Andrew I saw your comments in almost every similar discussion. May be you can help me understand this. Are you saying even 5% IDss matching become unnecessary if we parallel more than 2 Jfets?
I have regurgitated what I have seen reported.
You can read the same in the Forum.
I have not paralleled jFETs.
You can read the same in the Forum.
I have not paralleled jFETs.
Last week I managed to acquire 53 BF862s from a local distributor and tested their Vgs-Id performance.
Good news is that 44 of them could be matched within 0.5% of Idss, although there is very few perfect pair, say within 0.05% Hfe deviation for modern bjts.
Bad news is that their Idss scattered from 14mA to 25mA just as the datasheet described. In order to get as many of them matched as possible, one has to buy and test a lot of devices(compared to bjts). And testing those sot23 package is really painful.....simply too small to handle🙁
I'm using BF862 as sort of on board preamp for a bjt input headphone amp😉hope it'll work
Good news is that 44 of them could be matched within 0.5% of Idss, although there is very few perfect pair, say within 0.05% Hfe deviation for modern bjts.
Bad news is that their Idss scattered from 14mA to 25mA just as the datasheet described. In order to get as many of them matched as possible, one has to buy and test a lot of devices(compared to bjts). And testing those sot23 package is really painful.....simply too small to handle🙁
I'm using BF862 as sort of on board preamp for a bjt input headphone amp😉hope it'll work
Attachments
Cool load it up, good luck n share results please
Sent from my iPhone using Tapatalk Pro
Here is a jig I made to test BF862 Idss in-situ. The jig is the same as the actual circuit the JFET will go into so the values here represent real values in the actual circuit. It is painful but wear magnifier goggles, use good tweezers, use an eraser from a pencil to apply pressure to the JFET on the jig to get good contact. The painful part is waiting 20 sec for it to reach steady state.
Here is a jig I made to test BF862 Idss in-situ. The jig is the same as the actual circuit the JFET will go into so the values here represent real values in the actual circuit. It is painful but wear magnifier goggles, use good tweezers, use an eraser from a pencil to apply pressure to the JFET on the jig to get good contact. The painful part is waiting 20 sec for it to reach steady state.
Nice work!
When i was buying components at the distributor, the guy actually recommended me a fixture for durability test of SMT parts. It contains spring loaded pins to contact the part being tested, so all you need to do is put the part in and close the cover. Saved a LOT of time for me😀
Attachments
There was a link a couple of months ago when I suggested that a 3d printer could do the job for us and maybe a lot cheaper than the prices I was seeing at that time.
SOT23 Tester
Dear xrk971:
I bought this one:
NEW SOT23 SOT23-6 SOT23-6L IC Test Socket / Programmer Adapter / Burn-in Socket | eBay
They had a soldering error between their two layers of boards. Not a design fIaw, just sloppy soldering on my sample. I cut off the lower board and soldered some wires to the board with the adapter.
It is still fiddly to put in the SOT23 JFETs in a way that you get good contact ... 😛
The tester holds two SOT23 devices, and you can use it to differentially match two SOT23 JFETs.
OT: I use it to test MMBFJ271 that I want to use for the differential pair of my Aleph "Minimax".
Something like that goes nicely with the SOT23 JFETs:
NEW SOT23 SOT23-6 SOT23-6L IC Test Socket / Programmer Adapter / Burn-in Socket | eBay
The left adapter board in the picture holds two SOT23. I plan to use it to make a differential pair, closely coupled, and put it in place of the usual TO92 or TO220 devices ... 😉
Best regards,
Claas
Dear xrk971:
I bought this one:
NEW SOT23 SOT23-6 SOT23-6L IC Test Socket / Programmer Adapter / Burn-in Socket | eBay
They had a soldering error between their two layers of boards. Not a design fIaw, just sloppy soldering on my sample. I cut off the lower board and soldered some wires to the board with the adapter.
It is still fiddly to put in the SOT23 JFETs in a way that you get good contact ... 😛
The tester holds two SOT23 devices, and you can use it to differentially match two SOT23 JFETs.
OT: I use it to test MMBFJ271 that I want to use for the differential pair of my Aleph "Minimax".
Something like that goes nicely with the SOT23 JFETs:
NEW SOT23 SOT23-6 SOT23-6L IC Test Socket / Programmer Adapter / Burn-in Socket | eBay
The left adapter board in the picture holds two SOT23. I plan to use it to make a differential pair, closely coupled, and put it in place of the usual TO92 or TO220 devices ... 😉
Best regards,
Claas
Most modern audio input stages use a differential pair like this:
An externally hosted image should be here but it was not working when we last tested it.
From ESP website Current Sources, Sinks and Mirrors in Audio
This type of input reduces even order harmonic distortion, but is widely used in many amps and opamps because it reduces common mode interference.
Sometimes called a Long Tail Pair (LTP).
Not yet. Figured out the circuit with just BF862 and it is shared in another thread. If you are looking for a PCB have to wait bit more.
Answer is yes. You have to use lower voltage to be on the safe side as BF862 can only handle 300Milli watt and if you happen to pick high IDss ones (above 15ma) it won't be a good idea to run with 18 V supply .
- Status
- Not open for further replies.