The attached F3 schematic indicates that the bias conditions are Vds 2.4v and Ids 1.5A. I plan to perform a Vgs-Id curve at Vds=2.4V. I can obtain transconductance and Rds while performing the sweep. The measured parameters will be Vgs, Vpinch, Rds and transconductance at Vds=2.4V and Ids=1.5A.
We shall see what the population distribution looks like after some numbers of devices are finished characterization.
My suggestion is for the group buy to acquire 200pcs to 300pcs from Mr. Pass which should yield 100 to 150 matched pairs. If any are left over, we can send them back to Mr. Pass or send them to the DIY store.
For those who want to play at home, you can buy the LU1014 from an eBay seller and I will enumerate how to do the matching at home. If you want to know which eBay seller, just send me a private message.
This will not happen overnight. It will be a process to do the matching and refine the procedure for mass consumption.
We shall see what the population distribution looks like after some numbers of devices are finished characterization.
My suggestion is for the group buy to acquire 200pcs to 300pcs from Mr. Pass which should yield 100 to 150 matched pairs. If any are left over, we can send them back to Mr. Pass or send them to the DIY store.
For those who want to play at home, you can buy the LU1014 from an eBay seller and I will enumerate how to do the matching at home. If you want to know which eBay seller, just send me a private message.
This will not happen overnight. It will be a process to do the matching and refine the procedure for mass consumption.
Attachments
Last edited:
why are we so hung up on reflowing low temp solder? if two side holes are provided why not put a clamp across the LU and just use a thermally conductive paste to the metal substrate? there may be issues with embrittlement on that solder layer.
soldering - Any drawbacks to "low temp" lead-free solder paste? - Electrical Engineering Stack Exchange
great idea to synch the GB of matched fets and the adapter board.
soldering - Any drawbacks to "low temp" lead-free solder paste? - Electrical Engineering Stack Exchange
great idea to synch the GB of matched fets and the adapter board.
Last edited:
A big thanks to JPS64 for making the design for LU1014D to TO247 adapter in IMS. Full manufacturing data is availble here. I will order a few to make a verification build and wg45 will then provide them as part of this GB as an option.
I am curious about the thickness of the assembly vs. the thickness of an ordinary TO247. Is there a dimensioned drawing of the LU adapter assembly?
What about a through-hole mounting of the 3-pin header? A surface mount header may be difficult to align so that it is straight enough.
We want the PCB to be flat and smooth on the bottom for good heat transfer to the heatsink. So through hole is not possible if we want it smooth on the bottom. Note that the metal PCB is conductive and holes and vias in general, are to be made conductive to the substrate. Most IMS PCBs are strictly SMT. The soldering of the 3-pin header should not be hard using paste and a hot plate. Once it melts use a tweezer to nudge it around to align with the pads. Then cool it to lock in place. The pads have solder mask and are well defined.
The IMS board is between 1.6mm and 2mm thick. The SMT header is a standard part - you can look it up with the part number. JPS64 might have the part number handy.
The dimensioned drawing is in the manufacturing zip file I believe.
The pin headers use 1.14mm square pins spaced 0.200in apart.
The IMS board is between 1.6mm and 2mm thick. The SMT header is a standard part - you can look it up with the part number. JPS64 might have the part number handy.
The dimensioned drawing is in the manufacturing zip file I believe.
The pin headers use 1.14mm square pins spaced 0.200in apart.
Last edited:
Excellent info. Thank you. How about a soldering jig for holding the header pins in place while soldering?
I appreciate your experience, but in my real job I manage a world class mechanical property testing lab. My personal experience with H20E is that it becomes extremely brittle at room temperature after a small number of thermal cycles within it's datasheet operating temperature.
Worth noting: I love my F3.
It's the best amplifier I've ever built and I keep coming back to it. I found a nice sweet spot in the set-up with just enough feedback, and it just sings.
The power supply has gotten noisy now, and I'm a bit upset with myself that I re-used some old EI transformers that have developed a ton of mechanical hum that is now coming out of the speakers.
It's the best amplifier I've ever built and I keep coming back to it. I found a nice sweet spot in the set-up with just enough feedback, and it just sings.
The power supply has gotten noisy now, and I'm a bit upset with myself that I re-used some old EI transformers that have developed a ton of mechanical hum that is now coming out of the speakers.
I am getting started creating the setup for matching devices. Attached are family of curves and transconductance measurements for one of the devices in my personal stash. I had to resort to 6-wire connection to make the D-S measurement 4-wire Kelvin.
I can see that I need to measure beyond 1.5A which will cause a re-stacking of instruments and recabling.
I can see that I need to measure beyond 1.5A which will cause a re-stacking of instruments and recabling.
Attachments
Excuse me if I stepped on your toes. It didn´t sound as you have experience in the field.
And thanks a lot for your view. It´s very intriguing to me as I dare say that "half of our company is built on H20E" and that epoxy has seen many qualification tests over the years.
No sign of brittleness at all. Only problem we sometimes have is silver migration but only under extreme enviromental circumstances.
And sure nowadays, there is a much bigger choice in epoxys. Especially when it comes to thermal resistance. At the time H20E was the best compromise for us, AFAIK priority was electrical conductivity.
The GB sign up sheet should go live next week, on this thread.
The same posting will have details on cost. Payment will be through Paypal.
I'm guessing it will take a few weeks to match and ship the Power Jfets. Carsten will handle distribution to the EU. I'll handle shipping to the Americas and rest of the world.