And now for something completely different. DHT and boatanchors are super awesome, but sometimes you just want to build something simple that does not take 6 months of saving up to just buy the iron or obscure tubes.

My very first tube amplifier was from an Elektor (iirc) article in the 1980s, I was still in school studying electronics and stumbled upon a diy push-pull EL84 amp. I'd always loved the old tube radios, and I knew that the one we had at home had a single EL84 in it as well; so it had to be good if it had two. I built that amp with some DIY PCBs someone made at home for me, got amplimo transformers for it. It played for a few hours until I realised my 88dB Visaton speakers did not like the 8-ish watts this amp produced. So it got shelved and eventually the PCBs thrown away. 

So I want to build something simple, with as many parts that I already have here. Seeing as I have three pairs of amorphous Tribute SE iron, and one pair of Bartolucci SE OPTs, the amp would have to be single ended. The power transformer that comes close should also be in my pile somewhere, already had some russian 6P14P and 6Z9P tubes so that leaves me with a few resistors and capacitors. I could have listed Telefunken/Mullard tubes for the E180F and EL84, but seeing as I will not use them in favor of the russian tubes I alreayd have, I will leave that choice for someone else to make :)  They are identical, so they fit right in.


© VT52.com

A secondary motivation for this amp is that I might want to tri-amp my system later on, just because I can. One of the pairs of Tribute OPTs I have has a 7K primary and was wound for 1kHz and up; so it was already intended for this kind of system anyway. The circuit was specd to handle down to 8Hz -3dB in regards to the cathode decoupling and the coupling between the stages. No dount my final implementation will have something like a 10nF/100K RC filter for coupling and might remove the cathode decoupling caps cmpletely. The schematic however reflects full-range use, should anyone try to build it and find unsuspectingly that it blows in the bass and mids.

Both tubes have been triode strapped and behave VERY nicely here. The 6Z9P at 145V/15mA on my eTracer gives a predicted 0.6% THD with the 8K2 load and full swing for 1W out on the 6P14Ps, which is very low. I don't have a distortion analyzer, so can not back up those numbers in real-life. The EL84 shows about 2.5% at 1W out, around 4% at full swing (±1.5W). I'm labeling this amp as a 1W amp as that last 0.4-0.5W don't really mean much in the grand scheme of things.

The 6P14P/EL84 was chosen out of nostalgia, but the 6Z9P/E180F is quite new to me. My old PP EL84 amp used an EF86, so maybe you could say it is still nostalgia in some sense? Anyway... I bought this tube around 12-13 years ago and never used it, so this seemed like a good place to let it shine. It is an amazing little tube in triode mode with 1K plate resistance and a mu of 35. It biases nicely at around -2V, although as with all high gm tubes, this can vary quite a bit as I have similar tubes that go all the way down to 1.2V to get the same current. Even with resistive loading, the gain is around 27x meaning that it has plenty of swing to get the EL84 to maximum output (biased at -9V). We only need a gain of around 10-12x to drive the EL84 to full output, so the E180F mu is kind of high.


© VT52.com


The power supply is fairly simple and compact. The heaters are all plain 6.3Vac windings, a seperate one for the left and right channels. The B+ supply is split from a single solid state rectifier, here consisting of four fast-recovery diode I still had, but you can substitute with the UF4006 or UF4007 for example, an ultra-fast version of the ubiquitous 1N4007 and just as cheap.

I skimped on the rectifier tube to keep it simple and affordable, but you could ofcourse stick a russian 5C3 or 5C4M in there and add an extra heater winding for it. For the filtering splurged on a single choke as it greatly reduces the ripple, namely around 10-12mV according to PSUD2. After the choke the filter is split into two channels where we find two RC networks. The ifirst one has a low value resitor that will allow us to tune the B+ to around 300V with minimal voltage drop to impede the EL84. The next RC stage is to drop 30V and do some extra filtering for the driver stage. The resistor wattage ratings  I've used Nichicon KX electrolytics for the first three stages, and an MKP for the driver. You could swap all the other caps out as well, but that would raise the price considerably.

Note:  High-gm tubes like the E180F tend to need some extra TCL regarding oscilations. The E180F is not the worst in this regard, but when using tubes like these or the E810F, 5842, 6S45P etc,  it is absolutely necessary to take the following precautions:
- connect every grid and cathode pin, even if it seems pointless, as they can act like antennas
- put a ferrite bead or 100-150R grid stopper on each grid pin and oin them all at one point where you connect the signal wire to.
- place a 100nf ceramic capacitor directly over the heater pins of each tube
- place a 10nf ceramic capacitor on each leg of the heater to the chassis.