• The move to the new server is done. There are some software and database maintenance updates in process. This has us passing the hat around to help out. We appreciate any donations. Seriously, even a dollar helps. The payment page may be found here - https://www.audiokarma.org/support.html

Proper/Recommended procedure for setting bias

tensleep

tensleep

Addicted Member
OK,

I've got my HK 330c, my DVM, my tech manual for the receiver and a clean work area. The lid is coming off!:banana:

Here is on the list to do:

  1. Clean the interior with compressed CO2
  2. Replace suspect wire ties (some have wire cores, can short out)
  3. Inspect for leaky capacitors
  4. Power on receiver
  5. Set bias

I have the following questions:

  1. What are the pros/cons to using a variac for initial powerup?
  2. The HK manual specifies a 50W 8 ohm resistor as a load between the speaker terminals. I have spoken to a local audio tech and he doesn't use a load resistor to set bias. Is this unique to this receiver and what method do you use?
  3. Is there something special about Deoxit, or will other products work, such as CRC Contact Cleaner?
  4. Should DC Offset be measured at the output terminals before and after bias is set?

Please make all the suggestions you like - I've got thick skin. Piccies will be coming as I progress.
 
Register to hide this ad
Let me get some food, sleep, and a little research done, and I'll try to help you out. Also, give me time to find the 330c hiding someplace in my garage...
 
I got started on the receiver tonight. I cleaned the outside some and opened the case - man, is she dirty inside. I spent the rest of the night chasing dust bunnies.

Other than the dirt, the interior apprears to be intact - all of the circuit boards are there, as are all of the switches, knobs, terminals, etc. A couple of questions:

  1. The dust has gotten "tacky" and is not cleaning very easily - any suggestions?
  2. The thermal grease between the power transistors and the mica shield has gotten old and is somewhat dried up. Should I pull the transistors up, clean, regrease? I suppose that the solder joints will need to be inspected for cracks if I lift up on the transistors.
  3. The tuner string is hanging a little on one of the hooks as the dial turns. The string is secured to the hook with a knot and then, it seems as if some kind of plastic seal was then used to decrease the possibility of the string coming untied. However, the plastic seal overhangs the track in the pully that the string rides in, causing it to hang a little.
  4. One of the speaker fuses was open. I checked the power supply fuses and they are intact. I forgot to check their values; I will check them tomorrow.
  5. I intend to use a variac to bring the receiver up for the first time. Any suggestions?

Well, it's bedtime. Sweet dreams!
 
You can persuade the dust a little using a soft bristled paint brush,then continue with the compressed air.Yes a redo of the silicone heatsink grease would be a good idea after.....checking the output transistors for shorts with your DMM, especially the channel with the open fuse...before trying to power up the said unit.
 
Pictures

Here are a few pictures of the receiver:

HK+330c+repairs+060508_jpg+006.jpg

HK+330c+repairs+060508_jpg+007.jpg

HK+330c+repairs+060508_jpg+008.jpg


I will expose the bottom of the chassis later and clean it out today (got the afternoon off:banana:).

So I need to get some heatsink grease - unless told otherwise, I will assume that all are equivalent and price/availability will dictate which I use. As far as testing the output transistors go, does anybody have a suggestion on how to test them for shorts without removing/desoldering the transistor from the circuit board?
 
To test the power transistor "in situ": No speaker load, power fuses removed.

Since all power transistors are NPN, you should have conduction from Base to Collector, and from Base to Emitter, but NO conduction in any other direction. In your circuit, you will measure about 220 Ohms from Emitter to Base, it's OK (R419 to 422). Some DMM are not friendly when comes the time to check transistor with this method, their voltage source are too low to forward bias the pn junctions. If your DMM has a Diode check mode, use it in this mode. I prefer to use an old analog meter for this test.

If you want to test the other transistors, keep in mind that for a PNP the conduction is from Emitter to Base, and from the Collector to Base, but NO conduction in any other direction.

Always check for other components across the junction, like R419 that shunts the BE junction in Q419. They may fool you!

Since you intend to replace the thermal compound for the power transistors, why don't you remove them to check them?
 
ecluser said:
Since all power transistors are NPN, you should have conduction from Base to Collector, and from Base to Emitter, but NO conduction in any other direction. In your circuit, you will measure
......................
...............
Click to expand...

Anyone else's pants get tight reading this?
 
ecluser said:
Since you intend to replace the thermal compound for the power transistors, why don't you remove them to check them?
Click to expand...

Yeah, common sense and experience tells me to pull 'em. I had better not start rushing. I will do a before and after test to verify your method for future reference. Thanks for the help!

specialidiot said:
Anyone else's pants get tight reading this?
Click to expand...

Mine were; I suppose I should do this the proper way - desolder, test, clean, remount. I should also retie the tuner string - it would eventually wear through, the way it is.
 
Nat said:
I'm curious -- why are all power transistors NPN?
Click to expand...

In the days of the 330C silicon pnp power transistors where more complicated to make and more expensive. So this design had complementary npn-pnp (low power)driver transistors, called a quasicomplementary poweramp design.

Helmut
 
Answer to NPN question

I believe on the HK 330c specifically, all of the output transistors are NPN. In the main amp section of the schematic, all of the transistors are marked as NPN. Thanks for the extra information, Helmut. I will study that circuit more closely.
 

Attachments

  • bipo-sym.gif
    bipo-sym.gif
    2.1 KB · Views: 16
So what was meant was that all of the power transistors under discussion in this thread are NPN, not that all power transistors in general are -- is that correct? I'm not as well informed as I should be.
 
For clarity:

  1. All final output transistors in the Harmon Kardon 330c are NPN. However, the main amp circuit employs both NPN and PNP, according to both the schematic and Helmut's commentary.
  2. Once the discussion turns to output transistors in general, many different devices are used and I'm no expert.
  3. I understood the original comment by ecluser to be aimed specifically at my question regarding in circuit (in situ) testing of my receiver's output transistors, which are NPN type, NOT output transistors in general.

FWIW, I intend to pull the transistors out of the circuit and test them. I will also cross ref the spec sheets against measurments i can take. Please feel free to make suggestions on the process. More pics to come.
 
Nat said:
So what was meant was that all of the power transistors under discussion in this thread are NPN, not that all power transistors in general are -- is that correct? I'm not as well informed as I should be.
Click to expand...

You are correct Nat.

In general, you will find a NPN transistors on the final stage for the + supply, and PNP transistors for the - supply. But many combinations are possible.

In cap coupled output, it is common to see the same power transistors.

The fact that they are similar implies that they should conduct/not conduct in a similar way, unlike NPN vs PNP.

I also gave a way to test the PNP because some drivers are PNP, and it would be wise to test them also.
 
specialidiot said:
Anyone else's pants get tight reading this?
Click to expand...

It's not that complicated to test transistors in the circuit for a quick check, as long as we don't forget the other components connected to them. Sometimes, we have no choice, we must remove them.


Keep in mind that each time you unsolder and resolder a transistor, you may damage the transistor and/or the board. I don't know what is the soldering skill of every AKers.
 
ecluser said:
It's not that complicated to test transistors in the circuit for a quick check, as long as we don't forget the other components connected to them. Sometimes, we have no choice, we must remove them.


Keep in mind that each time you unsolder and resolder a transistor, you may damage the transistor and/or the board. I don't know what is the soldering skill of every AKers.
Click to expand...

To try to help everyone out a little bit, if I can that is:

With most output-sized NPN transistors, putting the positive probe of the ohmmeter on the base, and going to either the emitter or the or the collector should show anywhere from quite a few hundred KΩ of resistance, to a few MegΩ of resistance. Reversing the probes, i.e., negative on the base, should show infinite resistance. Probes either direction (and check both ways anyway) from emitter to collector should also show infinite resistance. Point being, don't expect a low resistance in any direction. If one is seen, pull that transistor and check again.

Another helpful check, if your meter has a diode checking function, putting the positive probe on the base, and touching either the collector or the emitter with the negative probe (again, this is just NPN), you should get a reading of about 0.5-0.7V. You should get no reading in any other combination of pins and probes, period. If you do, pull the transistor and check again.

If the transistor checks fine out of circuit, you have trouble elsewhere, begin your hunt!:D

If your meter has a transistor check function (Hfe), DON'T use it, it is only strong enough to check small signal transistors.

Now, looking at the photos, I would suggest replacing the bias potentiometer before proceeding. The last one of that age that looked like that one crumbled the second I tried to adjust it (yep, with the amp powered on even). Don't worry too much about it being powered up though, because it'll go to full resistance if the two outside pins stay intact, and as long as the wiper doesn't short any of the other pins when it falls off, or if the screwdriver slips and shorts any of the pins, or. . . well, nevermind, just turn the sucker off quick if you keep the original pot and it breaks . . .:D

. . . This is also a wonderful time for me to suggest getting a very small PLASTIC screwdriver for when the adjustment time comes, new potentiometer or not.
 
BrassTeacher said:
To try to help everyone out a little bit, if I can that is:

With most output-sized NPN transistors, putting the positive probe of the ohmmeter on the base, and going to either the emitter or the or the collector should show anywhere from quite a few hundred KΩ of resistance, to a few MegΩ of resistance. Reversing the probes, i.e., negative on the base, should show infinite resistance. Probes either direction (and check both ways anyway) from emitter to collector should also show infinite resistance. Point being, don't expect a low resistance in any direction. If one is seen, pull that transistor and check again.

Another helpful check, if your meter has a diode checking function, putting the positive probe on the base, and touching either the collector or the emitter with the negative probe (again, this is just NPN), you should get a reading of about 0.5-0.7V. You should get no reading in any other combination of pins and probes, period. If you do, pull the transistor and check again.
Click to expand...

This is exactly for this reason I said DVM are not friendly for testing transistors. If they don't have a diode check mode they may fool you, the transistor being in or out of the circuit.

For instance, a 2N3055 test with a DVM gives about 1MOhm for the BE and BC junctions (forward biased), 0.45V in diode check mode. I have 10~11 Ohm with an analog meter. With such a meter, the 220 Ohm resistor across the BE junction (Q419/420) would cause no problem. I seldom need to remove a transistor to see if it is opened, shorted or reasonably good. A quick check like this has some limits though. For a more precise test I use my venerable Tek 575.
 
We're going in!

I have made some initial measurements (see attached word doc) and have exposed the output transistors. I have not pulled them yet, I don't have that much time tonight.

  1. I attempted the diode test, Base to Emitter was between 0.5 and 0.7 VDC. However, all other combinations tested less than 0.4 VDC, so I don't trust the measurements.
  2. I intend to get the thermal grease at a Radio Shack. Do they also sell the little shims that mount between the device and the heatsink (if I remember right, these are an electrical insulator with good thermal conduction)? If not, where can I source those parts?
  3. I tried using an analog meter, but those "negative" resistance readings were moving the needle in the wrong direction. I'm sure it has to do with fwd/rev bias, but my theory escapes me.

Please comment/advise.

HK+330c+output+transistor+test+060708_jpg+001.jpg


HK+330c+output+transistor+test+060708_jpg+002.jpg


HK+330c+output+transistor+test+060708_jpg+004.jpg
 

Attachments

tensleep said:
I have made some initial measurements (see attached word doc) and have exposed the output transistors. I have not pulled them yet, I don't have that much time tonight.

  1. I attempted the diode test, Base to Emitter was between 0.5 and 0.7 VDC. However, all other combinations tested less than 0.4 VDC, so I don't trust the measurements.
  2. I intend to get the thermal grease at a Radio Shack. Do they also sell the little shims that mount between the device and the heatsink (if I remember right, these are an electrical insulator with good thermal conduction)? If not, where can I source those parts?
  3. I tried using an analog meter, but those "negative" resistance readings were moving the needle in the wrong direction. I'm sure it has to do with fwd/rev bias, but my theory escapes me.

Please comment/advise.

HK+330c+output+transistor+test+060708_jpg+001.jpg


HK+330c+output+transistor+test+060708_jpg+002.jpg


HK+330c+output+transistor+test+060708_jpg+004.jpg
Click to expand...

The Word doc clearly shows that you were measuring R419/420 when testing the BE junction of Q419/420, and R421+429/R422+430 when testing the BE junction of Q417/418. At least you know the resistors are good.

Since the transistors measure about 1Meg in all direction with the DVM in resistance mode, they don't seem to be shorted.

Did you remove the fuses from the power supply, as I suggested? I don't see how you can have a "negative resistance" with the analog meter, and a low voltage reading in any direction with the diode checker, except if the power supply is still connected to the circuit. Do you have less than 0.4V from Emitter to Base ??
 
You must log in or register to reply here.
Back
Top Bottom