• 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

Harman Kardon 330C power board Q601 pnp

CalvinBeck

CalvinBeck

Member
HK_330C_schematic_top.jpg HK_330C_schematic_bottom.jpg H-K_330C_power_board_pcb.jpg Long time lurker, first time poster. First let me say I am continually impressed by the helpfulness and free exchange of knowledge I have seen on Audiokarma. I've only been working on electronics as a hobby for three years. I went through that "know enough to be dangerous" phase, moved on to shotgunning, but things have been starting to click lately. I'm pretty shy about posting anything publicly. I don't use social media. So this is hard for me to do, but after a week of being stumped, I decided it's time to start participating and ask the community for help.

I have a Harman Kardon 330C that I acquired for a slightly inflated price from the auction site. When it arrived I was pleasantly surprised at its condition. It's beautiful, barely a scratch on it. The controls were dirty (dang tape monitor). Other than that, it worked. Clean sine wave, great square wave, negligible dc offset, barely had to adjust the bias. It put out nearly 30 watts. All of the voltages on the pinouts of the power board looked good, with the exception of two which were about a volt and a half low. Where +B2 and -B2 are on this machine, I have no idea, so I couldn't check those.

So why not leave well enough alone and just enjoy the thing? I suspect there would be fewer of us that do this as a hobby if we could answer that question.

There were a few slightly bulging caps and considering they're all 45 years old, I decided to replace them. I did the power board, tone board, am/fm board. Tested on the dimbulb and listened after each board. Everything seemed fine. Then I decided to get a little radical and change the diodes on the power board. Why, I have no idea. I should mention, the filter caps hold voltage for a long time. C3 will hold -6.5V for over a day and C4 will hold 0.5V. Knowing this, I bled the voltage down to 1mV with a 1k resistor. I installed the four diodes D4, D5, D601, and D602 with UF 4003-E3/73s. The next day, to my horror, I discovered that I had installed D4 and D5 backwards, and C3 had gone back up to -6.5V. I don't know if this voltage went somewhere it shouldn't have, but after putting them back the right way and testing the voltages I found that I had no voltage on LP11, which is tied to the collector of Q601.

I've gone on long enough. Suffice it to say I don't know if my diode mistake is a coincidence, since the diodes I reversed appear to be connected to a separate circuit with its own rectification and filtering, than the ones on the high voltage rails.

If anyone can explain to me where Q601 gets its Vcc and where that voltage/current could be, I'd really appreciate it.

I have attached three diagrams of voltages I have measured.
 
Last edited:
Register to hide this ad
The way this reads it sounds like the unit was never powered up with the backwards diodes? Or was it? on a dimbulb? What does it do on a dimbulb now?

Have looked very closely at your solder work on the power board after cleaning up the flux residue and drying it?

Also maybe post a picture of the top and bottom of the board that shows your work on the four diodes, somebody may be able to spot something.

It would seem likely that your answer is within that scope of work since everything was working until you did that.

Lastly did you save the 4 old diodes? If you get stuck here, test them with a multimeter and if they still work you can put them back in and see if that fixes your problem. If it does then I would test the four replacement diodes with the meter, perhaps one was damaged by being backwards.
 
Unit was never powered on with the backwards diodes. Just sat overnight with that -6.5V on C3 filter capacitor. I powered it on with the new diodes on the dim bulb and it reacted normally. The pinout voltages were not as close to spec as before the work. I tested them with my Peak Atlas DCA 55 and they were fine. The old ones tested fine too but had a higher voltage drop. In an effort to retrace my steps and get back to “stock” I put the old ones back in. On the dimbulb they work fine. Also work fine on mains power. The diodes I reversed originally seem to only power the lamps and am/fm board according to the schematic. That circuit, which is powered by the lower voltage green wires coming off the transformer secondary, works fine now. The problem I’m having is I’m getting almost no voltage on the collector of Q601. It’s supposed to be -36.6V. Upon powering on it briefly peaks at -27.5V then drops quickly once C601 and C602 charge and the Vb of Q601 rises to -0.55V.

The rectifier/voltage divider circuit surrounding Q601 is powered by ~21.5VAC coming off the transformer. Once it passes R601 (5.6k) the voltages/voltage drops add up perfectly according to KVL. To me, this little circuit appears to be functioning perfectly. The emmiter of Q601 is tied securely to ground (I have double checked), the base is getting -0.55V (spot on), but the collector is just sitting there at 47mV. Where does the transistor get its Vcc?

-B1 (about -28.5V) comes off of the negative side of the D1 bridge rectifier on the power board. From there it goes to each half of the main amp. For discussion I’ll just use the even numbered side. Once the -28.5V reaches the main amp it makes a negative rail. There is a positive rail on top powered by +B1 (about +28.5V) that comes off of the positive side of D1 on the power board. Both rails have parallel branches that power the various amp transistors.

From where -B1 enters the amp board (on the right side of the schematic) it’s pretty much a straight shot to the left side. The voltage has to pass through R444 (47k) which is in parallel with C406, and through D402 (Zener, marked incorrectly on schematic, presumably greater than -28.5V). Finally there is R442 (820k). At LP413 with the wire going to the collector of Q601 disconnected I measure about -26.5V. This is close to that peak I get on the collector as soon as the unit is powered on.

I don’t know how the collector voltage, which is supposed to be -36.6V, is related to -B1. All I know is somehow the collector is being fed through the negative voltage rail of the main amp, minus the resistor and diode drops.

That’s about all I know right now. I look forward to your comments. I really appreciate the help.
 
You are clearly more versed in electronics than I am. I may be of little more help. Your write should help another chime in perhaps. I recently worked on an HK330C power board. My board had no markings on it, does yours? When you installed Q601 how did you determine ECB position? In my case I used the datasheets for both transistors. They both read ECB so I installed my new transistor the same way. After I did that my unit powered up, lights came on, passed dimbulb, and all my audio circuits were dead - no sound anywhere. A real head scratcher. I installed the old Q601 back in, everything worked again, hmm why. After I stared at the datasheets I realized one showed ECB but the transistor was face down, the other showed ECB and the transistor was pictured face up, so they were not the same pin out as I had thought. I put my new transistor in again, this time backwards so I would match the pinout of the old one. The unit worked again.

I am not saying you put Q601 in backwards, but I just wanted to relate a similar problem that I had and what I did wrong.

All I can suggest at this point is that you heavily scrutinize every component you replaced on the power board, checking to make sure nothing is backwards. Next would be solder joints, looking for broken or newly bridged joints.
 
Thanks for the reply Mr. Oates. There are definitely errors in the service manual. My board is unmarked as well. The other boards are marked, but most of the labels are underneath the components, making most of them impossible to see. The picture of the power board in the SM shows Q601 as an NPN. I searched for 2sa844E and the data sheet says it’s a PNP with base-collector-emitter leads looking at the transistor flat face forward. BTW the “a” in a BJT’s name indicates that it’s a PNP. Upon removal it tested good on my Peak Atlas semiconductor tester, which also identifies the leads. I tested it with a DMM and it passed that test too. So I knew it was an B-C-E transistor before I tried any others. I mostly work on Marantz receivers so most of my stock is of the emitter-collector-base variety looking at the flat face with the code. The solution here is to just simply turn the transistor around facing 180 degrees opposite the 2sa844E. I tried two ksa’s (can’t remember the numbers right now off the top of my head) with the same results as the original. Right now I have a ksa992FATA, gain 440, installed. It’s underrated, but I’m just using it for testing purposes. It’s given me the best base voltage of the four I’ve tried. I don’t think 4 transistors could be bad, so I don’t think that’s the problem.

Anyway, I’ve been scrutinizing my work for a week and I can’t find the problem. That’s why I finally decided to join the community and ask for help.

Seems like ages ago, before GPS, when we would try to drive and read a map at the same time. I’m one of those men who would Never stop and ask for directions.
 
Thanks for the diagram Mr. Pete. My pnp problem is a little different. The emmiter is tied to ground without a resistor, the base is at -0.55V, and the collector is supposed to be at -36.6V. It’s currently at 47mV, basically nothing. Have you seen anything on the web that looks like this circuit? Where is it supposed to get its collector voltage from, Vee I mean?
 
CalvinBeck said:
I should mention, the filter caps hold voltage for a long time. C3 will hold -6.5V for over a day and C4 will hold 0.5V.
Click to expand...

Yep, that is normal.

CalvinBeck said:
Knowing this, I bled the voltage down to 1mV with a 1k resistor. I installed the four diodes D4, D5, D601, and D602 with UF 4003-E3/73s. The next day, to my horror, I discovered that I had installed D4 and D5 backwards, and C3 had gone back up to -6.5V. I don't know if this voltage went somewhere it shouldn't have, but after putting them back the right way and testing the voltages I found that I had no voltage on LP11, which is tied to the collector of Q601.
Click to expand...

As stated in the other thread, Q601 is "supplied" from LP413 and LP414 via LP11. Make sure all those connections and solder joints are good.

Are C405 and C406 OK? Are they installed (positive and negative leads) correctly? They are in the path to LP413 and LP414. Caps can easily blow or fail when their voltage is reversed.

Are D401 and D402 working properly? They are in the path to LP413 and LP414, too.

Are you getting the full voltage at LP403 and LP409? This is where that "through the main amp" rail goes.

Lastly, you maybe fighting an incorrect schematic. I don't know where anything in the 330c would be getting 36V. More likely they meant 26V. If you're getting -26V there and the unit is operating properly, chalk it up to mistake in the service manual.
 
I'm headed up to my bench now. I'll double check what you mentioned. I'm uploading a schematic that I put together in Paint Shop. Hopefully the zoom function will make it readable. It's as neat as I could make it.H-K_330C_schematic001.jpg
 
Wires from LP413 and LP414 test good.
D601 and D602 tested good when I installed them. They rectify the incoming AC.
C601 and C602 are oriented with positive to ground. They are filtering the ripple.
I measure -28.85V at LP403 and -28.83V at LP409. I measure -26.62V at LP413 and -26.63V at LP414 WITH the wires going to LP11 disconnected. Connected, I measure -50mV at LP11. The base of Q601 is at -0.556V, 4mV above spec. The Vb on the scope is a rectified and filtered straight line at 0Hz.

I downloaded the service manual from hifiengine. There are two scans of it. Both of them show the collector of Q601 at -36.65V. I measured LP11 at -36.6V when I first got the unit. If the SM is wrong why did I measure that? I do see your point about this model not needing -36V for anything. The next greatest BJT voltage is -28.55.
 
I'm no E.E. and I'm OCD crazy, so I'd pull C602, C601, D601, D602, and Q601... and test each of them. And triple check Q601 pins are installed properly.

If you've got enough saturation voltage (?) on the base, then Q601 should be open and letting current through... and you should see 26v at its collector.
 
I’ve got a worthless Associate’s degree in “Electronics Service Technology” from a local community college that’s not worth the paper it’s printed on. They didn’t teach us anything practical. Theory is great, but it doesn’t mean you know what you’re doing.

Anyway, I’m at the point with this thing that I’m open to anybody’s suggestions. OCD? I’ve got a Ksa992FATA standing in for my Q601. I’ve got it wired with 24 awg wire into the board with the emmiter connected to common ground point. This way I can break any connection and insert a meter to measure current. Probably a little unorthodox. Just measured base current at 60uA (Hfe 440, 26.4mA if my math is correct. That sounds like plenty to cause conduction.
 
I'd do the last things dlucy suggested, then report on them. That's the best way to keep help coming. You will crack this, just need to follow through.

AND - take a picture of the top and bottom of the board, post them. You never know it could be something easily overlooked.
 
Q601 is part of the turn on circuit. When the unit is powered on, initially there is no voltage on the base of Q601. C601 (not Q) should charge "down" to -6 volts or so, while C602 charges "down" more slowly. When the base of 601 gets to - .6 volts or so, the transistor conducts and pulls the collector up to near ground - it is basically a delayed switch. the collector goes back to the input stage of the amplifier. It will start at -26 volts until q601 goes active, then will pull up to allow the amplifier to operate. Tthe function is to minimize turn on thump. The delay is a function of C602 - if this were open everything would work but turn on thump would be much louder. So, after turn on is finished, there will be nearly no voltage on the collector.
 
Exactly as Steve said. If you disconnect Q601, the power amplifiers will not work because the current sources for the differential input stages will not work.
 
Thank you for the information, Steve. That explains a lot. What you describe is exactly what I have been measuring. Upon powering on I get a flash on my meter, connected to LP11, which is tied to the collector of Q601, of -27.4V. I'm also measuring the Vb at the point between R604 and R605. I've observed the charge down effect. As soon as that starts, the collector voltage rapidly increases to about -50mV. The Vb levels out at -0.556V (SM says -0.56) enough to break down barrier potential? It must be if collector is nearly grounded and at -50mV
 
I'm happy to report that my receiver is up and running again. My problem was that I never had a problem in the first place. Goes to show you can't always trust the SM. I'm so glad I learned the purpose of that little pnp circuit. A big thank you to sregor for explaining it. I tested the power board voltages, the pre-amp and main amp transistor voltages, and scoped the pre-amp and main amp. Everything looks good. Sounds fantastic. Along the way I discovered that my function generator is putting out garbage on one channel, so I had to do my testing on the one good channel. It's just a cheap no-name job I got when I first got into the hobby. If anyone has any suggestions for a replacement that won't break the bank, I'd appreciate it. All of your comments and suggestions concerning my non-problem were most welcome.
 
You must log in or register to reply here.
Back
Top Bottom