MMBF5457’s and where they at?

Very interesting and wide-ranging discussion. As noted, tubes are still being made and new tech to create other kinds of tubes.

For that matter, vinyl records outsold CDs in 2021 (admittedly because CD sales have dropped to give way to digital music, but vinyl sales are increasing year to year and that is driving some new turntable and cartridge business). I thought the efforts to use new tech to improve vinyl playback were interesting (scratch filters, rumble filters, some of the dbx tech that added dynamic range -- apparently it is very hard to nearly impossible now to find Dolby encoding /decoding chips for cassette tapes They went out of production years ago.) Maybe some of those concepts will come into play for digital modeling of vinyl playback as well. The interplay of old tech and new tech still makes some interesting hobby niches along the way.

Seems likely there is enough NOS on obsolete parts in the world to keep the boutique pedal biz going for a while.
 
zgrav said:
For that matter, vinyl records outsold CDs in 2021
Click to expand...

This surprised me so much to read. My parents still buy and use CDs and I would have assumed that there were enough people like them that would have a hard time learning spotify/apple music, let alone setting it up in a stereo/car, etc etc. Mind blowing! And very cool to hear about the vinyl tech.

I’ve tried to be better about keeping up with new/upgraded IC releases for basically the same reason (and I’m a dork who finds it interesting). Ordered a relatively new analog freq divider to see if it can do anywhere near decent suboctaves. Worst case scenario will be fun :)

Last few posts in this thread are making me optimistic again :)

Bummer to hear about the jfets @DAJE. Vaxny’s were, until this, my go to for avoiding the 5457 minefield on Ali. :/
 
Last edited:
Regarding semiconductors, it's important to consider the trend of consolidation with fabs over the past three or four decades (NB: this is particularly applicable to ICs and not necessarily discrete devices). It used to be commonplace for semiconductor companies to have in-house die fabs to test prototypes and conduct tape-outs. Companies also used to have domestic (CONUS) fabs that could produce at scale. I'm not sure of any company that still has one. Some still have overseas fabs that are company-owned. There have been talks of a new Intel fab going to Ohio and a TI fab in Texas, but those are still years out and a reaction to the current fabrication issues. Niche companies may have their own in-house fabs (e.g., InterFET) or have a business model dedicated to legacy support (e.g., Central Semiconductor), but there has to be a viable market for the parts in order for them to be made at scale and made affordable. These companies have many offerings for analog front end blocks of new products. They have legacy portfolios to supply legacy needs. Central Semi produced Ge devices through the 1980s. This is almost assuredly due to big industrial/military/government customers with legacy equipment. The fact that they make legacy parts doesn't automatically mean that they will be made available for retail customers or wide release.

Companies dedicated to (discrete) analogue parts are rarer and rarer. The general trend in the industry is consolidation. So, companies are acquired and their portfolios are folded into the the purchasing company's IP. This can—and often does—result in (near-)monopolies on certain parts. For parts like the J201 with limited demand and vanishing commercial application, production is de-prioritized for parts with higher demand.

COVID-19 has affected silicon supply chains for myriad reasons, one of which is that there are a handful of giant fabs that handle the silicon for all companies. Another issue is raw material availability. Chip development is iterative. When you have long lead times and increasing demand, it becomes harder and harder to work on your new products. Since those new products are often the ones that are exciting for investment purposes, it's paramount that you be able to get those developed quickly and have some sort of public intro. When that pressure is raised and development cycles are condensed, you have to prioritize what gets produced. Semi companies are prioritizing new development that can establish footholds and contracts in new areas of innovation. This can lead to shortages in other parts if they are de-prioritizing extant portfolio items in favor of new development.

I don't know how appropriate it is to compare discrete JFETs to vinyl. Vinyl is a big part of overall music sales and has a wide customer base. It's been proven that there is a demand for it (so much so that many indie labels/bands are unable to secure spots at pressing facilities). Discrete JFETs don't have an analogous market space. So, while I don't think that we're looking at an extinction of the parts (either in through-hole or surface-mount packages), I think that this particular niche has an outsized need for them while the market demand is too low to really care from a business standpoint.
 
DAJE said:
I recieved 18 (paid for 20) and I've tested a few. Can anyone tell me what numbers I should be looking at here?

Given my luck with pedals lately, I'm assuming they're useless.

EDIT: I compared them to a couple of MMBF5457s I have, and the numbers are way too low. I don't think they're fakes, but I think they're out-of-spec rejects. They only cost a few bucks, so if I can't get a refund it's no big deal, but I would strongly advise against buying from that seller. I have had good results before, but not this time.
Click to expand...
I recently received a few jfets: 4 2n5457s and 4 J201s from GuitarPCB, and 1 2n5458 from Tayda.

To get nerdy with it, I wanted to measure and compare to those bought back in 2013. I don't have a meter that measures transistors, so I used the technique found at the end of the article here: A closer look at the Fetzer Valve (runoffgroove.com)

I find the J201s and 2n5458s to be right in the range that ROG measures (the J201s averaged 0.71mA and -0.98V, and the 2n5458 was 4.37mA and -2.62V).

But 2n5457s were out of range, averaging at 1.87mA and -0.43V. Usually 2n5457s should have Vp around -1.5V. They are very consistent though, all very close to each other, compared to older batches which had more standard deviation.

I also compared my measuring technique against some old ones, to see if I was consistent with previous measurements and found that it was.

So I'm guessing while manufacturers are still doing some runs of 2n5457s, the don't seem to be using the recipe they used to.
 
I have some I've been somewhat hanging on to for prototypes, but realistically I don't think I'll ever use all of these, like ever...

I'll make them available to those who need them soon.
 
Kroars said:
Man!! Put 500 in my cart last night, went to finish up today and they’re gone…. There was 2,000+ last night, figured I’d have a few hours. Thanks for the heads up nonetheless!!
Click to expand...
Quest Electronics still has them in stock.

I bought some not too long ago. These these are the LT1-suffix ones.

Anyone know where I could find some MMBFJ201's? I haven't seen any of those anywhere.
 
You must log in or register to reply here.

Similar threads

Fingolfen
Three Scoops of Ice Scream...
Replies
8
Views
1K
Fingolfen
Fingolfen
Back
Top