Anyone Cascade Film Cap Bypass their ALTEC crossover caps ??

[LowOhms]

Hello,

In the last nine years, I have spent a considerable amount of time and effort, in multiple cascade bypassing the film caps in my Single Ended Triode amplifiers. I usually ended up using about five or six paralleled caps, of successively smaller values, across each main film cap.

I got to wandering WHO is doing a similar technique with loudspeakers. This weekend, I had a fascinating time, reading the web site on line " North Creek Audio.Com ", and the history and writings of its principal, George E. Short III.

George tends to use about a double or triple film cap bypass across each film cap position in speaker crossovers, whereas I seem to be using about six cascaded, in my triode amps, done by ear.

Casually perusing this web site, I see very little mention as to cascade film cap bypassing in ALTEC crossovers. I am wondering, does anyone do this here? If so, what were you able to determine? Thanks in advance.

LowOhms

Attachment 2748Attachment 2749
Attachment 2750

[GM]

Greets!

Obviously, Altec was prosound oriented, so didn't need HIFI quality components and the majority of its 'career' had the advantage of cheap, plentiful PIO caps, which even cheap grades [motor starting] at least measurably match the best film caps and even outperform them in some apps at least up to a decade or so ago, so only an issue after it was forced the 'bean counters' to get serious and why I historically recommended them, though AFAIK, it's always fallen on 'deaf' ears, so gave up ages ago after being soundly trounced upon and/or verbally abused one too many times by the various forum's self-appointed 'cognoscenti'.

Regardless, I've since occasionally recommended making small value matching, or at least cascading, 'grids' for both resistors and capacitors as I learned early on as an application engineer-in-training that the pioneers of electrical motor design found that this was the way of efficiently, smoothly, starting, speed controlling large motors beginning in the DC era and later of course to AC before auto-transformers were developed and even then was still the best choice in some apps up till SS controllers were sufficiently vetted; so it seemed reasonable to me that it theoretically would be of some benefit in a low voltage signal ckt., or at least a high power, HIFI app and even more so nowadays that high quality resistors, caps are apparently made from rare earth materials based on their pricing.

Audibly, in my relative youth onward, I?ve not heard/perceived much difference except in extreme comparisons, but considering how much I?d abused my hearing by then, I wasn?t surprised, though impromptu tests at the time with young teens [females in particular] and a few <30 yr old adults with [much] better HF hearing, the difference in several cases was enough for them to definitely prefer the grids, so making the ?leap? that cascading would have ~the same results or be a bit more obvious depending on the quality of the large caps.

Of course, the greater the system efficiency, the flatter the frequency, impedance, polar responses, room response, etc., the greater the potential for any major improvement, so with only having a very high efficiency and decent room response at the time, I?m not surprised the results weren?t more conclusive.

Bottom line for me, auto-formers, PIO caps when I can, otherwise build small value grids based on the lowest voltage or current required, which allows averaging out tolerances without ?choking? the supply, potentially matching/having superior overall performance to most [all?] ?boutique?/SOTA parts or if nothing else, so it can be [considerably] cheaper overall.

Ditto medium, large inductors; de-winding a relatively cheap bulk spool works just as well [assuming they?re still wound to the relatively high standards of decades ago] and the extra wire makes for great interconnection wiring of components, whether on a chassis or from chassis to chassis to speakers to drivers, just the number of conductors in the ?bundle? changes, so virtually no chance of differing material physical properties introducing spurious impedance, inductance mismatches in the signal chain assuming all connections are properly silver [or at least silver bearing] soldered and preferably cold welded [screw/bolted compression connectors] where practical, so ideally no component, speaker quick connects other than possibly Cannon or similar based on the UL/CSA testing we conducted.

Too much cost/construction ?overkill? for most of our budgets [hundreds to thousands just for the crimpers/dies depending on how many different types required], but for those few that can afford truly SOTA, there?s enough audible improvement in clarity, spaciousness, ?sweetness? to be had when it?s in the recording to justify it IMNSHO.

GM