Hi,
It was from European Ebay shop and they're claimed Fairchild devices BC560C.
I've seen the CDIL ones but wasnt sure of the manufacturers reputation. Those CDIL are plentiful however.
In the end I think it cost me perhaps £2 for ten, thereabouts.
If they're fake I dont think I would notice in my CCS circuit, as the performance seems about the same whether I use BC560C or something like BC557B, and I havent checked gain or anything like that (I cant find my DMM with transistor checker)
From this graph it looks like being able to work at approx 9ma, 550v for max10ms.
May i ask what was the current at which this transistor blew away?
Attachments
It was around 10 years ago. But fakes they were, low hFE and very low VCBO, half what it should have been. Once I tracked down some genuine ones, purchased as part of a bulk buy, with measured 900V plus breakdown and correct gain, everything worked perfectly as it was designed to do. And continues to work perfectly to this day. Switch it on, listen and enjoy.
I bought the fakes from Dalbani (AKA Nikko), who I will never deal with again. They did refund the money once I sent the evidence, but had no interest in refunding the wholesale destruction of the rest of the circuit.
I bought the fakes from Dalbani (AKA Nikko), who I will never deal with again. They did refund the money once I sent the evidence, but had no interest in refunding the wholesale destruction of the rest of the circuit.
Yeah - apologies - I completely changed the subject of the thread.
Let's get back on topic - sourcing credible sources of BC550/BC560.
It is worth pointing out (it probably has been already somewhere on this thread) that the 550/560 are not particularly low noise devices.
I went so far as to moan to On-semi though, who, since they took over Fairchild three years ago have been hell-bent on obsoleting anything of utility to audio. They said that it was a "management and marketing decision". No surprise there. Vishay did the same sort of thing when they went on an acquisition binge some years ago.
Let's get back on topic - sourcing credible sources of BC550/BC560.
It is worth pointing out (it probably has been already somewhere on this thread) that the 550/560 are not particularly low noise devices.
I went so far as to moan to On-semi though, who, since they took over Fairchild three years ago have been hell-bent on obsoleting anything of utility to audio. They said that it was a "management and marketing decision". No surprise there. Vishay did the same sort of thing when they went on an acquisition binge some years ago.
Last edited:
Would ZTX851/951 be better choices?
What about those 2SA/2SC suggested some pages behind?
Yesterday I did several noise simulation on LTSpice with the SLN, using several transistors, and compared it with the Adcom mod RIAA with AD745/LT1010 and a Luxman 5M21 discrete with dual FET input, and the results were remarkable.
The SLN beats them all below 100Hz and equals them above that.
Even if noise sims are not too reliable, and the results were quite similar, I think they might hint the SLN is among the best.
Sims are available for anyone to see.
rbb' is not the most important factor with a moving magnet cartridge. It is down to a trade off between voltage noise and current noise - hence the typical optimum low value of collector current of 70-100uA. What you are trying to get to is Vn/in ~ 5k-10k-ohms (the typical mid-band impedance of an MM)
So no real advantage going for the low rbb' ZTX devices, which are more applicable to MC stages. The 2SA/2SC surface mount devices I suggested are not a bad choice, as are the KSC1845/KSA992 which are at least available in TO92 (check pin out compatibility). There are others that have potential, but would not fit the board layout and there are no adaptors available.
So no real advantage going for the low rbb' ZTX devices, which are more applicable to MC stages. The 2SA/2SC surface mount devices I suggested are not a bad choice, as are the KSC1845/KSA992 which are at least available in TO92 (check pin out compatibility). There are others that have potential, but would not fit the board layout and there are no adaptors available.
Last edited:
Around 3nVrtHz mid-band. Also steep climbers regarding 1/F unfortunately. High hfe when in rank C and lowish for Cob nonetheless. The 2SC/2SA are also high hfe in rank BL with even lower Cob (PNP) while better in noise and linearity too, so generally better.
Medium gfs Jfets have not so high Cin though, mostly a consideration of Crss and Miller. Cascode can help down to 2x gain Miller equivalent but add Cin and layout parasitics. Source degeneration helps via local feedback too, since the source impedance is high enough with MM to allow Rs even in the hundreds of Ohm. Anyways, all those things should be carefully considered vs a certain MM model and cabling. Some AT carts for instance can be particularly sensitive to >100pF becoming pinched enough in the treble.
Beyond the practically zero current noise advantage with a Jfet input, be it discrete or op-amp of that type, there is the safer DC input coupling thing too.
A friend for example had killed his Benz ACE cart few years ago after he fried a 5534 op-amp input stage while experimenting live. He had fried K170 Jfets a couple of times too while the cart was directly connected, being as stupidly hasty as he was, but the cart survived. That must had spoiled him for further adventures.
A friend for example had killed his Benz ACE cart few years ago after he fried a 5534 op-amp input stage while experimenting live. He had fried K170 Jfets a couple of times too while the cart was directly connected, being as stupidly hasty as he was, but the cart survived. That must had spoiled him for further adventures.
That is absolutely right - you can get down to 50 to 100pF that way. But adding the cabling can quickly exceed the load capacitance of many MM's.
Very low noise JFETs like the LSJ74/LSK170 are still a good choice for MC, where load capacitance is not such a big deal. Although cascode is still a good idea to increase linearity.
Very low noise JFETs like the LSJ74/LSK170 are still a good choice for MC, where load capacitance is not such a big deal. Although cascode is still a good idea to increase linearity.
Kinda OT warning:
I dont know who looks for devices and checked availability of BC560....but they're blind.
On Semi BC560CG are in stock at major sources (RS) OK so they're mid gain band, not high gain band, but what the hey!
Oh and also BC550C are plentiful, Taiwan Semi (also RS), and On Semi BC550CTA...
I wish the "fake device alert squad" would get their facts right!
I dont know who looks for devices and checked availability of BC560....but they're blind.
On Semi BC560CG are in stock at major sources (RS) OK so they're mid gain band, not high gain band, but what the hey!
Oh and also BC550C are plentiful, Taiwan Semi (also RS), and On Semi BC550CTA...
I wish the "fake device alert squad" would get their facts right!
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.