Mystery 'sj74'? jfets

Status
Not open for further replies.
Jump to Latest
#41
Other Jfets (J174)

Just for fun I tested a couple of J174's I received the other day...
Idss's -> 38mA, 30.81mA
Vp -> 5.9V, 6V
Vgs @5mA -> -4.17, -3.24
Vgs @10mA -> -3.39, -2.45

Giving me a (mathematically derived) Gm of 12.88, and 10.27 respectively.

Hope that's of interest - If I get excited about it I'll set up a jig and test the 50 or so I have and see what results I get.

Veni, vidi, er... I reported some data? I definitely don't understand latin. I apologise for my contribution to the effects of bad latin... 🙂
 
#42
Advice on 2SJ74BL Devices

I tested 10 2SJ74BL devices I recently acquired. Attached is the distribution of Idss from lockey's curve tracer.

Using the Yfs calculation method: (Idss(Vgs@0) - Idss(Vgs@0.1))/).1
8 of the Yfs values were in the 8mS to 11mS range. 2 were about 15mS. Toshiba's spec minimum is 8mS. Parts appear to be OK; they have the yellowish marking (no silver) and no sign of any impairment or funny stuff on the body. See attached photos. (sorry, a bit fuzzy)

Can anyone with some knowledge provide some input as to probable authenticity? The only thing in doubt is Yfs, so I'm curious if anyone else has done some testing and found "real" 2SJ74's to be at the low (minimum) end of the scale.

Thanks...
 

Attachments

  • DSC_3185.JPG
    DSC_3185.JPG
    286.2 KB · Views: 1,072
  • 2SJ74BL_Idss.png
    2SJ74BL_Idss.png
    27 KB · Views: 1,055
#47
NicMac said:
Using the same tracer and calculation my 2SJ74BL have much higher Yfs. See here.
Click to expand...

WalterW said:
I think they are fakes, with testing I have never found a real J74 that low on Yfs...
They are always easily around 25 to 40 mS.

Walter
Click to expand...


Gents,
Thanks for your responses. I concur with both. The reason: I have a pair of 2SJ74V that I know are genuine. These tested at 15mA Idss, with Yfs at 45 for both, well above the 22 typical. I also tested my LSJ74 A grade and these came in at 23 to 30 range.

This "sample" I received from a source could have been selected for similar Idss and could be much worse had I bought an actual lot of them.
 
#49
WalterW said:
I have also been foolish, to buy fakes 😡 2 times 😱
Click to expand...

I was very skeptical on these based on the Yfs measurement. You're right, all other Toshiba parts, and even the new LSJ74 were nowhere under 20mS.

It's hard to believe the greed for money that's out there. Someone said that FET's are going the way of tubes...I can understand that rationale based on these re-branders that are taking logic gate FETS and remarking them.

I do have one other source that is *guaranteeing* - along with their own test certification - some 2SJ74BL parts. Probably enough leftover to sell some. 😉

Will keep you posted.
 
#50
roger57 said:
I do have one other source that is *guaranteeing* - along with their own test certification - some 2SJ74BL parts. Probably enough leftover to sell some. 😉

Will keep you posted.
Click to expand...

Heard from the source. The price their supplier was asking is silly, no...actually outrageous. With new production from LIS, I don't see how they would sell them at the asking price. 😕
 
#51
jackinnj said:
Bogus J74s have noise densities well over 10nV/Rt Hz while originals should be under 1nV/Rt Hz. Often those bought on EBay are P-channel switches with relatively high noise.

You can test for noise -- there are many articles on the web describing low noise amplifiers -- ideally you'd like a 1000X very low noise amplifier like this: http://www.mh-audio.nl/Articles\LowNoisePreAmp.pdf
Click to expand...

Great article. Since I do PCB CAD design, this will be my next project. Thanks!
😎
 
#52
roger57 said:
Gents,
Thanks for your responses. I concur with both. The reason: I have a pair of 2SJ74V that I know are genuine. These tested at 15mA Idss, with Yfs at 45 for both, well above the 22 typical. I also tested my LSJ74 A grade and these came in at 23 to 30 range.

This "sample" I received from a source could have been selected for similar Idss and could be much worse had I bought an actual lot of them.
Click to expand...
Could you post details of Idss, & Vp for each so that Members can compare to their devices?
 
#53
I also have the Locky_Z curve tracer and the SJ74BL's I have range from 32 to 43Yfs. I have not managed to find legit SJ74's on eBay, but surprisingly there are some genuine K170's, 372's and 369's.

By the way, if you have fakes just file a complaint with eBay and you will likely receive a prompt refund without having to send them back.
 
Last edited:
#54
alazira said:
I also have the Locky_Z curve tracer and the SJ74BL's I have range from 32 to 43Yfs. I have not managed to find legit SJ74's on eBay, but surprisingly there are some genuine K170's, 372's and 369's.

By the way, if you have fakes just file a complaint with eBay and you will likely receive a prompt refund without having to send them back.
Click to expand...

Thanks for posting this information. It reaffirms what I have received are "legit" fakes (hmm.....had to think about that...oxymoron maybe?)

Any other postings I've seen support what you are saying, alazira - that Yfs values are in the range of Toshiba's published typical values, or higher. I was also able to confirm this with 2 genuine 2SJ74V devices. One had a Yfs of 43, the other about 50. Curves matched data sheet.

What gets me is that I received these 10 samples from a component supplier based out of Florida. After my first email exchange, I expressed my concern that there were bad devices around, but he assured me that his supplier was valid. (How many of you out there in DIY land have heard that before???)

I received the samples and the locky tester same day, so was able to do some testing over the weekend. After bringing back the news to the distributor, his response was...quote:
"Ok thanks for the update, so as I understand from the testing on one portion of function the parts we sent run on the lower levels of range and you usually see in middle of pack. We haven’t had problems before with product, but from your concern I understand..."
I have no idea what he said about "middle of the pack". I've never said anything to him about what I've seen because I've never had much experience with this device. However, I do have a bunch of genuine 2SK170 and 2SK117 I bought from Mouser years back, and they are absolutely in the Yfs ballpark from the data sheet.

The only other thing I noticed on these samples was that the plating on the leads was tight to the body, that is no copper showing. Although I'm not convinced it's a 100% verification, it seems most posts that discuss it say their valid devices are not plated all the way up, exposing some copper.

This locky_z tester is just fantastic for the asking price. :up: As someone pointed out, not a lab grade instrument, but if used from device to device as a reference, it's perfectly fine, ensure the "calibration" is checked with each pass. It is certainly good enough to verify that the spread of Idss (at the required Vgs) is good for a group of devices for a project.
 
Last edited:
#57
jackinnj said:
before you burn any copper let me check the letters to AX to make sure that the error in the schema was corrected.
Click to expand...

Normally, I would probably do a quick run & check 🙄on LTSpice, but in this case I would usually just go with it because of the source. Maybe I should change that thinking (had pretty good luck in the past...)

Thanks for the heads up 🙄
 
#58
alazira said:
.........................By the way, if you have fakes just file a complaint with eBay and you will likely receive a prompt refund without having to send them back.
Click to expand...
Have Ebay changed the rules for disputes?

That is not the way it used to work.

I have "won" a few disputed deals.
In all three cases Ebay said I must send back the mis-described goods at my cost to get a full refund of the purchase cost + original postage.

That leaves me many pounds (GBP) out of pocket.
 
Status
Not open for further replies.
Top Bottom