I also have a couple of additional questions. Sorry if they have been asked before but this is a long thread...
1. For simplicity I would be tempted to use a lower voltage for the input tubes. If I remember correctly you have said that the higher voltage is needed for gain reasons. Would this in practice mean that only a part of the output power can be used? Could the issue be solved with a preamp with gain or would that overdrive the input stage?
2. Could a toroidal power transformer be a good candidate for OT? At least the DC currents need to be matched very well. Is separate screen supply regulators for each tube the best solution for accurate matching? Is there other possible solutions? Are there specific reasons why I shouldn't use a toroid as OT?
I haven't heard much interest in PCBs- I have been wanting the post the PCB stuff for the screen regulators but for a variety of bad excuses, haven't done so yet. I promise I will... I would love it if someone would do a PCB design for the input stage- it would easily be adaptable to other designs like the Bevois Valley or make a nice input stage for an ST-70 rebuild.
I am totally incapable of PCB design, so in the words of Blanche Dubois, "I have always relied on the kindness of strangers." If anyone wants to do it, I have no objections to them selling it, as long as they give me one or two to try out.
The higher voltage is needed for both gain and swing. If you run it lower, the distortion will increase since the first stage will be starved and the second will have insufficient cathode to plate voltage for linearity. But the current requirements are low, so it doesn't really add up to much expense.
I haven't tried toroids (though a local guy has offered me some), but there's nothing special about this design that would make it any more or any less friendly to toroids than any other push pull. I absolutely would use separate screen regulators in that case- toroids are very, very finicky about any imbalance in idle current.
I am totally incapable of PCB design, so in the words of Blanche Dubois, "I have always relied on the kindness of strangers." If anyone wants to do it, I have no objections to them selling it, as long as they give me one or two to try out.
The higher voltage is needed for both gain and swing. If you run it lower, the distortion will increase since the first stage will be starved and the second will have insufficient cathode to plate voltage for linearity. But the current requirements are low, so it doesn't really add up to much expense.
I haven't tried toroids (though a local guy has offered me some), but there's nothing special about this design that would make it any more or any less friendly to toroids than any other push pull. I absolutely would use separate screen regulators in that case- toroids are very, very finicky about any imbalance in idle current.
thomsva said:
If you use the antek 2T300 power toroid for power, use one winding with a solid state bridge for the driver and you'll get about 400V. Use the other winding with a hybrid EZ81 and solid state bridge for the output and you'll get about 325V.
These transformers cost $29 and have a huge VA rating, and plenty of heater power
SY said:
If this is split into smaller boards (one for the driver, one for the LEDs, one for regulators) it is pretty easy. A PS board is not too bad, either, maybe sans the EZ81. Not sure it is worth putting the output tubes on PCBs due to heat and whatnot. In fact, other than the LEDs perhaps, these would all be useful little projects on their own.
I could be talked in to doing them, but really only if I knew there were enough people that wanted them, and if someone else picked up the tab for ordering and assembling a round of prototypes (which would end up being under $50 plus parts for a single set. I can be emailed at 01312009@ecp.cc -- that email address will be disabled after Jan 31)
I just sent the artwork for the "Simple P-P" PC board off to the same PC board house that did the Simple SE. I will get a large stack of boards in 4 weeks. It will take another 2 or 3 weeks to shoot all of the pictures for a Simple SE style manual. The current financial climate has severely reduced PC board sales, which caused the delay until 2009. I couldn't put the $1K hit to the books in 2008.
The circuit is similar to the RLD or the Bevois Valley. The driver (12AT7) power supply (5AR4) and the output tubes (EL84) are all on the PC board.
I agree that output tubes on a PC board used to be a problem. With modern ceramic sockets on FR4 board material it is no nonger a problem. The Simple SE dissipates far more heat than this board, they have been in use for two years without issue. My personal unit runs EH KT88's at 40+ watts each. The only toasty thing in the amp is the poor overloaded power transformer, hot, but still running!
The Simple P-P was designed for the usual cathode resistors and bypass capacitors, and the PC board is laid out for those components. I made provisions for LED's on the PC board, but not a bunch of LED's. I have found that 3 or 4 jumbo blue or white 1/2 watt LED's in series on each cathode works great. The prototype board can be seen lighting up the whole room using white LED's from the flash of a camera phone in the following thread.
http://www.diyaudio.com/forums/showthread.php?s=&threadid=126291&highlight=
When I decided to do the Simple P-P, I knew that there would need to be two P-P designs. There is a "Simple" version, which was just sent out. There is also a "not so simple" version.
I have done a layout of a prototype PC board and started fabrication. I can't be sure what the final version will look like, but this one is a 4 by 5.5 inch board that contains the driver circuitry for a single channel. There are no power supply or output tubes on the board. Why? This board is intended for "universal" applications. Output tubes from the 6L6 types up to large transmitting tubes. Screen drive and cathode follower applications are supported as well. The power supply is off board since there are different requirements depending on output tube choice and circuit topology. A power supply board is being laid out now for some glowing sweep tube experiments.
The circuitry is not unlike the Crystal Palace driver, except the cathode followers are now mosfets and the bipolar CCS's are now 10M45's. 9 pin tubes are used but an octal version is also planned. The prototype has provisions for topology modifications which may be used in screen drive and cathode follower applications. Not all of these features will make it to the final version. Testing will progress over the next few weeks.
So do I. I might get another batch of the reg boards made- I used them in the ImPasse, too, and I have a project I want to do for an article which will also use this reg.
Problem has been the time to get my old computer set up, pull the files off, move them to this computer... I've been working 80 hour weeks so not much time for updating the website. Maybe this weekend, between work and the playoffs... (Go, Ravens!)
Problem has been the time to get my old computer set up, pull the files off, move them to this computer... I've been working 80 hour weeks so not much time for updating the website. Maybe this weekend, between work and the playoffs... (Go, Ravens!)
Blue Light District?
I have hundreds of blue, green, and UV LEDs from my old Xbox casemodding misadventures. They all seem to sit at around 3.2 ~3.5 volts. Since I already have these, would there be any way to utilize them in this design? That would mean I would only have to order the transformers for this project, as I already have everything else.
I have hundreds of blue, green, and UV LEDs from my old Xbox casemodding misadventures. They all seem to sit at around 3.2 ~3.5 volts. Since I already have these, would there be any way to utilize them in this design? That would mean I would only have to order the transformers for this project, as I already have everything else.
SY said:
For those who don't know what or whom a "demi-monde" is -- the aforementioned Blanche Dubois -- Vivien Leigh will make you sweat: http://www.youtube.com/watch?v=CxGN29njs3Q
Assuming that the schematic is the same as in Bas Horneman's DIYZINE, I will take a shot at a PCB for the input this afternoon and provide the gerbers to anyone who wants.
I'm partial to the LR8N3 from Supertex -- with a pass transistor like the MJE3439.
Jack, the current schematic for the input stage is the same. I made an update on the LED array (each LED string is six reds in series, with a small resistor in each string to verify current balance). I have a chassis layout for Front Panel Express- I'll send it to you tonight so that the physical layout of the board will drop into it.
Here's a cut at a PCB layout for the driver -- the LED arrays consist of a 6x6 matrix with a small current limiting/balancing resistor -- the plate/cathode resistors are 2W or 3W:
http://www.tech-diy.com/Amplifiers/RedLight/RDLT_3D.gif
http://www.tech-diy.com/Amplifiers/RedLight/RDLT_3D.gif