Hafler DH-200/220 Mods

Burning Amp 2016 Hafler DH220C Presentation

Some folks expressed interest in seeing the slides from my Burning Amp 2016 presentation of the Hafler DH220C revamp using a full-complementary JFET input stage with a floating tail. It is a powerpoint file, and, unfortunately, it is over 7 MB, even zipped (zip compression of ppt files is negligible). I don't know of any easy way to post it here. Please email me if you would like a copy.

Cheers,
Bob
 
Last edited:
Thanks mlloyd1, this is what it says at their site:
 

Attachments

  • Capture.PNG
    Capture.PNG
    25.8 KB · Views: 456
Thanks mlloyd1, this is what it says at their site:

For the relevant types 2SK1056/1057/1058 and 2SJ160/161/162 (TO3p replacements for the Hitachis used in Hafler DH-120/200/220) their site still states mass production. Also Profusion/Exicon offer the laterals in different cases (TO-3, TO247, TO-264) as single or double die version, even as selected parts.

So all possibilites are open for a direct replacement of the PC-6/PC-19 pcbs in the Hafler or a complete new design of pcb. For me the free wiring of the FETs in the DHs was a little suspicious, a bettter way will be to integrate the output stage on a pcb.

BR and Happy New Year for all DIYers

Günther

PS: Special thanks and :up::up: for Bob Cordell, that he keeps the good old Haflers alive
 
that's a good question.
Magnatec says theirs are still available: TT electronics Semelab Limited - Magnatec Datasheets
Exicon says theirs are still avilable: Lateral MOSFET | Profusion Audio Design and Exicon Lateral MOSFETS
I've heard nothing back (yet) from Renesas, the grandadddy of all, whose website claims they are still "In Mass Production", but who knows if that's real https://www.renesas.com/en-us/products/power-mosfets/mosfets-for-amplifier.html

mlloyd1

Please keep me posted as to availability, since I want to update that kind of information in my second edition.

Cheers,
Bob
 
I am currently running a 3pair of the double die versions at 200mA each for a total bias of 600mA and the bigger heatsink is ~50°C when Ta=20°C.
Not sure what will happen if I get a good Summer.
It sounds good, but with a very different input stage (ASTX 14 PCB).

I have already tried to adjust for higher bias and after temperature of heatsink (or mosfet) increases the bias is then automatically reduced because of the temperature coefficient of the lateral Mosfet...
However, with only 100ma per Mosfet you may not be in the good temperature coefficient region....

I think you have not shown much of your diy work so far thus would be interesting to see this one...:)
Can you post a link for the ASTX 14 pcb.

Fab
 
Last edited:
Yes, the latFETs do have a negative tempco once above their critical current which is typically around 200mA for these power LatFETs.
This allows the stage to self correct for small changes in heatsink temperature.
I find they benefit from being Vgs selected when using paralleled sets.
And I find they definitely benefit from having some value of source resistor to give a bit of feedback to help with that correction.

I see that the Hafler and Cordell's version both do without source resistors, relying on the near zero tempco to attempt matching of the operating currents.
 
ltspice of Bob's mods

Hello folks,

yesterday was a snow day, so I drew up Bob's mod's in ltspice. I am not finished yet, I need to figure out the mosfet models. maybe someone can help me out :)
also missing is the cfp bias gen, but got it going with a bjt ops. will post it here for everyone. it uses Bob's models, models from LS. I zipped up my libraries, so you can put them in
C:\Program Files\LTC\LTspiceIV\lib\cmp
Add the pdf, maybe some do not know ltspice
Enjoy.
 

Attachments

  • Cordell DH-220C _mod.asc
    20.6 KB · Views: 123
  • cmp.zip
    313.8 KB · Views: 106
  • Cordell DH-220C _mod.pdf
    19.9 KB · Views: 243
Last edited:
For the relevant types 2SK1056/1057/1058 and 2SJ160/161/162 (TO3p replacements for the Hitachis used in Hafler DH-120/200/220) their site still states mass production. Also Profusion/Exicon offer the laterals in different cases (TO-3, TO247, TO-264) as single or double die version, even as selected parts.

So all possibilites are open for a direct replacement of the PC-6/PC-19 pcbs in the Hafler or a complete new design of pcb. For me the free wiring of the FETs in the DHs was a little suspicious, a bettter way will be to integrate the output stage on a pcb.

BR and Happy New Year for all DIYers

Günther

PS: Special thanks and :up::up: for Bob Cordell, that he keeps the good old Haflers alive

Thanks, great news!