I was out of town for 10 days, which put me 10 days behind at work. With luck I might get a 3 day weekend. Project numero uno is my dual SPP amp. This is what I know now:
Wiring a single board for PPP mono is rather simple and I'll post the details as soon as I can take some pictures. This will get you a single channel at twice the power. This works for the EL84 version running at about 340 to 350 volts with a 3300 ohm OPT, and should also work on 6CW5's but a 1500 ohm OPT is needed which I don't have, so I haven't tried it. About 30 watts per board seems to be the limit in this configuration. Most EL84 types work fine at this level.
Of course I an not satisfied with a mere 30 WPC, and I have seen nearly twice that. After considerable experimentation it has been determined that some additional circuitry is needed to allow operation at these extremes without several minor annoyances like instability, total meltdown, and excessive distortion.
Operation at these extreme power levels requires running the plates at 435 volts and the screens at 325 volts. A screen regulator board is required to keep things controlled. I have a clip lead prototype, but some more development is needed. Obviously wimpy tubes will not live long at these voltages. True 7189A's, some NOS 6BQ5's and JJ EL84's seem to deal with it OK.
SY discovered that replacing the cathode resistors with an array of LED's can lead to increased power and better sound. I installed some white LED's in a prototype board last year and found this to be true. Further investigation revealed that doing this is essentially making the cathode biased amp operate as if it has fixed bias. The problem with LED's is that you have to start with a large batch and select a set that are relatively matched AND drop the right voltage for your particular tube. This will require a lot of LED's and the amp will light up the whole house. We need something better. Ideally it should have a very low effective impedance across the audio band, and be adjustable. I have a simple prototype circuit, and so do a few others. These can be seen at the end of SY's RLD thread.
Of course both of these ideas will be applicable to the standard Simple P-P board for those who want to improve the performance, or crank up the power.
So in my never ending quest to find a more suitable amp for some speakers I want, I've come full circle back to Tubelab (first amp was the Simple SE) and am thinking about the simple P-P. Maybe the Simple P-P setup as monoblocks will do it for me? The speakers are only 88 db efficient some I'm thinking 30-35 wpc is a minimum. Any progress on finalizing the design for a monoblock setup with a little more juice???
I would love to see the Simple P-P setup to run something bigger (6L6's, EL34's, or KT-88's) and I'd love to run those in monoblocks to get something like 60 wpc but that's no longer a Simple P-P... I know it's basically the 6L6GC AB2 amp worked on elsewhere but my skills are not up to the task of doing a scratch build like that myself.
I'll continue to dream...
I would love to see the Simple P-P setup to run something bigger (6L6's, EL34's, or KT-88's) and I'd love to run those in monoblocks to get something like 60 wpc but that's no longer a Simple P-P... I know it's basically the 6L6GC AB2 amp worked on elsewhere but my skills are not up to the task of doing a scratch build like that myself.
I'll continue to dream...
I'd be surpised if a 30W monoblock Simple PP is not enough power. Another possibility is convert a simple SE to PP, George has posted about this somewhere on the forum. But the difference between 30W and 50W is not huge. FWIW, I'm running 83dB full range speakers and a standard Simple PP has more than enough power, it just depends on how loud and how much bass content you like in your music.
Ian.
Ian.
Could someone point me to an actual schematic showing the different modes pentode, ul, etc, for the Simple Push-Pull. I've got it all stuffed up, no problem, iron is all hooked up but that open circuit on the L_FB-1 and -2 is whipping me. I've tried searching for 'pentode mode' and found a text reference to a drawing George had posted but I can't find it. Any help is appreciated.
with tube amps I would expect it to be a significant difference
but as you said, 30watt is still good
regarding speakers, I think its just as much a matter of general design and function as its about sensitivity, maybe even more
say, Im sure you can drive small 2way with 83db sensitivity, because its never intended to be a power speaker anyway
and if its 85db with 6" woofer instead of a 5", then even better
and not to forget crossovers, how its done matters
is it a lively speaker, or a dull one
lots of things to consider other than mere sensitivity
Thanks loads, that was exactly what I needed, and a special thanks to Ian444 for creating a complete schematic. Well done. Now for the smoke test....
Oh well, first power I got good sound for a few seconds followed but a high pitched howl (oscillation) and pretty blue lightning in the 5AR4. No smoked parts, can't see any solder bridges, whats the likelihood of a bad 5AR4?
Do you have the global feedback connected? If you do, try disconnecting it on both channels and see how that goes.
Edit: or remove the 5K1 feedback resistors and capacitors if fitted, whichever is easiest.
Last edited:
Some joy. My version of the SPP is as follows: JJ 6BQ5's on the output, 12AT7 up front, Edcor XPWR008 for power and CXPP25-MS-7.6k's on the output. Critical voltages seem high. R1 is 150 ohms with 366 VDC on the low side of R1. Pin 1 on V100 is 237 VDC. R1 is dropping 15.65 VDC. V101 plate has 358 VDC on it. No tubes are showing signs of being pushed too hard, no components are visually damaged, no smoke, no bulging caps, nada.
I disconnected one whole channel and fired it up. With the feedback disconnected the oscillation has stopped. Attach the feedback and it howls like a banshee. Sine wave looks ok with no feedback. Voltages are as above. Is the current drop across R1, (app 104ma for one channel), too high for the 5AR4 to handle when both channels are running?? No arcing with one channel, fireworks in seconds with 2 channels as the B+ approaches max.
I disconnected one whole channel and fired it up. With the feedback disconnected the oscillation has stopped. Attach the feedback and it howls like a banshee. Sine wave looks ok with no feedback. Voltages are as above. Is the current drop across R1, (app 104ma for one channel), too high for the 5AR4 to handle when both channels are running?? No arcing with one channel, fireworks in seconds with 2 channels as the B+ approaches max.
Swap the blue and brown wires of the OPT primaries. Your feedback is positive instead of negative.
That seems to have solved the oscillation problem. I've hooked up the FB connections and, for one channel, all is good. But within 60 seconds of putting all the tubes back in the rectifier briefly sparked a couple of times then stopped. As soon as I increased the input signal to about 50 mvpp the arcing started again. Have I murdered poor 5AR4?
The voltages you mention sound about right. The first filter cap C1 is 47uF and no bigger?
Sounds like a bad 5AR4.
Sounds like a bad 5AR4.
Swapping out the 5AR4 solved the remaining issue. The amp sounds fine now. Thanks all for the help!!
Once they arc the first time, they tend to arc again. Alas, the poor fellow is probably gone. 🙁
I was wondering if anything had advanced on running two SimplePP boards as a stereo PPP amp using 6CW5's? The last I heard about it was in May when testing was a ways off.
Right now all I have accumulated are tubes. I have a sleeve of 10 GE 6201's for the front end and a couple sleeves of Siemens 6CW5's. I didn't know if the stumbling block was the toroid main or trying to drive the 6CW5's with it. I am certainly willing to go with other power tubes if that is the prevailing wisdom.
I'd like to get started soon, but I realize that life has been hectic for a while for George.
Any advice would be appreciated.
-Steve
Right now all I have accumulated are tubes. I have a sleeve of 10 GE 6201's for the front end and a couple sleeves of Siemens 6CW5's. I didn't know if the stumbling block was the toroid main or trying to drive the 6CW5's with it. I am certainly willing to go with other power tubes if that is the prevailing wisdom.
I'd like to get started soon, but I realize that life has been hectic for a while for George.
Any advice would be appreciated.
-Steve
I guess there are simpler ways to get more power
It may be called simple, but doesnt look very simple to me, only easy to build
It may be called simple, but doesnt look very simple to me, only easy to build
The Simple P-P itself is about as simple as any P-P amp could be. Now, taking any amp and extracting twice the power that the output tubes normally make, tends to make it a bit "un-simple". The basic Simple P-P runs the output tubes in cathode bias with the screen grids at the same potential as the plates.
Extracting maximum power from any output tubes generally requires "fixed" bias with adjustment for each tube. As with many pentodes, the plates can be operated at a higher voltage than the screen grids. The "maximum power" Simple P-P requires both.
The usual method for fixed bias involves an adjustable negative voltage source for each tube. SY showed us another way with his Red Light District amp. He used LED's instead of the cathode resistors. This works like fixed bias due to the fixed voltage drop across the LED's. It however is not adjustable. It is possible to design an "adjustable led". Several versions have been proposed in the No Light District thread. You need one circuit for each output tube.
The screen grid can not run at the voltages we are operating the plate at. In the mega power version I run JJ EL84's with 430 volts on the plate and 325 volts on the screen. Since the screen grid voltage is the determining factor in tube current, the screen voltage needs to be regulated. A screen regulator is needed, but one can run the whole amp.
SO, the Simple P-P does indeed become not so simple if you want to double its output power. Are there easier ways to make 50 WPC? Probably.
Hectic isn't the word for it. Most readers know that Sherri (my wife) has spent a lot of time up north caring for her mother who has cancer. She hasn't been home in two months. I was asked to volunteer for a project at work that could involve some long hours. That was an understatement. I got two days off in July, and I was there for 10 to 12 hours a day. A new policy has been instituted that forbids "supporting a personal business on company property" That includes phones, computers and internet connections.
So, the PPP dual mono EL84 amp sits on a shelf along with several other projects. I think that I have turned on my bench once or twice since June.
I haven't seen my wife in two months, my daughter and grandkids in two years. They are all in West Virginia for the next two weeks, so I am driving up there on Monday. I plan to stay for two weeks, or until the phone calls from work get too demanding. Either way, I will be behind again when I get back to work. I wish I had better news, but I don't see getting time to finish this real soon. I need to finish some of the basic Simple P-P designs and update the documentation on my web site first.
Steve,
I'm sure there would be no problem running 6CW5 with the Antek 1T200 as George did, this should get you around 28W per PCB (PPP). You would need one transformer per PCB/PPP channel. See post 242.
From post 298, it appears very little is required to convert to PPP. See the left hand photo. Looks like one wire jumper is added, and one resistor is removed. It is a grid stopper to the "2nd" phase splitter, so more than likely George would recommend keeping that resistor in and lifting one end of it from the PCB.
So all you really need to do is buy 2 x PCB's, populate them according to the documentation, get 2 x 1T200 power trannies, and 2 x OPT's. I'm sure if you got the PCB's built George would help you out with the PPP mod. There is plenty of help here. The OPT's will need to be 1.5K to 1.8K impedance plate to plate, check this page here.
It is a simple amp and easy to build, I can't see too much going wrong.
Ian.