you built an awesome amplifier! and your skills are great, and I learned a lot from your pictures.
I just wanted to say 'before you build' that by using a smaller output transformer you make the transformer contribute more to the distortion or character of the amp. Does it sounds better now than your other built with bigger outputs?
Yes - and no.
I faced the same situation when once, not for conscious mirrorring (how many 'r's??), but building a stereo amplifier on the same chassis, it turned out to be a non-problem for audio.
Yes, lead inductances, and rather more severe capacitances, will be different. (I laced leads so there was enough capacitance anyway to make me apprehensive). But I found that between the two layouts the open loop frequency response only started to differ quite above the audio spectrum - where any discrepancies (e.g. in NFB stability) could be taken care of by slightly different compensation measures within the loop.
Yes - this will also certainly depend on the design. Some amplifiers are sadly so close to instability with NFB that a capacitance change because of wiring could make waves (!!), but I would think that, unless the layout was woefully poor, there need not be a problem (or else I would certainly have had a few).
Regarding some test instruments, one respectful word of caution though, JDeV:
One actually does need some of the instruments you mentioned for peace of mind in an amplifier build .... not because of audible differences so much as avoidance of a burnt-out tube or component because of inaudible spurious or even continuous instability. Any amplifier using NFB (and even some without it) may have undesirable performance somewhere, and I would certainly spend a sleepless night not having checked with at least a signal generator and oscilloscope that all was kosher. I have encountered enough reason in my life to make such an examination mandatory.
As someone who built Mullard 5-20's in about 1966 and fairly quickly changed them to Williamsons may I suggest you consider this option. Several of my hi-fi acquaintances had been down this road and made the conversion, experimenting with Mull-Will and Will-Mull combinations. We concluded triode operation sounded significantly better and certainly using the EF86 as a pentode without feedback (Radford) sounded distinctly upper middly.
I actually ended up using the same OPT and circuit, did not downscale it at all. Cost wise it was not much more. I can not say for sure how they compare to my old amps, did not hear the old ones for more then 2 years, but I enjoy the sound from these just as much.
I finished the one amp (the mirrored one), and it is actually working.
Here are some pics, nothing special, but it works and sounds GREAT.
Nice work! I wonder, though, why you put the filter 'lytic between both small signal tubes and why you didn't arrange these tubes in the same order as the signal flows.
Best regards!
I comment carefully regarding different designs because most have their staunch supporters. But there should not really be an audible difference between the Williamson and Mullard topologies, in that either comfortably has low enough distortion to make it a 'wire with gain'. Whatever distortion there remains come largely from the output stage, not the pre-stages.
Williamson has gained an undeservedly bad reputation because of some instability problems at l.f., but that is not common to the topology, only because of the relatively low value of electrolytic filter capacitors and component tolerances practical in its day. The 5-20 topology was first used by Leak as far as I am aware, thereafter by several other designers. (There are only so many basic ways one can really design a tube amplifier ....)
It was like that on the original 5-20, and on my previous clones. Worked without any problems.
Oh!
Can't remember; will have another look at the original booklet. As said, much of this has to do with critical component values; with tolerances of +/- 10% originally one wonders if the original test models perhaps had values all to one side ..... (I myself must keep that little datail in mind.)
Sorry, JDeV, here I go off on a tangent. Must also congratulate you on the build!
The one I built is actually an already modified Mullard 5-20, not the original, see the attached documents.
Finished the 2nd mono recently. Here they are, both working equally well.
Only problem I experienced was because off the high supply voltage in my area (245V) I had to use 47ohm resistors in serie with the supply. I ordered 2x 100W metal clad resistors from eBay (China), 1 was open circuit.
Only problem I experienced was because off the high supply voltage in my area (245V) I had to use 47ohm resistors in serie with the supply. I ordered 2x 100W metal clad resistors from eBay (China), 1 was open circuit.
Attachments
Finished the 2nd mono recently. Here they are, both working equally well.
Only problem I experienced was because off the high supply voltage in my area (245V) I had to use 47ohm resistors in serie with the supply. I ordered 2x 100W metal clad resistors from eBay (China), 1 was open circuit.
There as well??
Here (Pretoria, RSA) we also have that problem, as you probably know. Locally wound transformers (imports are expensive what with the postage) are mostly done with mains taps up to 250V. I often have to use series primary resistors of some 33 - 47 ohms for overseas' products as well, but found that a well ventilated 10W suffices.
Fine build there.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.