Possible - they will have to dissipate 700-800mW, I decided to start with 9240 just to be on a safe side. But if 9610s will feel good thermally - it's a good alternative. I think, TL431 and zeners are also available in SMT...
Too bad there's no fast way to get a Mouser order in your area. I've been wanting to build this front end since I first saw it.
Yeah... But it's going to arrive pretty soon now 😉
Soldered the second pcb without 9610s while waiting. Also running the TH version in my system now - plays very nicely. Low phase shift rules 🙂
I've been looking into SMT complementary devices in a single package. These look promising. http://www.mouser.com/ds/2/308/BC846BPDW1T1-D-111237.pdf
Jeff, these ones look very good. I'll try to order some 846 ones.
If the transistors in each pair are really close in terms of key parameters, additional matching will not be required. Simplicity level increases 🙂
If the transistors in each pair are really close in terms of key parameters, additional matching will not be required. Simplicity level increases 🙂
If you have access to Newark/Farnell, they have a bunch of different ON Semi devices on sale right now. I'm stocking up.
Thanks for the info - I will be in Las Vegas 10-17 April (presenting at NAB Show) - going to use the occasion and arrange delivery to the hotel 😉
Rather interesting observation
I have compared the Shuntie's regulation efficiency with IRFP9240 and IRF610 transistors (in simulation).
Ingesting 50mv ripple at the input (70V rails), I have at the output:
- IRFP9240 - 400nV;
- IRF9610 - 1500nV.
The reason is the difference in their Rds(on):
- IRFP9240 = 0.38R (up to 0.5R);
- IRF9610 = up to 3R.
Well, both results are very good. But the first one is slightly better 🙂
I have compared the Shuntie's regulation efficiency with IRFP9240 and IRF610 transistors (in simulation).
Ingesting 50mv ripple at the input (70V rails), I have at the output:
- IRFP9240 - 400nV;
- IRF9610 - 1500nV.
The reason is the difference in their Rds(on):
- IRFP9240 = 0.38R (up to 0.5R);
- IRF9610 = up to 3R.
Well, both results are very good. But the first one is slightly better 🙂
That's interesting. Do you think the ripple could be reduced farther with a device like IPB025N08N3 or another low RDS device?
Though hard to come by nowadays, I wonder if anyone has thoughts of using Toshiba 2SK2013/2SJ313 in place of the IRFs for some more sonic benefit?
Possibly yes, but the level of 400nV from 50mV is already rather cool 🙂
It will also require certain topology changes, as those Infineon devices are all N channel...
Hi Sam, I'm really not sure there will be some benefit. IRFs work very well in these conditions.
Thank you for the feedback.
What changes would you recommend if the OPS is to be changed to N-channel Mosfets only? Thanks.
Nice job soldering the SMT devices. All oriented the same way and right side up, even. I would like to know more about the jelly flux. That sounds useful.
Do you have a good soldering iron for flowing solder around the SMT FETs?
--Russell
Do you have a good soldering iron for flowing solder around the SMT FETs?
--Russell
These have been available for a long time now. I like/try to use/search for parts in 50miil soic,sot-23, sod-123 packages, if they are avail. for passive R, i try to use 0805/1206/1210 etc for these types of designs. Susumu are nice thin film R parts to use.
i read that the onsemi tl431 has the lowest noise of all the mfgs.
Enjoy your trip and show/conference !! What are you presenting? It is the national assoc of Broadcasters?
Any problem with declaring parts with the immigration officers? Just hand them the load of paper they send with the orders, it is unbelievable how much paper they send with the orders.
A very small chisel tip works for sot-23. for d(2)pak, you can use a bigger hotter iron.i use a 10x jewelers loup for inspection
Last edited:
Hi Russell,
I use the German flux paste - see the picture. Something pretty similar in Mouser stock:
>NO CLEAN FLUX PASTE<
With regards to the FETs - good point. I think, I will have to remove a couple of the output plugs to be able to use some lower angle while soldering them, just to have more "massive" contact with the iron. I use a regulated-temperature iron, so normally, if I slightly increase it - it's enough for soldering such "areas".
Cheers,
Valery
Attachments
Good to know - thank you 😉
Yes, this is exactly the National Association of Broadcasters. I am a video-encoding equipment developer and manufacturer - noisypeak. Five international partners are showing our products - have to oversee how it goes and support them in the meetings when necessary. Actually, I was not going to participate NAB show this year - preparing to the next year show - but too many companies are willing to show our gear, so it was the "last minute" decision to go 🙂 Although, I like to be in Vegas from time to time - no complains 😛
I have also got a pretty good loup, the bigger iron - would be good to have for bigger parts, will probably buy one soon.
Cheers,
Valery
Attachments
Hi Sam, it's possible to modify the OPS for N-channel only - I will drop the schematic update - but it will slightly suffer from additional distortion - not too much though, as the front-end is a relatively high loop gain one.
Cheers,
Valery
- Status
- Not open for further replies.