I’ve received several emails from people asking me why I’ve gone quite lately. Nothing to be alarmed, I’m just simply buried on the daily grind like many out there are on these Pandemic times. My work-life balance has changed dramatically. This left me with very little time on my hands to work on audio stuff. I won’t complain as many others are actually struggling to survive, that is tough.
As my system is playing amazingly well and I’m so satisfied with my 300B-Liionidas OB setup and the low-gain 01a preamp, I do spend every single time at nighttime to enjoy music. That is what is built for in the end, isn’t it’?
Having some time to update on a long-time design, I received last week the remaining iron needed for my 845/813 SE amp. Yes, the monster one. If you see the output transformers that Alexander built for me, you will say the same. They are 40cm tall and 10.5 Kg weight each! Massive, but needed for a good performance. I originally intended to run these at low-ish operating point, like my friend DHTRob does. With a Zaa of 5KΩ you can do 10-11W with 700V at the anode. Otherwise you can increase the anode voltage to the crazy 1kV or beyond to get much more power, if you really need that.
In my case, I want to aim for a sensible HT and enough power to drive well my OBs. You may have other needs and there are so many ways to skin a cat so I’m not going to open up the debate which is pointless.
If you’re interested, here are the specifications of the OPTs:
Impedance ratio: 4.97k/4Ω 4.42k/8Ω 4.97k/16Ω Primary inductance: 75H Nominal primary Idc: 120mA Max power output at 25Hz*: 35W Primary coil Rdc: 130Ω Secondary coil Rdc 16Ω: 0.48Ω Secondary coil Rdc 8Ω: 0.27Ω Secondary coil Rdc 4Ω: 0.12Ω DC flux density at 120mA Idc: 1.12T AC flux density at 420Vrms**: 0.47T *Theoretical power output without the losses included **This primary AC voltage is equal to the maximum theoretical power output Shunt capacitance: 1.55nF Leakage inductance: 13.2mH Cp/Ls resonant frequency: 100kHz Cs/Ls resonant frequency: 54kHz Frequency response (1700R Zout ; -3dB): 4Ω 3Hz-54kHz 9Ω 3Hz-52kHz 16Ω 3Hz-55kHz
Some driver experiments
I’m looking to build this amp around the 801a valve. The topology is very similar to what I’ve been using lately. A SUT 1:8 with the LL7903 and the custom IT 1:1 directly into the 845 or 813 grid in fixed bias. Simple and yet elegant:
I get 34.5dB gain here and with a very low distortion. See the 250Vpp swing at 0.1%. That’s brilliant with nice harmonic decay:
The alternative option I will test as well, is of course the board I have to complete to replace my 47 driver in the 300B. It’s the hybrid mu-follower version:
Here is with the VT-25, but can be tweaked to run the 801a in the same way. Only challenge is wiring carefully the regulator to avoid hum being picked up as the Rod regulator is high impedance and we have a 1K (un-bypassed) cathode resistor. If you fail to get this working well you have 2 choices, you either add a bypass cap or you use a different regulator (e.g. voltage regulator) which has low output impedance.
Meanwhile, I’m waiting for the superb sockets from Jakeband Audio.
Hope you are all keeping well and safe…
Ale
Thanks Ale for your article!
And thanks to Ale for being patient with the delivery of all the iron by me. As indicated by him, pandemic brought a lot of complexity to our lives and figuring out a way for shipping was a nightmare.
A few words on the iron. On the photo there’s a pair of heater chokes, HC chokes, interstage transformers and OPTs. The OPTs are actually 100W power output capable beasts, but gapped at 35W, delivering a tremendous amount of primary inductance. 4000 primary turns, a nice selection of air dielectrics and careful interleaving brought this result.
Hoping the pandemic will settle at some point, slowly bringing our audio hobby times back
Best regards,
Alexander
I completely forgot to add the Brexit contribution, yes. DHT Rob will joke about it, however is extremely frustrating. Needless to say, you managed to find a way to ship into the UK and am grateful for this. The OPTs are here. The amp will need a crane to be moved around for sure. I think I will end up building them on the IKEA chopping boards first, as usual. I never seem to find the time to finish them up properly in a nice wooden case with the top aluminium plate.
give 211 driver with a choke load connected to source follower to drive 845 🙂 a thought, increases complexity, but sounds fantastic
For movie monsters & transformers, size DOES matter…BTW, are those (all) amorphous C-cores, or GOST laminates?
Ale – thanks again for another great post. Do you find the 1:8 input transformer to hinder the sound and therefore, would it be better to go with a higher mu driver and avoid the need for a input transformer? I too use the EMIA Elmaformer. Just curious, as I am contemplating an input transformer with 801A vs. EML 30B without an input transformer. Thoughts?
In my opinion, it’s not about the transformer itself, it’s about the end-to-end combination an setup. For the transformer to excel it needs low impedance source drive and secondary balance depending the Miller capacitance of the valve driver. Otherwise the result will be a poor HF response. This is the reason why many people disregard them because they don’t take the necessary points in the design and more worryingly is because they don’t measure response to correlate their listening findings.
The EML30 is a good choice (although I haven’t tried it myself but heard feedback from other builders) if you want a DHT driver. I like very much the 801a, VT-25, 46, 45 and 47 as drivers. All these valves need some extra gain to work on most amplifiers so if you have a good iron then you’re in a good track.
I’m inclined to prefer the DHT driver but difference is very small. I had the D3a driver before in my 300B amplifier and have to say that it also sound great.
Hi Ale – thanks for the response. I am leaning toward the Elrog 801a or EML 30A because I could run either tube at B+ 400-420V, thus avoiding any dropping resistors and tapping directly into the C2 (L1 C1 L2 C2) power supply cap. Currently, C2 B+ = 425. I can easily bring this down to 400-420V. I would likely use a custom input transformer from Alexander or Dave Slagle. Are you intending on driving your 801a –> 845 amp with the EMIA Elmaformer? If so, what input transformer are you using? Lundahl? Alexander?
Hi, I’m a big fan of the LL7902 and LL7903 transformers. That’s what I use and am very pleased with the performance.
Hi Ale, I just purchased some Aa tubes as the drivers in my next amp. I am using the Elmaformer from EMIA which is fed by a 100ohm source. Regarding your comment for Miller capacitance, do you foresee any issues using the Aa with the LL7902 or LL7903? The Aa has a Cgp=3.5pF. I can’t find any documentation on the Ckp value for the Aa, so unsure how to calculate the Miller capacitance without the Ckp value.
Also, I can’t find any Aa curves, but using the Ba curves that you published, it looks like an Aa plate voltage of 240 would yield about 120Vpp. This appears to be insufficient when trying to drive my EML 45B to full capacity (your prior post suggest the 45B needs 176Vpp, preferably 200Vpp including headroom). So, is the 120Vpp a non-starter with the Aa as the driver for a 45B?