Simply put, is there a difference in sound quality between the likes of Lundhal/Tango/etc. and say Triode Electronics/Latino Dynakit Iron/etc. and if so, what is the difference that can be heard?

I know some designers that have used mega buck everything and say they hear no differences comparing to even 40 year old stock Dyna iron. Of course I have read plenty that think the $$ stuff does sound better.

Any thoughts/opinions?

Thanks all!

 

If I remember correctly, Dynaco used quite complex layout of primary, which is rarely available today and yield excellent results from all technical and sonic aspects. If you wish something radically different, and possibly, with better sonic qualities, try transformers with 50% nickel iron, Fe-based 1.56T amorphous, 1.2T nanocrystalline cores, or toroids.

 

Thanks LinuxGuru! I use right now, some Triode Electronics A431's in a Dyna MK3 with an odd-ball circuit. The design essentially barely even takes advantage of the iron and output tubes. I am achieving about 35-40 watts total power, about what I would get if I used a modern equipped ST70.

Reason I asked this question is because my theory (I'm sure I am wrong) is that with using overkill iron/ps/etc., the iron should be providing a lot less distortion than iron pushed harder. This said, it could be that iron will either be superior or inferior regardless of how it is being run.

 

This is right in some degree. Iron pushed hard only at lowest frequencies / power close to max (= max flux density), this is where problems begin with undersized cores or too low primary turns.
On the other side, larger transformer means higher leakage inductance and stray capacitance, and therefore, more phase shift and distortions at high frequencies.

In short, core should not be undersized or oversized, it must be balanced to target specification.

 

To put it very simply, big heavy iron is good for bass, but sucks for highs. So TX companies figured out ways to group windings into small bundles and do all sorts or weird stuff to get the best of both worlds. Dynaco was one of the best at this, which I assume is why their TXs still bring top dollar. There is a company called Heyboer with a dude named Alden that has "unwound" a ton of the best old iron to reverse engineer the recipes. So IMHO, the newer stuff with the old techniques is the best as there is no rust or other nasty deterioration going on inside where we can't see. I got that opinion mostly from Morgan Jones. But in short, winding an output is an art form.

 

I have not compared up (A431 to more exotic) like you are asking, but I have compared perhaps more equals. I built a copy of my KT120 amp for a friend using Hammond 1650R iron. After completing it, I compared it to mine using triode electronics A431S iron (everything identical in the amps except output iron). I can definitely hear a difference, with the A431S slightly edging out the Hammond. But mostly, I'd say they just have different strengths.

The A431S is an excellent transformer--deep articulate bass, with highs that are really quite good also. The entire system sounds very even top to bottom with the A431 clone.

The Hammond 1650R does not go as deep (maybe 1/2 an octave less)--and not quite as smooth or even keeled across the spectrum, but the highs are a bit more "delicate" I think describes it the best. The sound stage tends to be a wee bit wider with the Hammonds also.

 

All things being equal (which they seldom are, in answering your question)? I'm not sure there is a meaningful answer to be found. In the same circuit with feedback adjusted to optimum for that transformer would be the only way to make these comparisons. I doubt there are many out there that have done so. I have compared Dynaco transformer's (A-430 bell ends cloth wire, '57 A-430 potted cloth wire, and the ZTO194 '74 plastic wire) Nothing scientific, just my ears for results, I didn't hear a dime's worth of difference. Did the same thing with a bunch of Eico EL84 OT's. 4 different GRT's (today's Heyboer), Supreme's, and Chicago Standard. Same thing, no discernible audible diff to me. Hey what can I say, I am from Missouri (as in show me
? lol) Satisfied my curiosity YMMV. People will believe what they want to believe (I paid a grunch more so it must be better and I will like it more?) However, no one's good taste is better than your own. I used to think Mac's were the end all be all, only to find out that there were other amps out there that I liked as much or more. Ya' gotta keep an open mind and listen and experience as much as you can. That's my answer (which isn't really is it?)

 

There probably is a noticeable difference in going to the expensive brands you mentioned because of the additional interleaving and extra complicated winding you would hope they would have, but it may not be worth worth the extra $ to you. There is a lot more labour in that so the extra $

Back in the day for example the more expensive HK Citation II was supposed to have better OPT and they are still highly valued so go for more $ still today.

O-netrics (makes 5 different levels) and Hashimoto are others available in NA worth mentioning.

As mentioned Heyboer can replicate some of the best vintage ones also.

 

Thanks everyone for the responses.

Not to go off topic, but I am trying to understand a little more about the Ultralinear "tap" that my Triode Electronics iron OPT has and just what % of Ultralinear I can expect to have with it? Also, what % Ultralinear did Original Dynaco transformers have and how is this % determined? I really like the "concept" of ultralinear, but am not sure how to achieve it especially if my transformers are essentially/primarily pentode based. This all said, in going with a more exotic tranny, this may get even more troublesome or not in terms of achieving as close to a true ultralinear output.

Thanks!

 

I have the info somewhere, but it IIRC it was 20 or 30 on most Dynaco iron. Ultralinear just steals the voltage for the screens from the OPT winding. My personal preference is to run pure pentode with solid state regulated screen voltage. A completely separate supply would be my idea of ultimate.

 

The UL % is typically expressed as a percentage of the primary winding (as opposed to a % of the primary impedance), and can be found by conducting simple AC voltage tests on a given transformer. The percent of the winding to in fact cause UL operation varies from tube to tube, and based on how the tubes are operated. The required UL percentage can range from as low as about 23% (for fixed biased 6BQ5 family operation for example) to as high as nearly 50% (individually cathode biased EL34 for example). With some tubes (the 6L6 family for example) , the required percent to achieve UL performance is very narrow, while with others (KT88), it is rather broad. UL operation is specifically defined as a point of operation where distortion lower than that obtainable with triode operation, an output impedance equal to that of triode operation, and power output nearly equal to pentode operation are all achieved together.

The problem with pentode operation is that it has a very high output impedance. This provides poor damping qualities for a loud speaker, tending to cause the speaker to have a booming, or false bass tendency. But it does produce a lot of power output.

The problem with triode operation is that it has a wonderfully low output impedance to provide good speaker damping, but produces low power output.

UL operation operates midway between these two modes, with feedback provided to the screen that not only produces the benefits mentioned above, but also one other outstanding benefit: UL operation causes the tubes to operate in such a way as to always "want" the impedance that is actually being reflected back to the tubes from the varying load presented by the loudspeaker on the secondary. The result is that regardless of the varying load, maximum power is always transferred (within in reason) with UL operation. This is a benefit that triode and pentode operation simply cannot match. It's no wonder that engineers and consumers alike took to the concept so quickly when it was developed!

Dave

 

I tend to run pure pentode into lower than "usual" Zs in PP amps. I have several UL amps and like them too though. When it's all said and soldered up it's the resulting sound that matters, and I think all my amps have there own "signature" sound. One or two I don't like. There's something about the 6V6 tube that turns me off, but I fear it's all in my head. The tube looks great on paper, of course the HF81 looks terrible on paper, but try to buy one in decent shape for under 5 bills.

 

I am told by an employee at triode electronics that their A431S has 33% UL taps. I should have measured this for myself before I bolted mine in. :scratch2:

 

The A-430/431 and A-470 transformers all have their screen taps located at 33% of the primary winding.

Dave

 

Thanks for all you wrote Dave and also to others. Very much appreciated and always enjoy reading. I cannot think of a better forum than this one.

So that 33% figure, is that for the originals AND the newer types because Latino's website says 40% for his ST120 amp. I would be curious if and why all/specific transformers in his amps measure at the 40% spec.

 

I think you can measure this in place, when the tubes are cold they have
very high impedance and may be ignored. The feedback is in the secondary side and rather high impedance so it _should not_ interfere.

Thus the procedure : turn off amp. Open so the anode and screen can be measured ( could be done by removing the power-tubes and measuring on pin 3 ( anode ) and 4 ( screen). Then apply 1000hz from a signal generator on the 16ohm outlet.
1000hz should appear on anode and screen connection ( relative ground)
Then i'm uncertain how to interpret , if we have 1V on 3(anode)
and .33V on 4(screen).
Dave might come to my rescue here.

 

The turns count and voltage are a 1:1 relationship.If the screens are 33% of the way between CT and anode, there will be 33% of that voltage from CT-screen. Put 5V of 60 cps across the plates, and look at the screen-screen voltage. The OPT can easily tolerate 60 cps signal...it is quite capable of less too( or it had better be ).
cheers,
Douglas