25W Class A amp with Lateral MOSFETs

Jump to Latest
prasi

prasi

Member
Joined 2014
sites.google.com/site/analogdesign1234/
Paid Member
manniraj said:
Thank you, and I using the board from post . So I need to set it to 100mA bias on the 100R resistor. I usually use the voltage side of the DMM for precise bias, so 100R*100mA = 10mV and I hope I got my calculation right :)

Let me do thatA, thanks again.
Click to expand...

For 100mA bias, use a 10R-0.5W to 1W resistor, set 1 V across it.
V=IR= 0.1A x 10R=1V
 
prasi

prasi

Member
Joined 2014
sites.google.com/site/analogdesign1234/
Paid Member

Attachments

  • lateral (1).jpg
    lateral (1).jpg
    328.7 KB · Views: 871
  • maplin sch_1.png
    maplin sch_1.png
    15.8 KB · Views: 898
  • maplin1.png
    maplin1.png
    76.9 KB · Views: 1,002
Last edited:
M

manniraj

Member
Joined 2012
Nelson Pass said:
There is a small concern about the bias, as it appears to be sensitive to
supply voltage. If the DC supply is constant, no issue, but I expect the
voltage across RV1 to drift upwards with supply voltage.
Click to expand...

Thanks Papa, I am using a 22vac transformer which is giving me a very stable +/-34.5vdc. Able to use RV1 to set the bias without drifting so far. But I am yet to do this setup using the 10R/1-2W resistor as Prasi suggested for 1V bias across this resistor.

:)
 
M

manniraj

Member
Joined 2012
Using a 10R/2W resistor I was able to set the bias to 1V per Prasi suggestion which makes it to 100mA. Played few tracks and its playing beautifully. My small 120/100/50mm heat sink is barely warm even after an hour of playback. Offset is pretty stable around 5-6.7mV.

Thanks all.
 
M

manniraj

Member
Joined 2012
At last found sometime to finish the Maplin and it sounds excellent. Thanks to Prasi for the beautiful layout. I get slight hiss kind of sound which not sure because of my usage currently with closeby AC wiring. I am yet to check with my main speaker setup of Troels :)

Few build pics for you Prasi





 
  • Like
Reactions: 1 user
A

Alexandre

Member
Joined 2004
Hello

I like it, I find this beautifully simple. Single supply, no matching required, no thermal compensation headache. I will build it. I have sources for the parts that I'm hoping are honest (hopefully genuine).

edit: there it is again, that interesting "bootstrap ccs". Looking forward to build this. Thanks Juma

Thanks,
Alex


juma said:
Hi Skorpio,
I tested your 2-loops idea and it brings some improvement indeed :up:
I also changed the input device into N-JFET which made some other changes necessary and the whole thing now looks like this (and sounds somewhat better) :)


Please don't ask me about parts substitutes, power increase or to compare it with other amps/designs....
Click to expand...
 
Last edited:
D

Dave S

Member
Joined 2003
juma said:
That circuit is tested and stable. Clumsy prototypes often create problems.
Make reliable connections (PS feedback loops, input, grounding) - no long wires, alligator clips and similar, ground your heatsinks....

100pf on Q4's base and collector is way too much - 10pf or so is more like it.
Try to slightly increase C4 or introduce a couple of pf over R10. Keep gate stoppers (R3, R6, R13, R14) very close to the gates.
Click to expand...

I have this amp running on the bench. I used my own pcb design, which I believe is pretty decent.
I have encountered borderline instability.

Differences from the original:
I used J309 instead of J310 (and 2SK170BL)
R1 and R2 are 499K instead of 1M
P1 = 20K (not 10K)
R13 changed to 249R (although this made no apparaent difference)

Removal of the 10pF cap helps, but to ensure stabliity at all signal levels a miller cap of 10pF or greater on Q4 is needed. I have yet to try the suggested cap from Q4 collector to the R8/R9 node.

The amp has some features missing compared to "normal":
  • No input filter
  • No zobel network
  • No output inductor (seems to be unnecessary with cap coupled amps)

Please could you explain a bit more about your design ethos and why these features are not fitted.
Any further comments on the stablity issue?
 
juma

juma

Member
Joined 2005
Dave S said:
...could you explain a bit more ...
Click to expand...
Those features are ugly, harmful patches (from SQ PoW) but they are mandatory in commercial design and I tend to avoid them when it's possible. Of course, those techniques are well known and anyone can add them to the circuit in accord with specific conditions and needs (this is DIY after all). Stability depends on real-life circumstances too and it's easily achieved/mended once you understand the relations between circuit/components parameters (reactance/impedance ->phase, gain, load...).
My "design ethos" (big words if you ask me, but there you have it :)) is to make it as simple as possible, fun to build and good enough to listen to Haydn/Mahler through the afternoon without being bored or tired.
 
D

Dave S

Member
Joined 2003
juma said:
Those features are ugly, harmful patches (from SQ PoW) but they are mandatory in commercial design and I tend to avoid them when it's possible. Of course, those techniques are well known and anyone can add them to the circuit in accord with specific conditions and needs (this is DIY after all). Stability depends on real-life circumstances too and it's easily achieved/mended once you understand the relations between circuit/components parameters (reactance/impedance ->phase, gain, load...).
My "design ethos" (big words if you ask me, but there you have it :)) is to make it as simple as possible, fun to build and good enough to listen to Haydn/Mahler through the afternoon without being bored or tired.
Click to expand...
Thanks!

Not sure that I will be doing Haydn/Mahler but looking forward to having a listen.:)
 
D

Dave S

Member
Joined 2003
Sadly I was not able to suppress the oscillation by fitting 20pF from VAS collector to the feedback point (R8/9 node).
The oscillation is about 2V pk-pk in the MHz region and "rides" on top of any output waveform. Both channels are the same.
I have resorted to 20pF miller cap on the VAS to ensure stablity.



May I ask why the J310 was chosen (not another 2SK170)?


Anyway, I will build the cap multiplier PS next and then proceed to audition the amp.
 
mlloyd1

mlloyd1

Member
Joined 2001
sorry to hear you're having troubles. is the oscillation in single digit (1's of) megahertz or double digit (10's of) megahertz? wondering if the BF862 is screaming or if the output mosfets are or something else entirely ... care to show construction pictures so we can see things like lead dress, grounding, etc.?
also, did you change the gate resistor values at all?
 
D

Dave S

Member
Joined 2003
Thanks Mlloyd. I think you might have identifed the reason for the J310 - it provides a couple more volts of VDS for the 2SK170.

That could be a reason for the instability.
OTOH I don't understand how C4 actualy stabilises an amp. It looks to me like it rolls off the gain of the amp at HF which increases the feedback factor and therefore reduces the phase margin... or have I got it wrong?
Moving the end of C4 to the VAS collector effectively misses the o/p stage out of the feedback loop at HF. I can see how that helps phase margin.

I have tried a 249R gate resistor for the 2SK1058. It made no difference.

The oscilation is actually 7V pk-pk at 6.7MHz (sorry I should have been more observant before).

Here is my layout:

Layout.jpg


Apologies for my typo "Jumo" :(
 
You must log in or register to reply here.
Top Bottom