Harman Kardon FA-3000X questions

[Ziradog]

Started my refurb on this unit last night. I DBT'ed it, filled the empty tube sockets & fired it up. As received, it does play but has bad hum and the tuner is definitely wonky - stereo light comes on only when stations are mistuned to give loud noise. I did a basic chassis & faceplate cleaning, cleaned the posts, switches, and tube sockets, and started on the power supply caps. That's where it got interesting.

On the schematic I have (attached), R110 is 3900 ohms. On the actual unit, the original looking resistor is marked and measures 39K ohms. 3900 is more in line with what the other sections of this can cap use (they have 4700 ohm resistors, which match the schematic. Any thoughts on which is right? Anyone else have on of these they could look at?

I also found that R107 (1500 ohm resistor) was burned completely up, so I will probably replace with a higher wattage rating - 1 or 2 watts instead of 1/2 watt. Any risk there?

Thanks!

 

[gadget73]

Is the voltage out of that 39K correct ? Gut feeling is that 39K is probably accurate.

R109 is a 4.7K and its dropping 45 volts, so the current there is basically 1ma. Current there is going to be more than what is flowing through R110. R110 needs to drop 50v and you can't do 50v at less than 1ma on a 3.9K resistor.

that is of course assuming 4.7K is correct, and the voltages are reasonably right.

generally speaking I go with whats in the unit unless it shows obvious signs of having been meddled with, or it gives me some solid reason to believe it needs to be altered. It worked when it left the factory, so if the parts are original it should work if you leave the values alone.

 

[Ziradog]

Thank you sir, I just needed a sanity check. I haven't checked the voltages yet as I found out I do not have ant 2700 ohm resistors after tossing out all the old parts attached to that can. And since I cannot run down to the local Radio Shack (RIP) today, I will have to piece some together until I can order some.

 

[Ziradog]

Interestingly, the other 3 voltages on this cap are 5 - 6 volts high (probably because my wall voltage is high), but this one is 16 volts high - almost 10%.

 

[gadget73]

might depend on what mode its in, and possibly on the meter. If you're using a digital it will read higher than an old 1k/volt model will just because its not adding load to the circuit. Also depends somewhat on the condition of the tubes connected off that tap.

 

[Ziradog]

Well the power supply caps are done, coupling caps are next. Does anyone know what these used for dial bulbs? Thanks.

 

[gadget73]

Schematic says #47

 

[Ziradog]

D'oh! Thanks.

 

[Ziradog]

So coupling caps and dial lamps replaced. Still some buzz/humm but much better & hopefully putting the case back on will help that. The tuner is still not good - it will pick up stations, but the stereo light still only comes on when local stations are mistuned and the pointer is off. I don't have the tools or skills to do a real alignment. Any chance that this is something simple, or at least simple to check? I have not tested the tubes.

Also, there are two balance pots, one for each channel. Do I use these to match the bias current on each pair I assume?

 

[Ziradog]

I didn't expect much help with the tuner question, and digging around I see that these are common problems with these units.

I would still appreciate input on the balance pots. I've had no luck finding a service/owners/SAMs for this.

Thanks.

 

[gadget73]

you can try tweaking the antenna trimmer to see if it pulls in any better. No special alignment gear needed, just peak it for best reception.

balance pots are to get equal current draw in each channel, but the original instructions probably say to adjust for minimum hum which ought to be the same thing.

 

[Ziradog]

Thanks. I'm not used to playing with the tuning caps, but I guess I have nothing to lose. On the balance pots, I guess add the 10 ohm resistors to check cathode current (which I haven't done yet) & make it equal? Given the mis-matched output tubes I have, that should be interesting. Guessing I should see (ideally) ~35 mA on each tube?

 

[gadget73]

The problem with adding too much resistance into that circuit is that it changes the amount of bias the tubes get, and since it uses tube heaters as the resistor, you also mess with the voltage those see. If you can tweak it for minimum hum, that will be enough. Scope across the output leads may be easier to use vs your ears though.

You can't really set it to a level, it pretty much comes out where it comes out based on the tubes. This one really wants a matched quad if at all possible. In theory it should be about 37ma per tube but it may well be slightly under that.

 

[Ziradog]

I tried adjusting the balance pots by ear, but honestly heard very little change. There is noticeable hum with the volume all the way up, so I'm kind of surprised. If I use an oscilloscope, can I use speakers for a load or do I need to dig out the pair of non-inductive resistors I have buried someplace?

Still trying to figure out how to adjust the antenna trimmer. I ran the FM ~5 hours yesterday, almost no drift. Sound was mediocre though.

 

[gadget73]

Speakers as the load will be OK for this.

You may actually be better off adjusting it with the volume all the way down. That will limit hum to just what the output stage produces, and thats all that the balance pot is going to change.

It would be worth checking R114 and R115. Those act as a virtual center tap on the heater string. If either are way off value or just open it may tend to hum more. The actual values aren't all that critical, as long as they are close to each other. I usually use 100 ohm resistors for that job but I have a large bag of them.

The antenna trimmer you'd just peak for best reception. Adjusting it should make the station stronger or weaker depending on the adjustment. Its easier to use a station that doesn't come in terribly strong already so the peak is more obvious. You should be able to peak the RF trimmer in the same manner. All of the rest of that stuff you really need proper alignment gear to deal with. Weak tubes will also mess with performance.

 

[Ziradog]

I tried doing it by ear, but there is very little hum & adjusting the pots mad no discernible difference. I hooked up a cheap digital oscilloscope I borrowed from work, and it wouldn't even trigger unless I turned the volume all the way up. Again, the adjustment made no real difference. This was with a 1X probe, 10 mV/division range (smallest it had), using non-inductive 8 ohm resistors. Also with the bottom & tube shields in place. which helped a lot.

I'll check those resistors. How close is close enough - 1%?

The problem with the antenna trimmer is I can't really get to it and can't even see what tool it takes - I assume a straight screwdriver. The can capacitor there is too close to be able to see or reach it.

 

[gadget73]

even 5% is probably fine, but the closer the better.

 

[RWood]

I just did this on an A500 (different power tubes but similar bias balance method). I put the probes (actually Pomona Mini-grabbers) on the cathodes of each pair of power tubes and set the meter to DC volts. Polarity of the probes does not matter. Volume at 0, amp good and warmed up, adjust the balance pot to read as close to zero on the meter as you can get it.

Since the owner of this amp has eight power tubes and wants to potentially swap them around, I made it easy for him to measure this without going under the amp where all the high voltage lives. I put two sets of terminal strips on the rear apron of the amp and led wires from the cathodes, each to its own terminal. Now, to adjust the balance he can put his probes on these terminals, a pair at a time, and access the balance pots from above deck, adjusting for zero.

If you then want to see how close (or not) the tubes are matched, from the standpoint of idle current draw), you can probe the cathode to ground and compare voltage from one to another. As Gadget correctly pointed out, you cannot adjust the overall bias level of the power tubes, just the balance. But if you make sure your bias / filament string has proper resistance and new capacitors, it should get you within a very comfortable range for the 7408s.

 

[gadget73]

If you want to do some fiddling you can get individual adjustments for each tube but its debateable whether its worth it. I have that setup on a roughly similar Fisher TA-600 to let me use very mismatched tubes but tubes that are reasonably well matched won't have that kind of problems.

 

[Ziradog]

I'm not planning of adding individual adjustments. I did try the adjustment method that RWodd suggested. I also checked R114 & R115 - one was ~25% high so I replaced both with the best matched pair I had (~2%). Thanks gadget for suggesting I check that. I stuck with 33 ohm - just curious why you went with 100?