Transnova-Schade OS/Amp

@ Planet10:

FHXL, it shall be then...let me see about ordering a pair of MS-11s...various SNAFUs...
:rolleyes:

Harry Olson in his classic book says a flattened pyramidal driver mount gives the best overall smooth response. I thought a hemi/spherical mounting took care of that. Do you think I could go for a "light pyramid" by having bevelled edges around the driver...probably a sub-baffle sort of thing?

My experience is that thick absorbent padding (please don't laugh-- my (experimental) favourite is thick Turkish towel --two layers-- than foam, which IMHO does not work...) Your insights and tips on this surely would benefit many builders.

Warm regards
--UKP
 
if post 10 schema has details, i can make a layout.
here is post 1.
 

Attachments

  • ezgif-5-396010f1a2.pdf-1.png
    ezgif-5-396010f1a2.pdf-1.png
    142.5 KB · Views: 368
If you mount PSU closer to the bottom of the heatsink you'll manage to put MOSFETs in lower half of the heatsink (say 5 inches from the bottom) and the heat transfer will work better
Actually the best position for heat sources is towards the top.
Think of the bottom position for example: The heat going past the fins will actually worsen the convection and block cold air going past the heatsink.
If we can ignore this natural convection and use a strong fan to force convection, the geometric middle would be best for even distribution.
I think there is nice thread about the subject here in the forums.
 
frugal-phile™
Joined 2001
Paid Member
@ Planet10:

Sir, my attempt is to minimize diffraction in the FHXL by putting a bevelled baffle to mount the driver...do comment on its advisability from your rich experience with the FHs...

You can go as crazy as you want with a supraBaffle, all the way to emulating a nautalis midrange pod superimposed on the top of the FHXL.

To get much bevel/roundover at the edge you will need a supraBaffle or add baffle and side material thickness.

dave
 
Measuring proves otherwise, but measuring is not a scholastic method, that's what peasants do...
I wasn't precise enough - I wrote "closer to the bottom" and that's relative... I should have said: at about one third of the heatsink's height, measured from the bottom.

Quodlibeta: Francis Bacon and Horse's Teeth


--

I may only say that, despite being realized not exactly as per the (correct) theory, my amp works more than satisfactorily:D

I hope that some other people could do the same that I did, some search in the forum, one question to the master (after some trials and errors), and it's iron time... Enjoy
 
Last edited:
Planet10: Thank you Sir for your valuable inputs. :lifesavr:

Guido, I second you there...and the proof of the pudding is, as they say, in the eating, especially for us humble solder slingers, for whom the pleasures of involved measurement remain as pleasant mirages far in the horizon! For the "peasant", the harvest is sweeter when it is the fruit of his/her own labours... :)

As for the heatsink issue, things become very clear when one has a tall heatsink (rectangular, with vertical fins) say, about 12 inches tall. All one needs to do is to mount a power device about one-third from one end, run it for a few minutes and take a look with an IR camera. Simply invert the heatsink and repeat. When the device is below the "belt line", things improve overall. Of course, the assumption is that the hot devices are ALL on the same level and the heatsink operates by pure convection.
 
could you please help for couple of questions
  • do we need the lateral mosfets to be matched or any set of one N and P channel lateral mosfet we may use?
  • can we use non lateral mosfets like IRF9520, IRF520 in this topology of common source schade feedback symmetric amplifier. may be for that we may need to add additional components to handle thermal runaway, unlike for lateral mosfets
 
Hi sarathssca,
  • matching is always beneficial but not very important in this case. You can use what you got - Hitachi, Renesas or Exicons are pretty much the same. Read carefully the post #1 and you'll know what to do.
  • Hexfets will work too - if you are aware of thermal requirements nothing will go wrong but you'll have to work harder to make them work and sound good.
 
  • Like
Reactions: 1 user
Hi sarathssca,
  • matching is always beneficial but not very important in this case. You can use what you got - Hitachi, Renesas or Exicons are pretty much the same. Read carefully the post #1 and you'll know what to do.
  • Hexfets will work too - if you are aware of thermal requirements nothing will go wrong but you'll have to work harder to make them work and sound good.
thanks for the reply..
I dont have much knowledge on audio amp design, just some basic idea on different modes of MOSFET usage (common source/drain/gate) etc.. (amps we see more often has multiple stages, number of mosfets/transistors and other passive components, and are very difficult to understand).. But this amplifier is looking very simple, little understandable and very tempting to try .. thanks for bringing up such nice simple designs.. :) ..
But lateral mosfets are getting very rare and costly , and i think currently only Exicon lateral mosfets are only available and need to be imported from UK.
Nice to hear same topology could be used with IRF mosfets which are more readily available and lesser cost too..