|
Post by flateric on Jun 28, 2010 10:02:50 GMT -5
Posted this on the DIYStompbox forum but no decent reply yet. A lot of home pedal builds call for different general purpose amp transistors such as BS170's, J201, 2n3904, 2n5088. When you don't have the exact one (some are a pain to source in the UK) how do you go about checking the online data sheets to see if you have a suitable replacement available? Other than checking they are both NPN or PNP, the datasheets have 7 pages of figures and graphs. Any way of telling which bits of info are important before you decide to change your 2n3904 for a 5088? Any repercussions on biasing needed? - and if so, where is the info that can tell you where resistance would need a tweak?
|
|
|
Post by D2o on Jun 28, 2010 11:20:02 GMT -5
Posted this on the DIYStompbox forum but no decent reply yet. I am not saying that I'm about to change that, but it seems like there may be more than one J201? Someone asked a similar question and got a response from someone with the username “the_random_hero” (hmmm) here. I don’t know if you are asking specifically about the 2n3904, but if it is of any help, one of your fellow diystompboxes forumites also wrote: I don't know if there is truth to any of it, but I have read enough to believe that transistors can be swapped so long as the rest of the chassis is also adjusted accordingly (to deal with any change as a result of the transistor swap). Allow me to pre-respond to my post for you: Posted this on the GN2 forum but no decent reply yet.
|
|
|
Post by flateric on Jun 28, 2010 13:02:30 GMT -5
OK I have some 5088's, I'll use instead of 3904's but it looks like I can't swap a 3904 for a J201 as its a jfet.
|
|
|
Post by D2o on Jun 28, 2010 13:33:01 GMT -5
Maybe not, but a cursory search indicates that there may be some hope with respect to the J201 issue. This dude (J. Donald Tillman) was originally using a Motorola 2N5457 N-Channel JFET in this design, and swapped it for a J201 … he does, however, indicate that the 2N5457 “also works well” but is not as quiet. This MXR Phase 45 project at tonepad uses 2N5952, but this build has substituted them for J201s (using a 200K trimpot instead of 250K). If JohnH or sumgai checks in I would imagine one of them would have some more useful (and credible) input ( edit: I meant "more credible than I have offered" - no slight intended on the people I have linked to), but at least there are some builders out there who are indicating that the J201 JFET/FET can be swapped with some of the 2Ns. I hope some of that is of some help. Cheers, D2o
|
|
|
Post by flateric on Jun 28, 2010 16:09:22 GMT -5
Thank you for taking the time to post such a detailed and linked thread!
|
|
|
Post by D2o on Jun 28, 2010 18:03:41 GMT -5
My pleasure, flateric.
I hope you will get further input from JohnH or sumgai - but I appreciate the appreciation!
Cheers, D2o
|
|
|
Post by JohnH on Jun 29, 2010 0:28:22 GMT -5
Id say that for small transistors, the most important parameters are; are they NPN or PNP, and are they silicon (or germanium, still favoured in some old stompbox designs). After that, the gain is the most interesting parameter, which may or may not be important to the design. usiually more gain is good.
J201 is an N-channel JFET, and there are other types that can often replace it, but biasing may be affected and overall gain in an amp circuit. I tend to use 2n5457 for all JFET circuits, adapting the design to suit.
|
|
|
Post by flateric on Jun 29, 2010 7:27:07 GMT -5
I guess I've already fallen into the trap of wanting it to be simpler than it really is. There was a circuit calling for a 5088 but someone had put in a J201 instead, then I needed a J201 in a different build but only had a 2n3904 which I've used in the past in place of a 5088, but the gain was all wrong. I guess while it can sometimes be ok to replace A with B, if you know B and C are similar and you don't have either, slotting in A instead of C does not always work. Does that make any sense?
Anyway, the good stuff: 5088 = 3904 pretty much most of the time. OK? 5087 = 3906 J201 = 5457 BS170=?? Any other common trannies to add to the list?
|
|
|
Post by D2o on Jun 29, 2010 9:45:10 GMT -5
From what I can tell, the most commonly referenced substitute for the BS170 seems to be the 2N7000.
All references have a notation that care must be taken to install it correctly, however, as the pin out is different (180^ / opposite of the BS170)
Cheers, D2o
|
|
|
Post by flateric on Jun 29, 2010 10:05:57 GMT -5
I see my post on DIYstomps has gathered a bit of momentum now, may be worth collating both sources of info together for anyone else's future reference.
Posted by TRJones1 (thanks!):
|
|
|
Post by D2o on Jun 29, 2010 10:10:19 GMT -5
Thanks for sharing this, Flateric - this is good information.
|
|
|
Post by sumgai on Jun 29, 2010 14:08:24 GMT -5
♭eric, I think Jones missed some of your point, in his response above. As I understand it, you don't have so much as a "tone issue" as an "availability issue", am I correct? There used to be a Master Cross-Reference, printed up and published for free, and available to everyone, not just the Engineers and such. That's pretty much down the tubes (it's still there, but no longer free), but thankfully it's the Internet to the rescue. Over the past several years I've collected links to such references, only to see them disappear when I'm not looking. So now it's not just the Internet, but the Internet's best friend, Google to the rescue. Just search on the terms "2Nxxx (or whatever the type is) substitution OR cross-reference", and you'll get more than you ever needed, or wanted, to know. Jones is not very far off-base at all, in his technical assertions. However, I think gain (often denoted as Beta or ß) is not necessarily the most important factor, I think noise would be of more concern here. But the telling factor, to my mind, will be what we call "slew rate". This factor measures how closely a device follows an incoming waveform, without distorting it in some fashion. Slew rate is measured in volts per microsecond. The more voltage a device is rated for, the quicker the response, and that's what we desire. More or less, it means that we can pass more juice through the device, in that same microsecond, and it will come out the other end with less chance of being distorted. Figures of 5v/µs used to be considered pretty good, but.... nowadays we can get upwards of 200v/µs. That's considered top quality, pretty much industrial or military class operation. What you're likely to find on the open market will probably be of a lower rating, and probably for less money. Thankfully. Look for a combination of all these factors that are most likely to suit your needs. Obviously, availability is at the top of the list, then pinout, then slewrate, then noise factor, then gain. If you can't learn one or more of these factors for the transistor in your hand, then just go with it. Well, obviously you'll need to have the pinout, but beyond that, you really can stretch most operating parameters quite a bit, and still come up with a decent sounding unit. HTH sumgai
|
|
|
Post by JohnH on Jun 29, 2010 15:46:34 GMT -5
Should probably strike that one. 2n5457 can often be used for the same job as j201, but it needs the circuit to be adjusted or redesigned, it is not a drop in replacement
|
|
|
Post by flateric on Jun 29, 2010 17:18:31 GMT -5
ep it is primarily an availability issue that i'm after a table of alternatives for future ref john.
|
|