Hello friends,
I think of creating a Prefered Devices List for amplifiers. Many raisons can explain this aim (grouped buy for shipping costs, avoid fake parts, repetitive questions, easier to reproduce...etc)
This can be the must have for minimal lab. Useful list can mix vintage and up to date parts. I divide it in 3 stages :
- Preamplifier
2N2484
BC550/BC560
- mid stage
2N3440/2N5416
BD139 / BD140
- final stage TO3 and plastic TOP3
BDY58
2N3773/2N6609
At the end, I will make a downloadable file. Thanks for your advices.
I think of creating a Prefered Devices List for amplifiers. Many raisons can explain this aim (grouped buy for shipping costs, avoid fake parts, repetitive questions, easier to reproduce...etc)
This can be the must have for minimal lab. Useful list can mix vintage and up to date parts. I divide it in 3 stages :
- Preamplifier
2N2484
BC550/BC560
- mid stage
2N3440/2N5416
BD139 / BD140
- final stage TO3 and plastic TOP3
BDY58
2N3773/2N6609
At the end, I will make a downloadable file. Thanks for your advices.
See this list site: Driver Transistors
from Greg Erskine of Australia
Many parts available in Europe or Asia are not for sale in the western hemisphere except as counterfeit. Maybe vice versa. So lists should be customized for each market. For example many 2sa 2sc numbers are totally unavailable from US suppliers, except possibly in ******* copies from dodgy vendors. No datasheets from those vendors. Except after 5 years of bumbling around for TO126 drivers I found Fairchild/ON supplies some similar products to 2sa/2sc under the KSC prefix. However all the ones I have bought have been TO92 substitutes with a different "package code" suffix. !@#$#$%&^. You can't draw 2 W out of TO92. Back to MJE340/350 and ON BD139/140 none of which have a Ft spec that the Phillips original datasheet had.
It is telling that Peavey of Mississippi replaced the 2sa/2sc drivers originally in many of their 90's amps with ON semi MJE products. A tiny semi house not maintaining adherance to their datasheet IMHO.
BC/BD available from houses like NTE & New Jersey Semi are missing many specifications on the datasheets. Also true of their 2n products. Buyer beware.
The transistors you have picked are a bit odd. 2n2484 I have, I might use them some day for switching a LED indicator. Certainly not as an amplifier.
from Greg Erskine of Australia
Many parts available in Europe or Asia are not for sale in the western hemisphere except as counterfeit. Maybe vice versa. So lists should be customized for each market. For example many 2sa 2sc numbers are totally unavailable from US suppliers, except possibly in ******* copies from dodgy vendors. No datasheets from those vendors. Except after 5 years of bumbling around for TO126 drivers I found Fairchild/ON supplies some similar products to 2sa/2sc under the KSC prefix. However all the ones I have bought have been TO92 substitutes with a different "package code" suffix. !@#$#$%&^. You can't draw 2 W out of TO92. Back to MJE340/350 and ON BD139/140 none of which have a Ft spec that the Phillips original datasheet had.
It is telling that Peavey of Mississippi replaced the 2sa/2sc drivers originally in many of their 90's amps with ON semi MJE products. A tiny semi house not maintaining adherance to their datasheet IMHO.
BC/BD available from houses like NTE & New Jersey Semi are missing many specifications on the datasheets. Also true of their 2n products. Buyer beware.
The transistors you have picked are a bit odd. 2n2484 I have, I might use them some day for switching a LED indicator. Certainly not as an amplifier.
Last edited:
> a bit odd. 2n2484 I have, I might use them some day for switching a LED indicator. Certainly not as an amplifier.
'2484 was used a LOT in studio console amplifiers as the low-hiss input device (usually in diff-pairs). I'm sure some of your LPs were recorded through multiple '2484s. (The LM394 took over these jobs, so '2484 is old-school, except '394 went out of production, so maybe the '2484 will come back into style?)
'2484 was used a LOT in studio console amplifiers as the low-hiss input device (usually in diff-pairs). I'm sure some of your LPs were recorded through multiple '2484s. (The LM394 took over these jobs, so '2484 is old-school, except '394 went out of production, so maybe the '2484 will come back into style?)
Dear indianajo
you cleverly pointed all aspects of sourcing problems.
localization :
may we turn off buying difficulties with selecting some parts from trusted worldwide distributors (mouser, radiospares, farnell...) or from webshops validated by forum users ?
Others old parts may be mentioned as NOS "value" from "manufacturer". So clearly identified, don't buy anything else !
second sourcing (or even first) :
As you pointed, second line sourcing may have not quality level as original. Worst, even first bigs change manufacturing process altering specs. Answer can only be bring by wise members sharing knowledge.
thanks for the link. A way to not reinvent the wheel
PRR, good guess !
you cleverly pointed all aspects of sourcing problems.
localization :
may we turn off buying difficulties with selecting some parts from trusted worldwide distributors (mouser, radiospares, farnell...) or from webshops validated by forum users ?
Others old parts may be mentioned as NOS "value" from "manufacturer". So clearly identified, don't buy anything else !
second sourcing (or even first) :
As you pointed, second line sourcing may have not quality level as original. Worst, even first bigs change manufacturing process altering specs. Answer can only be bring by wise members sharing knowledge.
thanks for the link. A way to not reinvent the wheel
PRR, good guess !
.MODEL's! I almost hate to not act my age (old), but these days don't we pretty much have to have known-good models? And maybe another column in the table for 'model, known shortcomings'?
Then maybe a column indicating whether the part is available with alternate pinouts. You've already outlined a daunting undertaking -- so many discontinued p/n's, so many new part designs. Doesn't seem that long ago that the biggest obstacle to adopting an innovative new part was convincing the division manager that the second source deal would be signed in time. Now we consider ourselves lucky if a former competitor doesn't buy our favorite vendor and cancel two-thirds of their line six months after we're in production.
There've been at least a couple of projects on this site that methinks might've gone smoother for some builders if the alternate pinouts for some p/n's were more widely known and documented.
For another, I remember one using 2N5401 and 5551's as level shifters, that those parts surely didn't survive very long -- their Vcbo's were exceeded -- and likely ate some more expensive parts when they died shorted. But LTSpice showed it working just great.
This is a brilliant project, will be a lot of work, and probably bring out the Dark Side in at least a couple of us. But it'll be worth it. I certainly do admire your ambition!
Best regards
Then maybe a column indicating whether the part is available with alternate pinouts. You've already outlined a daunting undertaking -- so many discontinued p/n's, so many new part designs. Doesn't seem that long ago that the biggest obstacle to adopting an innovative new part was convincing the division manager that the second source deal would be signed in time. Now we consider ourselves lucky if a former competitor doesn't buy our favorite vendor and cancel two-thirds of their line six months after we're in production.
There've been at least a couple of projects on this site that methinks might've gone smoother for some builders if the alternate pinouts for some p/n's were more widely known and documented.
For another, I remember one using 2N5401 and 5551's as level shifters, that those parts surely didn't survive very long -- their Vcbo's were exceeded -- and likely ate some more expensive parts when they died shorted. But LTSpice showed it working just great.
This is a brilliant project, will be a lot of work, and probably bring out the Dark Side in at least a couple of us. But it'll be worth it. I certainly do admire your ambition!
Best regards
Last edited:
TO3's in a tin make a pretty loud rattling, TO92's less satisfactory I find 🙂Yes. I also want a column in the list showing how good or bad that transistor sounds
Jan
The presence of vibrato on top octave Steinway piano tracks does not require a spectral analyzer to measure. Steinway pianos aren't equipped with vibrato especially on RCA Dynagrove & Colombia Masterworks tracks. AX6 with TIP41c/42c drivers did have vibrato. AX6 with MJE15028/29 drivers did not have vibrato. Same NTE 60 OT's and D44R4 VAS both times. Faster Ft in drivers, anybody?Yes. I also want a column in the list showing how good or bad that transistor sounds
Jan
For calibration of ears, I own a Steinway piano.
As for using 2sa1837 2sc4793 drivers Mike Slavkovic specified, I can't buy those here. If I bought them a German distributor, they would have to go through customs. Newark (farnell US) can't even get a box through customs without pilfering. I have some parts from UK via farnell inter-warehouse transfer, in customs limbo for 9 days. Who wants to bet any On Semi parts aren't subtituted by fakes? It doesn't just happen in 3rd world countries. Farnell does have an incoming inspection process at their SC warehouse, that these parts have to go through.
Last edited:
Its dead easy to get vibrato-like effects at HF in a concert hall as air-movements, temperature gradients, moving audience etc can give fading of short-wavelength acoustic waves through interference patterns shifting in the space. The easiest effect to note is the fading as you move your head in a smaller space, but with a larger space there are so-many variables that can disturb the acoustics.
Its possible to get vibrato in a single string too - if the vibration plane precesses around.
One thing is sure you can measure vibrato with measuring equipment reliably. Hearing it reliably at HF mandates headphones, as head-movements w.r.t. speakers will produce volume fluctuations.
Its possible to get vibrato in a single string too - if the vibration plane precesses around.
One thing is sure you can measure vibrato with measuring equipment reliably. Hearing it reliably at HF mandates headphones, as head-movements w.r.t. speakers will produce volume fluctuations.
You guys are such fun . .
The Steinway having *vibrato* is a little harder to swallow -- even though you own one (count me jealous-green) . .
Reams have been written about the subtleties of the decay envelope -- there are 3 strings singing for most of the notes, after all. I have not seen the word *vibrato* anywhere in the modest bit I've read -- there is almost a whole other, different language to it than we're used to in audio.
Wonder how the rattling of TO3's compare to some of the *modern* (plastic, non-hermetic) power packages . . 😉
Please post those results, when available.
Thanks/Cheers,
Rick
Last edited: