Please say more . As far as I know my BNC's never give me problems until a wire breaks .
I like the way BNC fits together , the silver plating they use , joy of soldering them , PTFE insulator , profile of the tip . I take some trouble to find cable that suits them . The shield coupling is not exactly ideal long term in some BNC .
High grade DIN are very nice . The pins don't have to be used as intended . I bought a very big bag of 7 pin DIN . I often use them for lash-ups . Often they seem superior to the better RCA .
I use an OFC copper cable from Farnell that is about 20 cents a metre . It is something I trust which seems OK for experiments . I swear it is better than many cables . Friends love it fitted with the cheap Neutriks . It is loosely speaking a microphone type which is beautiful to work with . I use 4% silver 0.5% copper + tin solder . I would say under most circumstances it is good enough to not have doubts about . I only bought it originally so as to have something OK around . It is more than OK . Trust me I know expensive cables . I even respect some of them . This cable is not horribly inferior . It has the dreaded PVC , sounds OK to me . Quad said years ago people never compare the connectors they like using the standard cables they have rejected . If there is a difference if will be in the connector . The sad thing is Quad made no friends saying this . Those who thought the same continued thinking the same , those who didn't believe it added it to a list of reasons not to buy Quad . Very sad . Tell the truth as you see it and suffer . When I saw RCA on the latter Quad I asked why . They said they had given up fighting . Then with a twinkle he said " DIN is better " .
CBBR4166 - PRO POWER - CABLE, 18/0.1, SCRN, 100M | Farnell United Kingdom
If one did it seriously, you would have to be very careful to duplicate the layouts as closely as possible. The RCA element is only using the centre pin to make a connection, the return path metal perhaps should hardwired to the link on one side.
Yes, the impedance of the node would be highly significant, but at least going to extremes would clarify, for people doing the test, would sort of audible artifacts are introduced, if any. Then one can step back from excessive 'poor' construction, connection by connection, to see at what point it becomes inaudible for the majority of people.
Personally, I gave up on RCA decades ago; I spent weeks if not months in that period jumping through all sorts of hoops, attempting to negate the impact of the connection degrading. Finally, I bit the bullet and hardwired - problem solved! - and have always done so every since ...
I haven't tried this, but a friend who used to follow all the usual audiophile trails did: he had a fancy interconnect, and a very ordinary one - there were definite audible differences between the two, he said, when plugged in conventionally. Following my thoughts, he tried hardwiring one, and then the other, and all those differences completely disappeared ...
I have an acquaintance who is a small scale manufacturer of cables. He tried all types of connectors and terminations.One day I was over and he asked me to listen to one interconnect vs another. I wasn't told what was different between the two and they both looked identical including the phono plugs. One sounded cleaner, and for want of a better description "more direct" than the other. The difference turned out to be that the plugs on one were soldered; the better one had been laser welded by a local jeweller. I was sufficiently impressed to look at the cost of laser welders. I soon gave that up when I found out!
Caig.com's "De-Oxit" comes immediately to mind. And they have a couple of other products that seem very good; one for gold connectors and one for potentiometers.
I haven't done any comparison tests with Caig's stuff, for audio connectors, but I CAN say that De-Oxit, by itself, "fixed" a lot of problems with test equipment, back when I used to re-furbish and repair a lot of it. I always keep on hand both the spray can version and the concentrate version with the needle applicator.
They had some amazing testimonials on their website, last time I was there (a decade or so ago).
I haven't done any comparison tests with Caig's stuff, for audio connectors, but I CAN say that De-Oxit, by itself, "fixed" a lot of problems with test equipment, back when I used to re-furbish and repair a lot of it. I always keep on hand both the spray can version and the concentrate version with the needle applicator.
They had some amazing testimonials on their website, last time I was there (a decade or so ago).
Yes you do, N-type is a little better and the (huge) DIN 7/16 is much better
Absolutely to be avoided are pressed connections as found in most RCA and many BNC
It is so common to see one of the tighter RCA plugs pulling the RCA socket outer off the back of an amplifier. The connection in the socket is only by (too light) pressure, with no wetting current to keep the contact clean. I am not at all surprised that other posts here say that direct connection of the cable shield down to the PCB removes all of the differences
Unfortunately, these type of treatments are not good enough, at least for me. In my experience they initially seem better, but do still degrade - and when they do so, they are actually worse than a 'raw' connection. The De-Oxit Gold seems to be ranked the best -- but interestingly enough a member of forum who was asking for advice recently, said that he had partially hardwired and partially used the De-Oxit. I suggested going for broke, and hardwiring those treated connections; initially reluctant, he finally did so, and was hugely enthusiastic about the outcome - a major improvement occurred ...
Last edited:
Agree entirely. The quality of construction of virtually all RCA sockets is quite dreadful ...
in heavy duty test equipment, such coated pins have superior oxidation stability and are longlife
had reliable contactation problems (few amps tru circuits) with s-steel or nickel, these works wonderfully. no signs of surface problems, just clean it alcohol and looks new
these last bit longer than gold variants
Composition of INGUN Test Probes - Ingun UK and Ireland spring loaded test probe supplier
had reliable contactation problems (few amps tru circuits) with s-steel or nickel, these works wonderfully. no signs of surface problems, just clean it alcohol and looks new
these last bit longer than gold variants
Composition of INGUN Test Probes - Ingun UK and Ireland spring loaded test probe supplier
I noticed a comment about pressed construction RCA's being bad . I think the cheap Neutriks I listed are one piece ! BNC criticism I totally accept .
I recently did some measurements on distributed load or ultra linear feedback of pentodes . Had the equipment we now have been available in 1937 I suspect another UL point would have been recommended . I would call it the Jean Hiraga point . I have a hunch that a near true triode point is available in pentodes . Not the statistical best distortion / power compromise . That is to say g2 has been nullified in the new point . Pure conjecture as I haven't done enough tests . Also , various makes of lets say EL 34 might not agree . I suspect something like a true triode point would be available . Not wanting to say definitively I did find something . Just to say how we might better understand something using test equipment . Doubtless on a simulator all of this can be found . I did it the hard way . 80 % would be about right . I think if it was missed it was because it looks just like a conventional fake triode on the old graphs . What I noticed is the harmonics look more like a true triode at 80 % . On my tests there are some higher harmonics in the spectrum if g2 is strapped to anode that disappear at 80 % UL . They come back as pentode end UL is used . The pentode end is actually better ordered than the strapped end . Not proof of anything but interesting . I am connived an EL 34 used this way is not inferior to a 300 B excepted the effects of direct heating the cathode and bottoming .
One thing I didn't try before sending the amp on it's way was proper pentode with shunt feedback ( potenial divider g2 supply ) . It topologically would look like an Alex Kitic design . I have built an all pentode design like that where one capacitor deals with all the shunt feedback requirements input to output . One capacitor that most of use would use anyway . It didn't look exactly like triodes on the analyzer . I think I did find that pentodes with feedback are outperformed by UL in terms of distortion verses gain . I had expected pentodes to have a distinct gain advantage . Not really when crossing the less than 1% distortion line . I dare say the damping factor was better .
I recently did some measurements on distributed load or ultra linear feedback of pentodes . Had the equipment we now have been available in 1937 I suspect another UL point would have been recommended . I would call it the Jean Hiraga point . I have a hunch that a near true triode point is available in pentodes . Not the statistical best distortion / power compromise . That is to say g2 has been nullified in the new point . Pure conjecture as I haven't done enough tests . Also , various makes of lets say EL 34 might not agree . I suspect something like a true triode point would be available . Not wanting to say definitively I did find something . Just to say how we might better understand something using test equipment . Doubtless on a simulator all of this can be found . I did it the hard way . 80 % would be about right . I think if it was missed it was because it looks just like a conventional fake triode on the old graphs . What I noticed is the harmonics look more like a true triode at 80 % . On my tests there are some higher harmonics in the spectrum if g2 is strapped to anode that disappear at 80 % UL . They come back as pentode end UL is used . The pentode end is actually better ordered than the strapped end . Not proof of anything but interesting . I am connived an EL 34 used this way is not inferior to a 300 B excepted the effects of direct heating the cathode and bottoming .
One thing I didn't try before sending the amp on it's way was proper pentode with shunt feedback ( potenial divider g2 supply ) . It topologically would look like an Alex Kitic design . I have built an all pentode design like that where one capacitor deals with all the shunt feedback requirements input to output . One capacitor that most of use would use anyway . It didn't look exactly like triodes on the analyzer . I think I did find that pentodes with feedback are outperformed by UL in terms of distortion verses gain . I had expected pentodes to have a distinct gain advantage . Not really when crossing the less than 1% distortion line . I dare say the damping factor was better .
Last edited:
Does anyone here have any experience with THAT Corp.'s 340 chips?
These are precision matched 2 NPN and 2 PNP transistors, with a very low noise, below 1 nV/sq.rt Hz, reasonable 36 V max, Ft of 350 MHz, and a very reasonable price, in Europe €3.65, or about $5 per chip, single piece price?
On paper, they look damn well ideal for a fully complementary input stage, and 36V max is not really a problem for an input stage. The idea of close matching of NPN to NPN, PNP to PNP and NPN to PNP is soooooooooooooooo appealing.
Same substrate, same housing, should be thermally stable like the Devil himself and matched beyond our own capabilities without a hellf of a lot of work.
These are precision matched 2 NPN and 2 PNP transistors, with a very low noise, below 1 nV/sq.rt Hz, reasonable 36 V max, Ft of 350 MHz, and a very reasonable price, in Europe €3.65, or about $5 per chip, single piece price?
On paper, they look damn well ideal for a fully complementary input stage, and 36V max is not really a problem for an input stage. The idea of close matching of NPN to NPN, PNP to PNP and NPN to PNP is soooooooooooooooo appealing.
Same substrate, same housing, should be thermally stable like the Devil himself and matched beyond our own capabilities without a hellf of a lot of work.
Last edited:
My friend John uses them . I think I read 2N4403 and 4401 are almost that low in noise ? 2SA1085 lower ( 0.55 nV ? ) . The matching isn't so hard and I doubt any circuit needs that . I have always said 6 transistor in a long tail pair is best . If emitter resistors are used so much the better as they will have dual function . If emitter resistors are used it calls into question the noise reduction priority . All the same a good idea . My friend uses them as his company is rich and they would never question it . He does because he can . Wouldn't it be ironic if MPSA 92/42 do some things better ( often criticized if used ) . Come to that does anyone use small power transistors for input stages ? I know at one time it was fashionable to do things like that . Some think noise is the thing to avoid . They say lower noise = greater warmth ! Other say differently and prefer J-FETs on noise type .
BCV 61 is great also as it has one substrate . For a current mirror it no longer is a caprice . I never built a discrete one to match it .
BCV 61 is great also as it has one substrate . For a current mirror it no longer is a caprice . I never built a discrete one to match it .
I don't believe the NPN is matched to the PNP (it would help if they showed curves of Vbe with current, but note they say excellent matching of transistors of the same gender). Base spreading R is pretty low but there are lower. Beta is not very high. Someone there told me once that they tried making beta higher but the breakdown voltages got too flaky.
For some applications, especially those that require good log conformance, like translinear circuits including analog multipliers, they should work very well.
If there were packages with more pins you could have eight independent devices, and they will sell the chips if you want to manage the packaging yourself.
I worked with government agencies in the past to rectify intermod from competitors. It wasn't connectors.
My Brain is as old as yours,
you expect me to remember everything, who was Sy Involved with before linear audio, was it Glass ......?I must be Mixing up the whole shing ding .....
- Status
- Not open for further replies.