Class A-PP-Amplifier with EML AD1

Get technical support here for your Lundahl Tube Power Supply Chokes, Filament Chokes, Plate Chokes, Interstage transformer, Grid Chokes, Signal Input transformers, Moving Coil transformers, Microphone transformers, Digital Audio transformers, SE or PP Tube output transformers, air gapped mains transformers, or the Lundahl problem solver series.

Class A-PP-Amplifier with EML AD1

Postby mvd » Fri Jan 27, 2012 1:00 pm


I have a Question about the correlation of the Airgap in an Lundahl PP-Transformer and the Quality of matching of a pair EML AD1.

I would like to buildt the "Amity" from Lynn Olson, but with a Pair of EML AD1 instead of 300B or 2A3. Here is the Link to the circuit diagram:

It looks like the Lundahl LL1620 wired 6,0kOhm to 8Ohm would be the right choice, but i dont know how big are the unbalances between two matched AD1.
The LL1620PP has an Airgap of 25µm, so says the Datasheet. I assume that this Airgap allows an unbalance of 15mA between the two Tubes. Is that enough for a matched Pair EM-Tubes? And how about the wear of the Tubes over the years? Did the Tubes get old so evenly, that the 15mA unbalance is enough?

Same Question goes for the Interstage. Is the LL1635/5mA a good Choice in combination with a Double Triode like the russian 6N6P or a 5687?

Sorry for my somewhat "broken" English.
Posts: 3
Joined: Fri Jan 27, 2012 12:23 pm

Re: Class A-PP-Amplifier with EML AD1

Postby Admin19 » Mon Jan 30, 2012 11:48 am

Hello Michael,

I once met Lynn Olson on the Triode Festival. He knows theory and the practical side very well, and his schematics are very good. When you want to use AD1 instead of 2A3, you need a 4Volt heater. For the rest it is a plug and play replacement. Two tubes of a pair have always some difference, and as you calculated the Lundahl LL1620PP can do 15mA offset. The LL1620PP is one of the best there is, so don't worry about the transformer! It will do what you expect from it, out of the box. Just as a hint, and only if you know what that is, there is LL1620P-CFB available, this is a normal LL1620PP with an additional cathode feedback winding. You do not have to use the winding, it is just additional, and perhaps interesting to experiment with, after you have it all working the classical way.

For the driver interstage, you see here Lynn advises a DC offset of 20mA. This is really a lot, and when you have MATCHED 6H30 tubes, I would say you can do with less. However indeed 6H30 is hard to match, and match result can never be ultra precise, and aging is likely not to same for both tubes. So you can target for 20mA offset, though it is a high value. You could consider for this any Lundahl Single Ended transformer. This is extremly interesting with the Lundahls, since the windings package is universal. Though originally not intended by the makers of the transformer, with LUNDHAL you can indeed use any SE transformer as PP transformer. This is a most interesting effect from the windings package always being available twice with the Lundahls. Check if the Henry of the SE transformer is enough. If not, just take the next larger one. So LL1660 is larger than LL1635. Check the datasheets for this.

So we use here a SE transformer to get a PP transfomer with large offset. Now 6H30 is very low impedance, so that will work good with LL1635. But you do need to check it with a calculation yourself. Most Lundahl SE transformers can do very high voltage. In case you do not use this high voltage, you can also use a 15mA Version, and get more Henry that way. Also consider the LL1660 which is a larger size than LL1635, it will result in more Henry, at the same offset.


With EML tubes, aging is comparable with two tubes of a pair. This is very good with EML. (Though at End of Lifetime this is not so any more). At first you have burn in. That will result normally in two of the same tubes still. Then you have normal tube life, during which changes are quite small. Like the tubes go down to appr 70% of the plate current, over a long time. This 70% is defined at a tube tester, with FIXED grid voltage. So in a circuit with autobias, you see very little change. Then, when end of lifetime is reached, things start to change more quickly, and then one tube fails before the other one. So the aging curve is a straigt line, which however bends down rapidly at the end of life time. I speak only here for EML tubes. For EML AD1 you will not see 15mA offset, I do not expect that.

With such tubes as 6H30, large aging effects are to be expected. So the 20mA offset makes sense for the 6H30. Aging is mainly with DC plate current. If you force the tubes are the normal plate current, such as a good circuit does, the gain however stays constant.

All together this looks a very good circuit to me.

Here are the datasheets:

LL1620: ... 7_9202.pdf
Lundahl Overview:

Best regards,
User avatar
Site Admin
Posts: 65
Joined: Sun Dec 11, 2011 1:55 pm

Re: Class A-PP-Amplifier with EML AD1

Postby mvd » Mon Jan 30, 2012 2:35 pm

Hello Jac

Thank you for your long and enlightening response, now i am much more convinced to be on the right way. I will follow your suggestion and check the interaction between Interstage and the possible driver-tubes.
Like the 6H30 also the 6N6P, E182CC, 5687 or JJs new ECC99 has very low impedance and are suitable candidates for the Job.
And maybe someday i can switch from these cheap Double-Triodes to real nice DHTs like EML 20A-Mesh as the driver for the AD1. So maybe your advice using the bigger LL1660 is the right way to go, since the 20A has approximatly double the impedance of the modern Double-Triodes.
The people say that the sound is so nice driving a DHT with a DHT. I have never heard one, but maybe someday...

My choice towards the AD1 is not only because this is an old Telefunken tube, the crucial factor is the exceptional good socket of this tube. In germany we say "Topfsockel" ;-)
In my collection of old radios i have a Tandberg Solvsuper2 back from 1938 where all Tubes (AZ1, EL3, EF9 ...) have this socket and there are no problems with contacts ever. It is a brilliant construction.

Despite the Amp will be split in 2 Monoblocks, these will be very heavy-weightet because of all the iron inside. So i consider to split even the Monos into Power-supply and Signal-chassis. Since i have not much experience in building Amplifiers, my first attempt will be a prototype on a breadboard before i look for a final design.

Best regards and thanks again
Posts: 3
Joined: Fri Jan 27, 2012 12:23 pm


Who is online

Users browsing this forum: No registered users and 1 guest