I'll do that in the next 3 days: post both the 2SJ74V and LSJ174 curves and numbers.
Cheers,
Gary
It is possible to get through that process, get your hard-earned cash back without sending back item. Case in point:
I bought a USB Micro to HDMI adapter from a China eeeeBay seller recently. Cost was $8 that included shipping. Great deal. It requires a small outboard power supply (of which I had the correct type) It's meant to plug into a Samsung Galaxy III phone, and voila! You have your screen on a bigger screen. All nice, but didn't work in any regard. Nothing but the blackness of space. I tried everything from menu options to searching the web for help, but found nothing except a few others that had a similar issue.
Within 4 days, of getting it, I messaged the seller telling them of the issue. They replied back saying they would refund me if I sent it back. Nice offer, but our ebay friends forget something. In the western world (and especially in the social republic of Canada-stanz), we don't have the luxury of subsidized post fees. To send that parcel from here back to China it would have cost me $14. That makes it a great bargain, doesn't it? The post office ends up at the top of the income pile.
I opened a dispute with Paypal and explained my side of the story. Now I gotta tell you that since I've had an eBay account, I have spent countless of thousands of $$$ on ebay, used to have an ebay store and used Paypal - so Paypal has been pretty good if I ever needed to open a dispute, and only do so when there's really a need. I explained what the issue was, and basically stated that if the decision didn't favor me that was OK. But if they did do that, how would they address a problem like this in the future if they expect to be a internet conveyance medium of transactions where these kinds of issues should be handled given the theme of a global store? We can't just stroll back and return the item. It requires costly 3rd party logistics that want their slice to move packages around. I said I was OK with losing that $8, but really needed to hear a plan from both eBay and Paypal on how this issue in future can be better handled and what their thoughts were. Someone's gonna lose coin, and who will it be?
2 weeks later, they sided with me, and the seller lost his $8. Believe it or not, I wasn't all that happy about the outcome. The seller may have had a genuine belief that his device worked, maybe on the phone he tested it with, and sent it out with good intent. I tossed it into recycling. I didn't get what I wanted except lost time in communicating with the seller, ebay and Paypal. In addition, Paypal spent $$$ in wages to deal with this, ultimately contributing to higher fees.
No winner here.
(apologies to the forum moderator for being way off topic here...it will NEVER happen again
)
Is it possible to build this and offer a GB for boards without any issues? Not likely since this appeared in a trade magazine? Anyone know?
This amplifier would be a great buddy for the locky_z tester.
Hi Andrew,
No, eBay hasn't changed their policies. However, in the case of bad SJ74's, as soon as I opened a case I received a refund without them asking for a return. I think if you say they are counterfeit as opposed to defective you will get a better response because eBay might remove their listing. Even if they initially ask for a return of the product, if you counter propose that THEY should pay for return shipping then they will probably drop it rather than receive poor feedback.
Some 2SJ74 Test Results
Last month I requested and received 10 samples from a ComponentsMax (an obsolete component vendor) who understood and believed his supplier had genuine devices. Below are some images of test results.
Image 1:
These are Idss curves from a pair of 2SJ74V devices I know are genuine. Note the sharp knee as the devices reach > 95% Idss.
Image 2:
The objective was to clearly display the Idss change at the "knee" between the 2 pairs of devices. The image text provides what I observed in comparing a pair of known good devices (LSJ74) against the poor "samples" I received from the vendor. Note: I did not use my 2SJ74V devices here as the Idss of those was close to 15mA on those devices and wanted to display devices that were closer in value. Although I indicated they don't meet the Toshiba specification, in fact they do, but at the very lowest value.
Image 3:
This is an image showing the calculated Yfs for one of the "sample" devices I received. The other tested device was identical. Although Toshiba's 2SJ74 data sheet says a minimum Yfs of 8mS, I have yet to test a known genuine device, or new manufactured device that is lower than 23mS.
Image 4:
Yfs for one of the 2SJ74V devices (as in Image 1)
Image 5:
This group of Idss curves is from a lot of 4 genuine Toshiba devices I purchased from DIYA member Tony Woods. I didn't ask Tony what test setup he used, but he included his test numbers. My absolute value was out by just 600uA, but the lot variance was identical to his numbers. I think this is important information for anyone needing to know how close (or how far) the locky_z tester to someone else's test data.
Last month I requested and received 10 samples from a ComponentsMax (an obsolete component vendor) who understood and believed his supplier had genuine devices. Below are some images of test results.
Image 1:
These are Idss curves from a pair of 2SJ74V devices I know are genuine. Note the sharp knee as the devices reach > 95% Idss.
Image 2:
The objective was to clearly display the Idss change at the "knee" between the 2 pairs of devices. The image text provides what I observed in comparing a pair of known good devices (LSJ74) against the poor "samples" I received from the vendor. Note: I did not use my 2SJ74V devices here as the Idss of those was close to 15mA on those devices and wanted to display devices that were closer in value. Although I indicated they don't meet the Toshiba specification, in fact they do, but at the very lowest value.
Image 3:
This is an image showing the calculated Yfs for one of the "sample" devices I received. The other tested device was identical. Although Toshiba's 2SJ74 data sheet says a minimum Yfs of 8mS, I have yet to test a known genuine device, or new manufactured device that is lower than 23mS.
Image 4:
Yfs for one of the 2SJ74V devices (as in Image 1)
Image 5:
This group of Idss curves is from a lot of 4 genuine Toshiba devices I purchased from DIYA member Tony Woods. I didn't ask Tony what test setup he used, but he included his test numbers. My absolute value was out by just 600uA, but the lot variance was identical to his numbers. I think this is important information for anyone needing to know how close (or how far) the locky_z tester to someone else's test data.
Attachments
Thanks for posting this info
Did you do the same Idss measurement comparing the "normal way" vs. locky_z for K170s?
I know there is the same kind of Idss "offset" for these devices when measured by the two methods, but I don't know if it has the same value (i.e. about 0.6 mA).
I have not done N-Channel JFET's, but could do some genuine 2SK170's and 2SK117's that I have.
This amplifier can be built quite easily on a piece of PCB material, "dead bug" style or "point to point" a la:
An externally hosted image should be here but it was not working when we last tested it.
Samuel Groner had an article in Linear Audio which described a very, very low noise amplifier using BF862's -- he also provides the Gerber files on his website.
ok. why not use the idss on the curve? They were genrated at Vgs = 0, so pick your Vds point and that's it. Unless you want Vgs at something other than 0?
Idss is when Vgs=0 and Vds=10V and Tj=25°C
Vp= pinch off voltage and is difficult or impossible to read off the standard Id vs Vgs graph, because the Id is ~10uA
The test for Vp can be carried out at the same time as an approximate Idss using the test jig posted by some of our Members.
It's the correlation between Vp and Idss that tells us we have a high gm device.
It's this high gm that the fakers can't fake.
All us poorly resourced amateurs can afford the very simple and cheap equipment to enable the Idss and Vp to be measured.
Vp= pinch off voltage and is difficult or impossible to read off the standard Id vs Vgs graph, because the Id is ~10uA
The test for Vp can be carried out at the same time as an approximate Idss using the test jig posted by some of our Members.
It's the correlation between Vp and Idss that tells us we have a high gm device.
It's this high gm that the fakers can't fake.
All us poorly resourced amateurs can afford the very simple and cheap equipment to enable the Idss and Vp to be measured.
Andrew,
Thanks for the informative post. The pinch region Vp being the point where the FET just switches on. It's an easy test to do with no equipment other than an good DMM.
I'll retest a pair of valid 2SJ74V, and a pair of "problem" devices, then repost those numbers.
Cheers.
That was a repeat of my post52 request that got passed by.
Here's one test jig that I referred. fig5 and the calculation of gm from Idss & Vp: gm = -2*Idss / Vp (the correction can be ignored if Vgs=0)
http://www.google.co.uk/url?sa=t&rc...eB8Mxi9TEXrJfCT1pnrgdhg&bvm=bv.60444564,d.ZGU
Probably worth searching for all of Borbely's fet papers. and W.Jung and etc....
Here's one test jig that I referred. fig5 and the calculation of gm from Idss & Vp: gm = -2*Idss / Vp (the correction can be ignored if Vgs=0)
http://www.google.co.uk/url?sa=t&rc...eB8Mxi9TEXrJfCT1pnrgdhg&bvm=bv.60444564,d.ZGU
Probably worth searching for all of Borbely's fet papers. and W.Jung and etc....
Last edited:
An idea: instead of using the right-click function in the software "copy to clipboard", just use the Windows snipping tool. This way, when you post the test, it's easy to see what the settings were. See the attached image of some devices I bought from eBay seller "polida". They are marked 2SJ74BL, but there's just no way they are BL grade, these are GR grade. There isn't a single device out of 8 pieces over 6mA, and they are grouped closer to 4mA. Yfs seems about right for GR grade as well, about 20 to 30. Can't imagine how these could be marked with a BL, given those measurements. I have genuine BL devices that measure exactly in the middle of the 6mA to 12mA range, Yfs about 40-ish.
The second capture is a closeup of the 0 to 0.1 to see the Idss to calculate Yfs.
By the way, I bought some 2SK170BL devices as well from ebay seller polida, those were genuine and tested perfect in the center of the BL range, about 8.5mA to 10.5mA.
Another way to tell suspicious (possibly re-branded) devices: under a good desk magnifier, notice the face of the device where the printing is; it seems to be a shade lighter than the rest of the device's body, as if it were sanded off and then printed on. I have several genuine Toshiba devices and the gloss of the black face of the device is consistent with the rest of the device.
Attachments
Gentlemen, thank you for your suggestions.
I traced the same part again, as well as a second one I have.
If I'm interpreting things correctly, the idss and Yfs
of the two parts are ~12mA, ~45mS and ~16.5mA and ~49mS
respectively. These seem consistent with the V grade marking.
Roger, I'd like to think your "BL" parts with the low Idss are
either badly out of spec or relabeled 2sj74 GR. If they are not
2sj74GRs they certain mimic them well.
Thanks,
Dennis
I traced the same part again, as well as a second one I have.
If I'm interpreting things correctly, the idss and Yfs
of the two parts are ~12mA, ~45mS and ~16.5mA and ~49mS
respectively. These seem consistent with the V grade marking.
Roger, I'd like to think your "BL" parts with the low Idss are
either badly out of spec or relabeled 2sj74 GR. If they are not
2sj74GRs they certain mimic them well.
Thanks,
Dennis
Attachments
Yes...as I pointed out, the curves seem to be for GR devices, not BL.
I've notified the seller and awaiting their response.
I also have V spec devices whose Idss at VGs = 0 @ -10Vds is about 17mA.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.