The MONGREL (supersym II)

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Member
Joined 2006
Paid Member
Man, fall is beautiful (last PIX) .. the "Doghouse" is in a monster tree forest... :)
OS

1. Cover up speaker with a bit of camo and pile on some leaves.
2. Run some wires to amp and put on spooky sounds such as things growling or moaning.
3. Scare the dirt out of little trick or treater's :eek::mischiev::rofl::Pumpkin:!!!

WARNING - if Mama finds speaker you can kiss it good-bye!
If Mama finds you - RUN!!!!!
 

Attachments

  • Trick_or_Treater.jpg
    Trick_or_Treater.jpg
    21.3 KB · Views: 490
Last edited:
Ex-Moderator R.I.P.
Joined 2005
Is there any simple way to clamp outputs to a heatsink using ordinary household materials?

if you use alu bars, or similar, you could use ordinary metal screws, and do without other tricky tools
just dont mount through the devices, but place the screws beside them
try on a small piece of alu first, to see how big the holes need to be for the chosen screws
its important not to break a screw
or it could also get stuck halfway, with a damaged head
torx screws might be the better choise

btw, drilling and tapping in aluminium is best done using household spirit, and plenty of it
dip the drill or tap into the spirit
and be very careful
those small drills and taps breaks very easily
and the bits will sit in the alu forever
slowly drilling
and the tapping is carefully back and forth, bit by bit
and by hand, it takes practice to get it perfectly straight
and remember to carfully level the surface afterwards
wont hurt to level the back of devices too
hold it up against the light, and its easy to see how it is
 
PCB file for the PB120LP v2.0

Hi OS,

I have incorporated all the improvements that you suggested earlier. I put together the PDF file for the PB120LP.

Can you please put the attached file onto your web site as this version is an improved version of the PB120LP-45V.

Cheers, Stanley
 

Attachments

  • Mongrel-PB120LP.pdf
    96.5 KB · Views: 223
Hi OS,

I have incorporated all the improvements that you suggested earlier. I put together the PDF file for the PB120LP.

Can you please put the attached file onto your web site as this version is an improved version of the PB120LP-45V.

Cheers, Stanley

VERY nice. It's on my site. I DL'ed KICAD. All you do is draw out schema and it makes the board?? how do you import new components into the library. I have a lot to learn.
Below is new creation (BX) to add to "collection"
OS

OS
 

Attachments

  • BX.jpg
    BX.jpg
    155.2 KB · Views: 443
EX .. LEECH meets hawksford

Thanks a lot, Krisfr .. that is a lot of 3D "candy". I actually like that the library is generic since I do this primitively anyways. :cool:

I HAVE decided , my november amp will be a fully complimentary design. The EX (attachment 1) has the FASTEST slew of them all (way above 120 V/us). That might not matter .. but what the hell !!! I am using a modified Hawksford similar to my GX but it has a beta enhancement feature like was discussed in the goldmund threads. It can be thought of as 2 very fast "BLAMELESS" amplifiers symmetrically opposed. It is very linear , LOW THD ... .0005% 1k / .008% THD20 @40v p-p.

It is below in LT form , too. Not too many leech type projects around here , why not create one.. a BETTER one. :)
OS
 

Attachments

  • EX.gif
    EX.gif
    24.9 KB · Views: 432
  • mongrel_EX1.0.zip
    8.3 KB · Views: 113
My PB250 has the resistor pads for a passive EF3 , so I can still do that. The EF2 is flawless in a high negative feedback power amp. The only 2 situations where a triple or CFP would be needed would be for many outputs or a non-switching AB setup. I experimented and implemented a EF 3 in my "trouble with triples" thread to augment the performance of the triple.

The sound quality I hear now with the luxman/hawksford is close to absolute perfection. My only purpose to build the EX is that it is the "better"luxman (more expensive and is a lateral OPS amp. My EX could do almost 200v/us , so I'm curious ... and it's just going to cost me another $20 , thats all. The CSS EF's do sound a little "tighter" with drums but this can be compensated witha better faster voltage stage.

OS

PS: by next year ,I should be up to a "ZX" or close to it. Then I will do a subjective / objective "shootout" between all these class AB designs. I am trying to find an analyzer (real one).
 

Attachments

  • LUXMAN L58.gif
    LUXMAN L58.gif
    25.1 KB · Views: 427
Last edited:
OS,

You could replace the differentials with a comp, pair and take the feedback to the sources and have a decent gain in bandwith.

Something like I designed here but I would keep the VAS cascoged like you have

Jam
 

Attachments

  • 6-3-04 LAST SCHEMATIC1 - PAGE1.pdf
    53.4 KB · Views: 232
Last edited:
OS,

You could replace the differentials with a comp, pair and take the feedback to the sources and have a decent gain in bandwith.

Something like I designed here but I would keep the VAS cascoged like you have

Jam

Where are the transistors ??:headshot::D I want to try some fet's , but not ALL of them !!! :eek: Paralleled/single Jfets in a cascoded original would be the next logical step forward. (EX + cascode + 8 Jfets) The original's hawksford VAS LOVES to run at 12mA + as well , making it a better candidate to drive laterals direct ... at least 2 pair. Another benefit is the EX also is high loop gain 70db ... it would also do well fet'ed. :) Thus is born the "FX".
EX- 75 X 100mm and FX- 100 X 100mm board.
OS
 
OS,

I also ment to say replace the front end Mosfets with bipolars, Sorry I did not mean to put you into shock. :D

The output stage would be up to you. I ran the original VAS (mosfet) at 40 mls.:redhot:

Regards,

Jam

P.S. In my younger days I put Mosfets everywhere..........:eek:
 
Last edited:
OH ...yeahhh. FX it is. .003% THD20 in all it's "fet glory".. :cool: (below) FET's at 3.05ma each...

PS - I notice it takes a lot less current swing for the N-channel FET's than the P- channels , BJT p-n pairs are MUCH closer (ksa/c's)
(attachment 2)
OS
 

Attachments

  • FX.gif
    FX.gif
    30.7 KB · Views: 574
  • FX THD20.gif
    FX THD20.gif
    43.7 KB · Views: 555
Last edited:
Yes, but when you replace R20 and R21 with current sources (or current diodes) you improve the performance of of the original current sources by about 20dB, I believe. Something to do with better rejection of the rails.....

Jam
 
Last edited:
Yes, but when you replace R20 and R21 with current sources you improve the performance of of the original current sources by about 20dB, I believe. Something to do with better rejection of the rails.....

Jam

But my rails are "squeaky clean" , either cap multipliers or full blown shunted separates , so this would not matter. A simple improvement would be the 2 device style CCS for those without these "enhancements". In reality , I get about 5-10mv ripple when slamming my current GX w/multipliers. Sounds like a dual monoblock.
OS
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.