B&K ST-140 RECAP BOM

I do not have a variac. I used a dim bulb tester instead. After i adjusted dc and bias current i switched to 220V mains (Argentina)
 
JoseHH said:
Hi JR,
I am sorry about his set back. R9 is supposed to be a 66.5 Ohms and it is passing some 18 mA if I understand the circuit correctly. Check Q9, Q10, and the two gate resistors.
Let me tell you that I have tested the adjustment procedure on a prototype that I built by copying the schematic for the ST-140. I could adjust the idle current as well as the DC offset without problems, without signal generator, without dummy load. I just used two DVM, but this may just as well be done with only one.

I built my prototype with things I had laying around, scavenged some parts from junk PCBs. I only tested it with +/-36V. because i do not have a suitable transformer to generate +/-62v. I can confirm that after I adjusted the idle current to 100mA on the negative fuse, and adjusted the DC offset to some 10 mV, i tested it with speakers and I could play music on it. Sounds all right considering my crude prototype.
Click to expand...
That is amazing jose, very interesting to learn that. You probably know as much about this circuit as anyone these days. Nothing like building something to really gain insight. If R9 is passing 18mA, it's only a 1/4 W resistor, I wonder how much headroom that leaves before it overheats. I'll have to mess around with the power equations/ohms law, but it would be interesting to know how many watts with that current, at 62 V, would need to be dissipated.

R-9 was replaced with Vishay 1/4 W metal film @ 66.5 ohms. I'll try the process again tonight and monitor the temp of R9. Stay tuned.

jrh
 
JR Harley said:
R-9 was replaced with Vishay 1/4 W metal film @ 66.5 ohms. I'll try the process again tonight and monitor the temp of R9. Stay tuned.
Click to expand...
1/4W should be fine. Please don't forget to check Q9, Q10 and the gate resistors which should be 1000 Ohms. Always back P2 so that the parallel combination of P2 and the 475 Ohms resistor give you some 60-75 Ohms before powering on to prevent turning it on with too large of a bias current. If your DC offset was correct, i think it should be close now so no need to touch P1 now.

Good luck
 
!Gracias, Jose'! Nice! !Es un genio! Any ideas on increasing the wattage output?
I always replace them with 1/2 watters- considering that manufacturers always go cheap.
 
Last edited:
Thank you Ray. I do not know much about this. I am simply trying to understand how this (very simple) mosfet amplifier works, and since It is fun for me to play with prototypes i decided to build one. This way I can see what it takes to make the adjustments.
With regards to your question about increasing power: As you know B&K has made several revisions of this circuit so some of the difficulties with it were cured. B&K also made several models with larger power by doing two things:
1-Increase the voltage rails to up to 75V
2-Using multiple mosfets in parallel.
Some say that you need to match the mosfets to put them in parallel or may require using source resistors. There is some discussion in this other forum.
Now this model seems to be a very crude design. There is no speaker protection circuit, and bias current is so crudely controlled that it may be drift in which case I am sure the speaker will burn to protect the output fuse. Keep in mind that the the current that a fuse can hold before it blows is proportional to the speaker price. I wouldn't try to make this beast more powerful because of this.
 
diseño crudo/ básico. speaker protection fuses-2a. mild valve/tube sound. sounds better when the bias is set hot- over 200. i'm looking forward to moving on to my other projects.
Take care.
 
Last edited:
ST-140 update: Good news.... and bad news. The good news is, that after replacing Q10 and R9 on the right board, we're able to bring things up and adjust both DC offset and bias. The bad news is, that at 70 V AC on the variac, things are just too damned hot. Q10 is 157 degrees F and R9 is about the same. It just didn't feel right with the temperature rising rapidly, so I pulled things back and shut it down.I think I'm going to pull the right board again and test more components. I couldn't get to everything I wanted to with the board installed, so I'm going back in. I'm just happy that I'm not trying to run down a short circuit that doesn't exist, like I have been for the last 2 years!

We're getting there. Hope everyone had a good weekend.

jrh
 
Great News!
If you are using the sine wave then you are right. May be something is still not right in that board. Does the other board do the same?. What happens if you leave it at 100mA. Are they still running too hot?.

EDIT:
I meant to say that with the procedure I suggested you it may be the idle current should be left a bit lower than 200MA. I think that when the sine wave is applied you may get a larger idle current if you left it adjusted in the absence of the 20KHz signal.
 
data sheet:
"Maximum Operating Temperature: + 200 C".
392 deg. fah.
those suckers get hot! mount away from board- 1"
Q10 butterfly heatsink- had thermal grease. now red loctite. temp decreased substantially.
i read about 110 deg. farh. at 70v after 30 min. about 130 deg. farh. at 120 v.
try cleaning alcohol , reinstalling heatsink , dousing loctite. 24 hr dry time. see if the temp drops.
R9 reads about 81 deg fahr. after 30 min. at 70v. about 90 deg farh. at 120v.
1/2 watt resistor run cooler
 

Attachments

  • IMG_20201004_204239.jpg
    IMG_20201004_204239.jpg
    71 KB · Views: 42
Last edited:
Just to clarify - Still just testing the right board ray/jose style, no input signal.
You may be right ray, the heat may be normal. I'm strictly relying on my instinct, even as I'm aware of the specs. The transistors I'm not as concerned with, but the resistor heating up so much had me concerned. If it ran that hot continuously, it would discolor the board. And that hadn't happened in the previous 30 years of using the amp. I'll test some more and see what I can learn. Thanks for the comments guys!

jrh
 
installed an Inrush Current Limiter to limit / eliminate the thump! installed it like the schematic and it did not work correctly ( on the bridge rectifier). works between the fuse and the power switch, before the transformer.
 

Attachments

Last edited:
Hey ray,
Interesting, I've thought about doing that as well, seems like a good idea with such big filter caps, but I never had an issue there. Also, 1/2 W resistors would definitely run cooler. R9 and R10 seem to be good candidates for that, but it just puzzles me that I never noticed the heat before. Anyway, I've tested things and looked the board over pretty well, and it all seems fine, so I'll re-install and see if I can get it there. We'll see if I can keep control of the reins this time around.

jrh
 
Hello Ray andJR,
R10 should not be passing much current unless something is not right either at Q8 or Q9 orR11. What do you have installed for Q8 Nd Q9?. Are they correctly oriented?. Also is R11 (going from Q8 emitter to V-) 402 ohms?.
 
Q10 and R9 getting hot means a large current passing through them. That current can only go through q8 and Q9 but this means that P2 should also be passing that current in which case P2 and the parallel 475 oms resistor should also be getting toasted. If all is normal along P2, then your N channel is a suspect, because the gate should not take dc current because it should look like a capacitor unless your 2sk136 is doing something funny...
If I were you i would do the following measurements when your amp. powered at 70V on your variac:
1-measure the voltage drop across R9. Determine te current passing trhough it.
2-measure the voltage drop across the 2sk136 gate resistor. Determine the current. There should be no voltage drop there.
3-measure the voltage drop through pins 1 and 3 of P2. Power off, and measure the resistance between pins 1 and 3 of P2. Determine the current through it.
My expectation is that the current through R9 should be equal to that through P2.
 
To JR,
Lets make some further reasoning aling my last post. You said that you applied 70V with the variac. If i take the mains in your country to be 115V, then instead of having +/-62 Volts at the ouput of your power supply, you will have only +/-37.7 V. which is very close to my prototype running at +/-36V. Then my measurements should ve very close to yours.
1-
I adjusted my idle current to 110mA. DC offset adjusted to zero.
2-
Then my R9 (69 ohms) had 1.195 volts across. This is some 17.5 mA. In good agreement with my calculation of 18mA that i mentioned before.
3-
My P2 and its parallel installed resistor combination measures 77 ohms, and i registered a voltage drop of 1.353V across it. Thus 17.5 mA are passing through it. Just as expected.

4-
When i measured the voltage drop across the ends of either of the gate resistors (1000 ohms each) i get zero volts, thus no gate current. Again this is to be expected at idle.

Q9 and q10 get barely warm. nothing to worry about.
You can try to bring down the idle current to 110mA and do these measurements and compare to my values.
 
JoseHH said:
Hello Ray andJR,
R10 should not be passing much current unless something is not right either at Q8 or Q9 orR11. What do you have installed for Q8 Nd Q9?. Are they correctly oriented?. Also is R11 (going from Q8 emitter to V-) 402 ohms?.
Click to expand...
Good point jose, I did notice that R10 was not hot like R9. R11 is very close to spec at 401 ohms. I am confidant at this stage of the game, that ALL the transistors are the right substitutes and oriented correctly. More to follow, and thanks for your study of my efforts!

jrh
 
R10 was the resistor from collector of Q5 to base of Q8, which should be a low value (I used a 20 Ohms resistor there).It does not gt hot at all.
R11 + q9 must have the same current as R9. If not then you must have a current path that is not accounted for. May be a solder bridge?
EDIT
The current through R11 plus the current through Q9 must equal the current that passes through R9.
 
Last edited:
JoseHH said:
Q10 and R9 getting hot means a large current passing through them. That current can only go through q8 and Q9 but this means that P2 should also be passing that current in which case P2 and the parallel 475 oms resistor should also be getting toasted. If all is normal along P2, then your N channel is a suspect, because the gate should not take dc current because it should look like a capacitor unless your 2sk136 is doing something funny...
If I were you i would do the following measurements when your amp. powered at 70V on your variac:
1-measure the voltage drop across R9. Determine te current passing trhough it.
2-measure the voltage drop across the 2sk136 gate resistor. Determine the current. There should be no voltage drop there.
3-measure the voltage drop through pins 1 and 3 of P2. Power off, and measure the resistance between pins 1 and 3 of P2. Determine the current through it.
My expectation is that the current through R9 should be equal to that through P2.
Click to expand...
All knowledgeable comments jose. Let's see how my next attempt at biasing on the right board goes, and we may re-visit these measurements. Thanks, as always. We'll get 'er right.

jrh
 
You must log in or register to reply here.
Back
Top Bottom