diyAB Amp - The "Honey Badger"

Jump to Latest
ostripper

ostripper

Member
Joined 2008
OK ,
let me clarify the poly caps...

C1 can be any good quality 4.7uf cap (poly/film). It will only see a couple AC volts with a worst case DC fault (input DC) , hopefully not exceeding 25v (OMG!!). This makes 25v the minimum rating.

C3,6 - these .1uf's only decouple the DC blocking cap and cascode. The DC blocker will only see millivolts and the cascode will be at 13-15vdc (zener or resistive divider) . These can also be 35-50v (or 100v).

C10,12,14,16 - standard wima/vishay .1 /100v poly cap. These decouple the 2 local supplies (OPS and VAS/IPS/CSS).

C21 - can be the same .1uf/100v poly. Preferred is a 4/5mm wide 250 v poly.
As the Zobel network , if something went really wrong like 10R LPT emitter resistors or if the builder forgot Cdom (C7,8) :D , A 250v unit would allow R50 to burn right out (oscillation :rolleyes: ).

C7,8 - HAVE TO BE silver mica. I have tried poly caps for Cdom , no smoke ... but the amp suffers sonically.

C18,19 - these are capacitive "stoppers" for the drivers. multi-layer ceramics or best case silver mica 47-150pf. The reason for the range is that different drivers will respond favorably to different values of these caps. 100pF would be ideal.

C20- is the "suckout cap" for the outputs. The reason for the range is that different output devices will require more or less capacitance . This cap should only see the Vbe + small voltage variations caused by Cob and the base-stoppers.

Semi's seem good on the BOM(s).

R34,35 are 1/4w /22R - these are the base-stoppers for the drivers. No real current here. PCB is set for 8.5mm small resistors.

R36 is 1w/ 150R... more current here (20ma class A driver current) best to be safe than sorry. :)



Everything else "jives" - (is the same) ... unless you can find more :eek: .

OS
 
4

454Casull

Member
Joined 2004
ostripper said:
OK ,
let me clarify the poly caps...

C7,8 - HAVE TO BE silver mica. I have tried poly caps for Cdom , no smoke ... but the amp suffers sonically.

C18,19 - these are capacitive "stoppers" for the drivers. multi-layer ceramics or best case silver mica 47-150pf. The reason for the range is that different drivers will respond favorably to different values of these caps. 100pF would be ideal.

Semi's seem good on the BOM(s).

R34,35 are 1/4w /22R - these are the base-stoppers for the drivers. No real current here. PCB is set for 8.5mm small resistors.

Everything else "jives" - (is the same) ... unless you can find more :eek: .

OS
Click to expand...
C6 is a 0.22 uF according to your BOM. Should it be 0.1 uF and added to the long list of decouplers?

Will 1/2W R34/R35 not fit on the PCB? I put them into that group to minimize unique part count.

Why don't you have a D3?
 
ostripper

ostripper

Member
Joined 2008
454Casull said:
C6 is a 0.22 uF according to your BOM. Should it be 0.1 uF and added to the long list of decouplers?

Will 1/2W R34/R35 not fit on the PCB? I put them into that group to minimize unique part count.

Why don't you have a D3?
Click to expand...

I have used .1uf to 10uF for C6. This would depend on which option you use.
R34/35 could be fitted with 1/2W units ... but 1/4w units are so common.
as far as D3....
Here is a excerpt from " DIYA-AMP -construction-advanced-options.txt" (below)
The cascode (Q3-4) can be setup in 2 (or more) ways.

- The first is a zener referenced design - omit R18 , replace
with a 1/2w- 1w 12-15V zener diode and set jumper pad "C" to "Z" with thin wire. The "CRZ" jumper pads
are located above D1-2 near "OFFSET". You will have a cascode reference of either 12 or 15V , depending
on Zener diode choice.

- The second way is to use the "luxman" way ,a crude resistive design. R18 is left as is (NO D3) , jumper "C"
will bridge to "R". This will set the cascode reference to roughly 1/3 of the supply voltage ... whatever that is.

- Either of these 2 Options can use the "C to R" or "C to Z" jumpers in any combination ... all will work !
Click to expand...

-The zener (D3) is an advanced option for a different cascode reference. By default , one could just follow the base BOM and use the resistive divider.
4 combinations can be realized here - (either the 2 resistors to ground or CCS or resistor/zener to ground or CCS). As I said ...all will work as long as the cascode sees a 12- 20vdc reference.

OS
 

Attachments

  • DIYA-AMP -construction-advanced-options.txt
    7.1 KB · Views: 471
N

niss_man

Member
Joined 2008
niss_man said:
Ostripper

With the soft start I will be using 40Vac input. What R1 size would you recommend with this. Also is the relay a 12Vdc coil relay ?

Regards

Simon
Click to expand...

Dont worry OS I just simulated using LTspice. Was using wrong time length. Looks like I need 82 ohm R1 and the relay activates at around 4 seconds. Also looks like relay is a 12Vdc coiled one.

Regards

Simon
 
Last edited:
D

discrete

Member
Joined 2010
audiofan said:
I tried to find parts for the project and 220 ohms 1% are difficult to find ( DIGI KEY ) for R20 & R21 200 ohms are more standards, should I go with 200 ?

In BOM 100 ohms 1% are used for R8 & R9 ... according to me ..... should be for R15 & R16
Click to expand...

1% resistors are usually E96 series. The standard E96 value would be 221 ohms:

Preferred number - Wikipedia, the free encyclopedia

R15/16 should be hand matched to 0.1%. I did not include the offset adjustment pot and still managed 0.1 mV offset on the one channel, 0.7 mV on the other. Of course Q1 and Q2 were also matched.
 
D

discrete

Member
Joined 2010
AndrewT said:
Value 2.2 is included in E6, E12 and E24 series of values.
That value is one of the easiest to source whether in resistors or capacitors or inductors.

E48 and E96 series values are more difficult to source and usually more expensive due to the increased stock that must be carried.
Click to expand...


Agreed Andrew, 220 is as common as grass and 1% resistors are normally available in both E24 and E96 series.

But if you look for 1% resistors on Digikey, they tend to be E96 series. In fact if you search for 1/4 W or 1/2 W with 50 ppm/C you will see only one model at 220 ohm and three at 221 ohm. I guess it makes sense to be able to reach within 1% of any value (the E96 series gives you this) when you use 1% resistors.
 
P

PauloPT

Member
Joined 2010
Hi OStriper and all others following this thread,

I just started to read it and I see this is an amazing efford OStripper is doing! I'd like to compliment him for it.

I have a few questions (I hope I'm not repeating questions already asked, I didn't read yet all the 32 pages of the thread):

Can I use "2SA/C" types instead of "KSA/C" for the bipolars without compromizing performance and sound quality? The "Ks" are a lot harder to find.

When is expected the pcbs be available at the european store?
Has anybody already finished the amp - any listening tests? How does it sound?

Thank you.
 
N

niss_man

Member
Joined 2008
Hi all. Have nearly finished this amplifier. One thing I have found is when using the soft start circuit ostripper has kindly provide I hear alot of hum(think it's transformer hum). It starts to hum after the 4sec delay. When the 40v AC (which is from my main power trnsformer) input to the soft start board is removed to hum disappears completely. I have not connected the amplifier boards to +/-ve, gnd yet so this has nothing to do with them. Has anyone got any ideas to stop this noise??

Regards

Simon
 
D

discrete

Member
Joined 2010
niss_man said:
Hi all. Have nearly finished this amplifier. One thing I have found is when using the soft start circuit ostripper has kindly provide I hear alot of hum(think it's transformer hum). It starts to hum after the 4sec delay. When the 40v AC (which is from my main power trnsformer) input to the soft start board is removed to hum disappears completely. I have not connected the amplifier boards to +/-ve, gnd yet so this has nothing to do with them. Has anyone got any ideas to stop this noise??

Regards

Simon
Click to expand...

Is it possible that you have some DC on your mains? Big VA toroid transformers will complain loudly when the the core saturates due to the DC. Adding a small resistor in series on the primary is sufficient to prevent core saturation. This is what the soft start is doing during the first 4 seconds.

Unless your incoming mains power is very bad, the DC is often caused by other devices using half wave rectification. In my case it is a floor lamp with a 50%/100% switch when set at 50%.

ESP has a good article and solution:

Mains DC and Transformers
 
N

niss_man

Member
Joined 2008
12v relay activates after about 4 secs bridging out the ntc thermistor. Then hum starts. Please note that the hum is still there when the tranny is not connected to the soft start relay but the 40v input is still connected. Noise is gone with soft start power voltage40 removed.
By the way, this amp cranks.

Regards
Simon
 
You must log in or register to reply here.
Top Bottom