FirstWatt F1 Gathering materials

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
wayne325 said:


A delta of 34 is about as high as you want to go. That will
give you a temp of about 55C in the winter - approaching 60C
in the summer if you keep your apartment at 80F. Keep in mind
also those are free air figures and not stacked.

The other heatsink solution gave me a delta of 16 degrees but I had to give up too much back panel space and the connectors would have been jammed all together in a 4" by 5" area...

--Chris
 
Nelson Pass said:


The line capacitor should be specifically rated for AC line - it
will say so in the catalog and it will bear the appropriate
safety markings.

:cool:

I'm assuming the value isn't critical since its just here to filter high frequency noise from he AC line? I looked up the part number you used in your Pearl Phono design on pasdiy.com "Digikey part P4603-ND" .. But the closest they have in the panasonic Interference Suppression Metal film type is 0.0047uF.

I'm assuming the0.0047 will be fine so the BOM and shopping list is done! Now I just have to wait a week for all the stuff to get here :(

--Chris
 
Gents,

Many thanks to Mr. Pass, Wayne, Alain Dupont! I've placed my order for parts with PEX, Mouser, and Digikey. I've read Mr. Pass's article "Mosft Testing for Audio" (http://www.passdiy.com/pdf/mos.pdf) and I think I'm all set. I plan to make a breadboard with jumpers according to figure 5 in the article for matching my FETs. I'm assuming the bi-polars don't need to be matched? Anything else I need to know before I get started?

Thanks again guys, I think this is going to be one heck of an adventure ;)

--Chris
 
Thank you Nelson.

Here's the BOM I'm using with WIMA 1uf signal caps.

Sounds like I need to match the IRFP240's Q1 & Q2 to each other. Also match the IRFP9240's Q3 & Q4 to each other, and the ztx550's Q5 & Q6 don't need to be matched. The single Q7 IRFP240 current source does not need to be matched to anything.

Thank you for any oversight help.

-David

An externally hosted image should be here but it was not working when we last tested it.
 
Aren't IRFP P-channel devices possessed of some unwholesome type of distortion when used common drain?

The Fairchilds are rumored to better (by those who really know mosfets). There is a secret device number -- I think it is SFH9240 from Fairchild that is a good sub for the Vishay (aka IRFP).

JJ
 
Here's my thoughts based on my recent F1 clone build:

* I'd trouble sourcing the little PS C9 "interference" cap that goes across A/C IN so I just pulled one from one of the dead PC pwr supplies I had around.

* Even tho I'd purchased (supposedly) matched FETs, I needed R27*,R28* (RES 10 OHM 1W metal - ordered 10) supplementary resistors to attain < 50mv offset. Somewhere around are details about these, essentially you may need a couple paralleled on R27 OR R28 on each channel based on its direction of offset (*if* you find it > 50mv on initial build).
Even if your matching turns out better than mine did, I'd have them on hand.

* AFAIK, the 20V zeners are for static/hookup protection only, *think* I read somewhere that NP leaves them out of his personal builds and that removing unnecessary zeners is a good thing.... So I left them out of my build.

* Tweaks:Can't say about specific MOSFET devices... I used the ones in your BOM & they sound good to me:) Some have pushed building bridges out of "better" diodes VS using bridge rects (I've not gone there - used bridge rects in my builds).

Good luck!
 
Thanks for your guys help and ideas. I thought I'd try the tech-diy matched parts, since they are actually cheaper than if bought separately from Digi-key.

Cfcubed, thank you for sharing your issues and solution with matched parts. Would it be possible to PM me with the supplier you used?

I'm also building a pair of OB speakers using MOTH drivers I got from Nelson. The combination of the F1 and the new speakers will keep me out of trouble over the holidays.

-David
 
Nelson Pass said:
I used the Panasonics from Digikey.

I matched the devices to .01V, and the two P channel are
matched to the two N channel. If you match to .1V for Vgs
you can trim the Sources resistors to get the offset down to
the 50 mV spec without much difficulty. If necessary, I usually
trim the P channel Source resistors, as that has little effect on
the gain figure.

I would probably buy 2X to ensure matching, and the single
current source N channel can be the odd device out of match.

:cool:


Nelson Pass said:
1 = 2, and 3 = 4. Q5 is the single current source as I recall.

:cool:

I'm hoping to verify which Mosfets are matched.
* Q1 & Q2 gain devices matched
* Q3 & Q4 Upper CCS matched together

Do the gain mosfets need to be matched to upper CCS mosfets? Do all 4 devices need to be match to each other?

-David
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.