B&K ST-140 RECAP BOM

[JR Harley]

Hey neil,
I tend to agree with jose about the filters, but I also found them here in case you want to browse:
https://www.tedss.com/Catalog/Browse?searchString=101163UO75DD2B&inCategory=-

jrh

 

[raycarr00]

These protect speaker investments: Polyswitch Resettable Fuse 72V 0.75A TYCO RXEF075. five for three bucks & soldered to each driver's wire in series.

NeilBEN, that's one pretty heatsink!

 

[NeilBEN]

Ray, I had one of those polyswitches in the crossover on the tweeter, but I decided to take it out, thinking it would degrade the sound a bit?

The heatsink used to be black, long, long ago. Thanks for catching my misreading of the cap value. Those look to be very similar if not identical physical dimensions to my current ones, JR.

If I replace the signal caps, do I have to rebias? Because I don't have the variac, or scope.

 

[JoseHH]

It is unlikely that the bias and dc offset would change by changing those 4 capacitors. But if you need to, you can adjust the bias with a dvm, a dim bulb tester and a set of dvm probes provided with minigrabber hook ends. Dont use regular needle end probes for this.
If i were in your situation what i would do first is to measure the voltage on the speaker terminals with the dvm to see if you have any dc offset. Nothing should be connected on the input RCA connectors for this and wait 10 minutes with the unit powered untill the dc offset is stable.
Record your dc offsets and decide if you need to adjust P1 to get zero volts.
If both channels have zero volts within a few mV, then check what is the bias current that you have now prior to replacing anything. Do this by setting your DVM yo measure current on the 10A scale, and then go one board at the time by removing the V- fuse and clipping the dvm mini-grabber hooks onto the terminals of the empty fuse socket. Only then power the amplifier on and write down the current you read on the amp meter.
Repeat for the other channel.
ideally, you need to wait 10 minutes to allow for stabilisation of the current.
You can adjust P2 to get 0.2 Amperes if you want, while you monitor the current after 10 minutes of warm up. This is sufficient to get your amp biased.

It is good to know the bias current and dc offset before replacing any components. If you find that while trying to set the dc offset to zero, the adjustment of P1 does not get to zero, it means that you have to replace Q2, Q3 and/or Q5, Q6.
Thus, you order your new capacitors and new transistors, and you may also get new trimmers for P1 and P2 and save time and shipping costs.

 

[raycarr00]

a local shop could set the bias/ DC offset- 20 bucks.
4 caps (two 47uf/6.3v bipolar and two 100uF/100v polar) 10 bucks; digikey. See BOM.
my tweeters are about 300 bucks each. I'll stick to the polyswitches. music still sounds schweeet!

 

[JR Harley]

Hey neil,
jose is right on. Replacing those caps should not alter the bias or DC offset values significantly, assuming they are good to go initially, which it seems they are. That said, if you perceive the amp to be "running hot", they may not be. Monitoring these values as jose suggests, is a good procedure to undertake in that case. The factory trimmer pots, P1 and P2 on the schematic, are unfortunately not very "precise", and are touchy to adjust, since even the slightest torque to them can shoot the current way past what you want. As jose suggests, Bourns does make some very nice substitutes that are very smooth and precise, but they're a bit pricey at $7/each or so. "Plan your work, work your plan". Best of luck friend.

jrh

 

[NeilBEN]

If i were in your situation what i would do first is to measure the voltage on the speaker terminals with the dvm to see if you have any dc offset. Nothing should be connected on the input RCA connectors for this and wait 10 minutes with the unit powered untill the dc offset is stable.
Record your dc offsets and decide if you need to adjust P1 to get zero volts.

Okay I did this, and with my meter set in the 2000mV range, the left channel measured -225 and the right -096. I have to assume this is way out of whack?

 

[JoseHH]

I'll stick to the polyswitches. music still sounds schweeet!

That is fine Ray, if it works for you why not using it.
Like I said before i am studying the subject out of my own interest, I am by no means criticizing how you deal with your system.

I am now building an amplifier using the same schematic of the ST-140 but I built a speaker protection board with a relay so in case that DC offset appears at the output, it disconnects the speakers.
I also have the fuse in place. I think it is fine to have a speaker fuse just in case. Another thing I am wondering about it is to add a zener diode to prevent that the gate-to-source voltage never exceeds +/-14V, as this will destroy the output mosfets. This could happen if something goes wrong with Q9-Q10. Normally a zener diode in series with a regular diode is used between each of the gates and the common sources node. This scheme is commonly used in mosfet output stages, for example see this or this or this tor perhaps this o name a few.
One interesting document to learn about this, as well as how much idle current should be used based on measurements for very similar lateral mosfet devices, is (pages 8-12) this application note. If interested in knowing about the Hitachi power mosfets in the ST-140, it is worth reading this old application note.

 

[JoseHH]

Okay I did this, and with my meter set in the 2000mV range, the left channel measured -225 and the right -096. I have to assume this is way out of whack?

You should adjust P1 to get those values close to zero. At least get within +/-10mV of zero. If possible insulate most the shaft of the screwdriver and leave only the endmost 2 mm of the screwdriver blade to prevent shorting anything while you reach into the trimpot to adjust it. Do a very small adjustment, and be sure to measure not to have any speakers connected while you are doing this. Only your DVM at the output. When you get less than 200mV (as you have now on the right channel) use the 200 mV scale. The adjustment has to be done with the amplifier powered and after some 10 minutes of warm-up time.

 

[JR Harley]

ST-140 fan boys, update here:
I am getting increasingly confidant that I am very near completion on this epic amp journey. Thanks to all of you guys for your feedback, commentary, and insight, as I have worked to get this amp back up to speed. The information has been invaluable, and perhaps more importantly, enjoyable.

There is no problem passing an audio signal thru the beast at this point. Even at very low AC voltage, the amp pumps a strong audio output into the test speakers. Since my CD source doesn't have it's own volume control, it's a little too much to let go for long, so real music listening will have to wait for a bit, but the testing has been very satisfying. Check out the attached, annotated pics for where I am with it. If more detailed analysis with the digital scope reveals anything interesting, I'll post it up. Many thanks all!

jrh

 

[raycarr00]

Jose',no problemo(Terminator version)or,tiada masalah.
My motto: save a driver,use a polyswitch. Otherwise,use a smaller rating on the speaker fuse.
Check out the schematic of the BK ex 442 version ...was wondering how to increase bias safely.to 250mA.

NeilBEN, don't mess with the bias unless you follow the BK procedure. You don't need a signal generator. Dim bulb tester would work.
But with a DMM, you can check and adjust the DC to close to zero. It'll save your speakers from damage.
No signal, amp warmed up. DMM set to DC. DMM probes placed on one channel of speaker terminals.
You can wrap a small screw driver with tape to prevent shorting out the amp.or making your hair stand from shock!
choose the correct trim pot. slowly turn and watch the DMM. Take your time.
proceed to the other channel.
swap those caps!

Great job Harley...

 

[JR Harley]

Jose',no problemo(Terminator version)or,tiada masalah.
My motto: save a driver,use a polyswitch. Otherwise,use a smaller rating on the speaker fuse.
Check out the schematic of the BK ex 442 version ...was wondering how to increase bias safely.to 250mA.

NeilBEN, don't mess with the bias unless you follow the BK procedure. You don't need a signal generator. Dim bulb tester would work.
But with a DMM, you can check and adjust the DC to close to zero. It'll save your speakers from damage.
No signal, amp warmed up. DMM set to DC. DMM probes placed on one channel of speaker terminals.
You can wrap a small screw driver with tape to prevent shorting out the amp.or making your hair stand from shock!
choose the correct trim pot. slowly turn and watch the DMM. Take your time.
proceed to the other channel.
swap those caps!

Great job Harley...

Thanks ray, your "devil may care" enthusiasm has been an inspiration!
Best,
jrh

 

[NeilBEN]

You should adjust P1 to get those values close to zero. At least get within +/-10mV of zero. If possible insulate most the shaft of the screwdriver and leave only the endmost 2 mm of the screwdriver blade to prevent shorting anything while you reach into the trimpot to adjust it. Do a very small adjustment, and be sure to measure not to have any speakers connected while you are doing this. Only your DVM at the output. When you get less than 200mV (as you have now on the right channel) use the 200 mV scale. The adjustment has to be done with the amplifier powered and after some 10 minutes of warm-up time.

Okay - thanks! I need to get a better DVM - this one was one I got from my local hardware store in a pinch, and it only goes down to the 2000mV scale.

What DVM do you folks recommend? I have seen some with temperature probes - would that be useful?

And, naturally the pots are facing the back ...

Edit: they do have a small slot facing the front - whew!

Piher Spain is the marking - are these the touchy ones?

 

[JoseHH]

Those trimmers are good quality.
The dvm to buy depends on how much use you are going to make out of it, but when you buy a dvm, try to get one that brings a 200mV scale. Mine is UniT brand, and it is decent byt there are much better ones (fluke, tektronix, etc but tey are not cheap).

 

[NeilBEN]

I am glad it has good trim pots. I will get them a close as I can with my meter, and get a better one. To change the battery, you have to unscrew the back ... it is not great. I will get a better one, and check it again.

I am getting the new tubes for my Modulus today, so I want to get things close, so I can listen.

 

[raycarr00]

Certainly, your DMM should be sufficient. Set it on DC( V DC =) and connect its terminals to the speaker terminals. adjust dmm scale until you get a reading. just turn the trim pot that's towards you(not the rear one) gently in either direction until you get close to 0. long and taped-up, small tip screwdriver (flat or Phillips). reach the trim pots from under the board sitting on the large-blue can caps.
BK says: "Adjust the DC pot (P1) for 0 VDC ± 5 mVDC."
pawn shops have quality, used versions for cheap. amazon and the bay have them for about 9 bucks-
Neiko 40508.

 

[NeilBEN]

Well, my luck is good, so far! I found the plastic trim pot screwdriver, that I knew I had somewhere - it was literally inches from my left hand on my work desk:

And, using a vinyl glove I was able to get both channels to settle on 000mV, a drift only to -003 or so. But they seem to stay on -000 or 000 on my cheesy meter much of the time. The amp had been on for over 20 minutes.

And my new 6299 EH tubes just arrived from Arizona, via USPS Priority!

 

[JoseHH]

was able to get both channels to settle on 000mV, a drift only to -003 or so. But they seem to stay on -000 or 000 on my cheesy meter much of the time. The amp had been on for over 20 minutes.

And my new 6299 EH tubes just arrived from Arizona, via USPS Priority!

 

[JoseHH]

Great. This is excellent, because it means that the hfe matching of Q2Q3 and Q5Q6 is good. No need to change them. I am glad you could adjust your DC Offsets to specification.

If you are confident, set yout DVM to measure current (watch out where you plug the dvm probes for this: most dvms have a separate socket to measure currents, in a 10Amp scale. For your first measurement, i suggest you to use the 10A scale because the ST-140 uses a 200 mA bias current and this is in the limit of most DVM low current scale which is typically 200mA.
If your DVM has a 2000 miili Ampere range, then that would be perfect. If not, use the 10Amp scale.
What I suggest is NOT to adjust it just yet, just measure and report here.
Remember, that you need to power the unit off, remove the V- fuse and securely clip the dvm probes, preferentially, put the red clip on the socket clip that is closest to the corner of the board so you read a possitive current.
Then power the unit without any speakers connected, and without anything connected at the RCA inputs.
Leave the unit run for a couple of minutes and write down the idle current.
Power off and repeat on the other channel.
Let us know what do you get.

 

[NeilBEN]

Thanks Jose - I will measure the current across the V-fuse clips, and let you know. I have alligator clips for my meter.

Just to confirm - the fuse I need to remove is on the edge of the circuit board next to the Molex connector? (Not the fuse in the middle of the board, closer to the heatsink.)