diyAB Amp - The "Honey Badger"

Cool! So it's quite easy to build it as an integrated amp!

You might want to lower the global gain a bit for an I-amp.

My DIYA boards are going to grace just such a unit.

(below 1-2) is the output for a 30db gain Badger.
Schema is updated to reflect the exact V2.4 PCB.

(Attached) is the V2.4 LT(.asc), preset to show 10K THD @ 62ma OP device current.
Comments are included for gain,bode plot - add AC1 to power supply(s)
for PSRR (and more). :)

PS (edit) R3 , R6 =27K - lowers the gain....
OS
 

Attachments

  • newbadgerLT.jpg
    newbadgerLT.jpg
    157.5 KB · Views: 1,120
  • badgerTHD.jpg
    badgerTHD.jpg
    80.5 KB · Views: 1,091
  • DIYA-amp-LT.asc
    20.4 KB · Views: 151
  • DIYAamp.txt
    39.6 KB · Views: 171
OS,

I think an error has crept into your drawing: it looks that point R (where Q1-2 emitter resistors meet R14) is connected to the ground.
Also optional anti-saturation diode is not shown. It might be worth showing that optional connection in different colour on v2.4 circuit diagram.

Could you explain to me please how C5 optimum value (here 47uF) is determined? That part of my old notes got lost. Thank you.

cheers,
 
Hi OS,

I'm going to build a pair of Badgers to drive these : Statements

I'm planning to build dual mono with independent supplies for the front end (which means four toroids in total - fun!). Would 35-0-35 secondaries be okay for these speakers?


I'm also planning to build a subwoofer using ESP's Project 71 (Linkwitz Transform Subwoofer Equaliser). You've mentioned a BassBadger that you're working on. When it is eventually released into the wild, I would like to use that to power the sub. :)
 
ImmortalZ 35VAC for the outputs ought to give you plenty of power into 4 ohm Statements and help keep your outputs in their safe operating area. Probably a good idea to use MJL outputs to get the biggest package for the best heat transfer, too. That's what you should be most concerned about with low impedance loads.

My guess is a suitable Bass Badger will use MJL4281/4302 outputs and 70+V rails to allow 400W or more into 4R loads. OS was talking about a 2R load IIRC so he likely won't go quite so high with the rails. Horses for courses.

Use the Linkwitz transform judiciously. Don't try to make your driver go too far. It's ugly when an amp clips. Uglier when you bottom a voice coil.
 
ImmortalZ 35VAC for the outputs ought to give you plenty of power into 4 ohm Statements and help keep your outputs in their safe operating area. Probably a good idea to use MJL outputs to get the biggest package for the best heat transfer, too. That's what you should be most concerned about with low impedance loads.

My guess is a suitable Bass Badger will use MJL4281/4302 outputs and 70+V rails to allow 400W or more into 4R loads. OS was talking about a 2R load IIRC so he likely won't go quite so high with the rails. Horses for courses.

Use the Linkwitz transform judiciously. Don't try to make your driver go too far. It's ugly when an amp clips. Uglier when you bottom a voice coil.

If you do end up building this, read my posts in this thread as I am doing 63v rails for 6.5 ohm Short Thors MTM towers with 6" woofers, pretty similar speakers. I'm building a dual mono Badger; with my transfos, the soft start board, and the speaker protection board I'm stuffing a 5U chassis to the max. With bigger tranfos I'd seriously consider dual monoblocks.
 
The best way to minimize chances for clipping is to trim pream+amp gain for preferable music sources and the chosen speaker driver/s.

For high power operations one could choose either mg6330/9410s transistors recommended by Ostripper or mg6331/9411s (18A). With PS below 100V they offer 8A for up to 100ms and 3A at DC.

cheers,
 
No real problem ...

But you would most likely clip at low volume settings for the preamp
OR lower supply rails for the amp.

QSC specifies a standard of 32db for it's 300w amps and 36db for it's 600w
amps- input sensitivity of 1.2vrms.
To Standardize the Badger, the 27K/820R resistor combo gives 30.5db
and the 33K/820R provides 33db.

Another factor is the rail voltage , 30db is better for the low end (50-60V),
33db is better for 65V or higher supplies. With 65V , hefty outputs like the
Semitech MG series or even the larger MJL series (4302/4281 - 21193/4)
by ON is recommended.
All other specs are REAL close with either gain choice.

OS
 
When you find them, tell us how much you have to pay to get 4u7F polystyrene delivered.
Is there sufficient room inside the chassis for them?

You are correct - I have two external 44 gallon drums for these caps. (It was supposed to say Polyprop)

Q1,2 were transposed with Q21,22 as well. R8 was supposed to be 100R.

The values of the "emitter" resistors on the CFP output stage (R43-R47) have a direct influence on THD as sim'd by LTspice, hence 0.05R here. I have 5 x OP to remove the "need" for SOA protection (for the most part....) I note also that Self mentioned increased THD with the increased number of paralleled transistors in a CFP output stage. Something I have not sim'd myself yet.
 
K71 polystyrene
K72 ptfe
K73 mylar
K74 mylar, high voltage series
K75 pio
K76 stacked polyester, lacquer dipped, similar to e.g. Panasonic ECQ
K77 polycarbonate
K78 polypropylene

(second number gives the type, e.g. K71-4 is axial, 5 is round radial, 7 is boxed radial. B and/or V are MIL grades)
 
Last edited:
With the wide selection (range) in global sourcing , a CFP would be problematic.

I would add a stable triple to the badger. 4-5 OP pairs driven by another
set of njw0281/0302's.

OS

Thx for the comment ostripper.
What do you mean about global sourcing? Supply of Components? Lack thereof?

Anyways, I have found (with simulation) of the CFP stage quite noticeable crossover artifacts at high frequencies which I cannot resolve. I am wondering now if this is due to the comments made by Cordell and Self in their books about turn-off issues with CFP stages and also being worse with more paralleled devices?

Jas
 

Attachments

  • Amplifier - Transitional Miller Comp CFP - integrated models.asc
    21.8 KB · Views: 85
Transistor selection for Q1 & Q2

Happy New Year.
My MPSA18 transistors (Q1 + Q2) measure hfe 900 on my cheapo meter.
If I use KSA1015GR (hfe70-400) or 2SA970BL (350-700) for Q5+Q6, will this be an issue?
Or should I just go out and buy the hard to get SS9014C (200-600) transistors for Q1 + Q2?
 
Increase value of Q1/2 emitter resistors from 100R to 220R.
And increase R24 (tmc feedback resistor) to 1.5K.
This compensates the gain of the IPS to offset higher beta devices.

It will lower the unity gain freq. to 327Khz , but this will not be audible.
Bandwidth is still way above 100k ! :)

Bode below .. rock solid.

Ps - 150R would also work well - 470Khz UG....
OS
 

Attachments

  • badgerbode220.jpg
    badgerbode220.jpg
    37.5 KB · Views: 755