PeeCeeBee

Dear Shaan,
if I changed some parts is that okay...
on the picture I replace BD139-140 with something different ;)
not soldered yet, if it is okay I will solder if not I will replace with 139-140 instead...

In my PeeCeeBee also we can add a pair more 1000uF at the rails if needed :cool:
BTW what is the HFE of your input pair?
I don't have any matched yet,

Regards

Nice progress Bli, are you using toshiba's 1360/3423 ? Yea..good choice, i used it on my my SSA peeceebee..
 
My layout, as promised.
Some components aren't standard, for instance input cap is very big (RM 37,5). I've made peeceebee according to components I already had at home. Basically, it's shaan's peeceebee.
 

Attachments

  • VSSA Shaan layout.pdf
    818.7 KB · Views: 1,161
  • VSSA Shaan bottom.pdf
    249.6 KB · Views: 739
  • VSSA Shaan silk.pdf
    220 KB · Views: 734
...................Approximately 380. BC559B/BC549B.

Try using the ones having the highest rated ones, i.e. 'C' suffixed ones with hFE of 600 or more. :up:
bc5xxC have hFE from 400 to 800.
That 400 minimum limit is only 5.3% more than the 380 he measured.
There is a high probability that a group of C grade will be substantially in the hFE range of 400 to 450.
 
Last edited:
Most of the PNP group 'C' trannies I own have higher hFE than 650. But I didn't use them coz the all the NPN counterparts are all between 500 to 550. The 'B' rated pairs, although low gain, are more close in this regard, hence chosen.

Hi dovrivoje. Thanks for uploading the layout files. :worship:
 
Approximately 380. BC559B/BC549B.

Try using the ones having the highest rated ones, i.e. 'C' suffixed ones with hFE of 600 or more. :up:
Ok thanks, as I match BC550-560B I also get two matched pairs around 380 but different manufacture/brand.
I'm also matched two pairs 2SA-SC and get around 400, & I will use it.
My PeeCeeBee is designed for BC but I think I can put it on with a little trick :)

bc5xxC have hFE from 400 to 800.
That 400 minimum limit is only 5.3% more than the 380 he measured.
There is a high probability that a group of C grade will be substantially in the hFE range of 400 to 450.
Hi Andrew, it is hard to find real C grade here, lately I found few 550C after measure hFE with cheap DMM it show bellow 300. I only have BC549C & hFE 700+ and some 2SC that around 600-800+ but no 2SA.

BTW can we hear the difference between HFE of 400 & 600 or more?
I think it is okay to use it (hFE 380-400) because I can wait any longer that I want to made it :)

Speaking of transistors, I wonder if Shaan, or someone else, should say a little about measuring/matching the transistor pairs, at what current, and how closely they need to be matched.

What I do NOT know is the best method for this specific circuit. I can make a guess, but perhaps this is something that has come op in one of the other SSA or TSSA threads... ?
Hi PMI just use cheap but trusted DMM, I have use two kind DMM & still like the cheaper one :)...
We are matching to "compare" one & another to get matched one but with some variable (time, temperature, what we use, etc)
My point is:
use same DMM
almost same time
same temperature & don't let your hand touch the DUT because it is very sensitive, the higher hFE even more.

BTW I use my bare hand to put it on the DMM & wait for a moment and then note the result. After it is done I leave the DUT so it is back to room temperature & measure again to make sure the result :)
I know it is not so accurate also time consuming too :D:D:D
 
Last edited:
how to first power up

Hi Shaan, I wish I have a link too :), later...
BTW
There is no guidance how to safely power up our PeeCeeBee & what we need to measure at first time... (maybe I miss it)
Actually I don't know how to measure current that flows on the LatFet, usually I measure at the power resistor (0,22-0,33R) if any :p

Added link for quick guidance for newbie like me is very good :)
and
This how I put 2SA-SC transistor on BC PeeCeeBee
 

Attachments

  • WP_000560.jpg
    WP_000560.jpg
    123.6 KB · Views: 2,144
  • WP_000561.jpg
    WP_000561.jpg
    147.2 KB · Views: 1,428
  • WP_000569.jpg
    WP_000569.jpg
    246.9 KB · Views: 1,393
Hi Shaan,

Alfet now is beautifully singing. Based on your peeceebee design. I re-designed the output side in paint to accommodate the alfet. The vocal is crystal clear, very very detailed, the highs very smooth and the bass is there well balance:):):). I modified some components values to suit my taste. Running at low temperature in normal listening volume with small heat sink. I like this very much than previous one.:):):) Dc offset 0.0mV - 0.5mV. When using analog tester, the pointer stays at zero at 0.1 DCV setting.:):)

Regards,
Boyet
 
...how to safely power up our PeeCeeBee & what we need to measure at first time...

We need,

1. A digital multimeter.
2. 1R/1W x 2.
3. 22R/2W x 2.

For first time power-up solder two 22R/2W resistor to the PS rails, one end of each of them to each rail. Connect the PS to the open ends and turn on power. If nothing explodes/smokes then replace both of the 22R with 1R/1W and connect the PS again, but don't power it up instantly.

Take notes for the measurements from now on.

Set your digital meter to 'mV' and connect the probes to both ends of one of the 1R resistors. Turn on power. Check meter reading. That number shows total Quiscent current Iq.

Measure the front end plus VAS current I(f v) by probing the voltage at the ends of the 10R after the 4007 diodes.

Measure VAS current I(vas) by probing the voltage at the ends of the 10R at their emitters.

Now,

Total current= Iq
Front VAS current= I(f v)
VAS current= I(vas)

So, from aquired data we get,

Front end current= I(f v) - I(vas)

and

FET current= Iq - I(f v)

I hope this will be helpful.

shaan
 
Hi Shaan,

Alfet now is beautifully singing. Based on your peeceebee design. I re-designed the output side in paint to accommodate the alfet. The vocal is crystal clear, very very detailed, the highs very smooth and the bass is there well balance:):):). I modified some components values to suit my taste. Running at low temperature in normal listening volume with small heat sink. I like this very much than previous one.:):):) Dc offset 0.0mV - 0.5mV. When using analog tester, the pointer stays at zero at 0.1 DCV setting.:):)

Regards,
Boyet

Thats very good news boyet.

Just curious, do the Alfets have internal HV protection zeners? I guess not. If so, then it is a good idea to solder two of them between G and S pins of the FET(s), after the 100R gate stoppers. The (-) terminal of the zeners goes to Gate of the N-ch FET, and to Source for the P-ch FET. 15V rated ones, like 1N4744 will be perfect. :)
 
Last edited:
Just curious, do the Alfets have internal HV protection zeners? I guess not. If so, then it is a good idea to solder two of them between G and S pins of the FET(s), after the 100R gate stoppers. The (-) terminal of the zeners goes to Gate of the N-ch FET, and to Source for the P-ch FET. 15V rated ones, like 1N4744 will be perfect. :)
The ALF08N16V and ALF08P16V do not, which is why I asked how he was using them. I really wanted to try them myself, but could not see them both stocked by anyone here. N-channel version only stocked... :(

Maybe LC bought up all the good parts - :D

Bottom of second page of the datasheet linked below.

http://www.semelab.com/pdf/magnatec/ALF08N16V ALF08N20V.pdf
 
Yea these are fairly new beasts. On the other hand the classic Hitachi 1058/162 pair is abundant and Renesas site pages makes me think the pair isn't going out-of-market anytime soon.

If anything I "feel a higher desire to have" and would allow to replace my lovely and lively 1058/162 pairs with, are the SIC jFETs from Semisouth(are they 'created' in pairs too?). A dream that probably will become reality within in a few years, wishfully.

:D

edit: Semisouth is about to be closed. As says the news, likely to be old news. ... I need a shoulder. T_T
 
Last edited:
Thats very good news boyet.

Just curious, do the Alfets have internal HV protection zeners? I guess not. If so, then it is a good idea to solder two of them between G and S pins of the FET(s), after the 100R gate stoppers. The (-) terminal of the zeners goes to Gate of the N-ch FET, and to Source for the P-ch FET. 15V rated ones, like 1N4744 will be perfect. :)

Hi Shaan,

I used ALF08NP16V5. No zener diode. No changes was made. As is what it is in the schematic only I change output device. This is what in the Application note:
"Gate Protection & Safe Operating
Area Limiting


Semelab’s previous family of lateral mosfets
included internal gate-source zener diodes to
protect the junction against overvoltage
situations. On the new family of Alfets these
zeners have deliberately been omitted for a
number of reasons to enhance the performance
and as a result some protection does need to be
added externally to the devices (in addition care
should be taken in handling the devices because
of the high impedance sensitive gate and static
handling precautions should always be
observed)."

But I try your suggestion to put zener in the gate.
2sk1058/2sj162 has potential to improve the sound quality in peeceebee..I try to modify again after my work and i will update you the result.

Regards,
Boyet
 
Hi Shaan,

Alfet now is beautifully singing. Based on your peeceebee design. I re-designed the output side in paint to accommodate the alfet. The vocal is crystal clear, very very detailed, the highs very smooth and the bass is there well balance:):):). I modified some components values to suit my taste. Running at low temperature in normal listening volume with small heat sink. I like this very much than previous one.:):):) Dc offset 0.0mV - 0.5mV. When using analog tester, the pointer stays at zero at 0.1 DCV setting.:):)

Regards,
Boyet

:up: ALF08NP16V5 :sing: