Harman Kardon A-402 Restoration

[jheu02]

I just received this amp and want to give it a listen before I start the restoration. As always, I first checked it on DBT, no shorts. Then DC Offset... both channels were over 100mV...hesitant to hook to speakers. Then check Bias IAW the SM to see if that could be affecting the DC Offset in any way (doubtful, but always good to check/set anyway). That's where I'm a little stumped perhaps. IF the SM is the correct procedure (assume it is) it's a real PITA to get minigrabbers on the legs of each resistor mentioned (especially the lower legs due to the main filter caps being so tall. Also, the nice ceramic leg protectors leave very little metal exposed for the grabbers to grab. So, I'm not positive that I got a good connection. However, if i did, it was not pretty... 70ish mV DC on each channel and the trimmers didn't appear to do much for the time I was comfortable leaving it on. Will reattempt later today. Could the SM be incorrect? I find it hard to believe they would make it this difficult to attach a multimeter for a routine adjustment. Can anyone confirm from the pics added if I'm doing something wrong, or if there's another way to get an easier measurement location?

 

[jheu02]

There's a 730 thread going and the SM for it says to connect to the collectors of the outputs in each channel. Voila! When I did that with this amp, I got readings close to 15mV...so adjusted the trimmers to maintain close to the 15mV as called for. Offset is still high, but at least I can test the sound before tearing things apart. First order of business when I do is to get some nicely matched Qs for the differentials 401/403 and 402/404.

EDIT
Using this procedure, the adjustment should be to 35mV across the collectors not 15mV. See below posts

 

[leesonic]

I'll be watching this thread with interest, since I have one of these amps that I want to restore for a friend.

As a related note, I always bring the test points out on a couple of pieces to scrap wire. It makes it much easier and less risky than trying to test with clips or probes.

Lee.

 

[sregor]

Well, I discovered that the SM instructions are WRONG! There's a 730 thread going and the SM for it says to connect to the collectors of the outputs in each channel. Voila! When I did that with this amp, I got readings close to 15mV...so adjusted the trimmers to maintain close to the 15mV as called for. Offset is still high, but at least I can test the sound before tearing things apart. First order of business when I do is to get some nicely matched Qs for the differentials 401/403 and 402/404.

The manual isn't wrong, it's just that it's easier to go collector to collector - but now you are measuring across both emitter resistors, which, I believe, means that it should be 30mV if you are going that route, Or you can go from one collector to the output for 15 mV. The collectors connect to one side of the resistors, and the output is at the junction of the 2 resistors. All routes should give you the same (and correct) results.

 

[jheu02]

Thanks for bringing that info out. It's virtually impossible to get a good connection across the resistor with the way the amp is set up. Much easier to clip onto the collector extension between the heatsink and the board. That's what the 730 manual has you do. Looking back at that manual and adjustment procedure, it does say to adjust to 35mV +/- 5mV, so will need to go back in and readjust and then give a listen.

@leesonic... I would have made test point extensions, but that would have involved removing the amp board too...no way to get an iron cleanly to the resistor leads. I wanted to give a listen and get some initial impressions before the restoration. While back in there I'll also take some basic voltage reading to make sure the PS is functioning properly.

 

[dlucy]

First order of business when I do is to get some nicely matched Qs for the differentials 401/403 and 402/404.

I'm using KSC1845 in the Q401-Q406 positions in the hk730 and having excellent performance.


power amp Q401-Q406 2SC1400 (UA) ECB hfe 700 --> KSC1845 (F)
power amp Q407,Q408,Q413,Q414 2SB549 (Q) --> KSA1220A
power amp Q409,Q410 2SC853 (L) --> KSC2690A
(this line could be wrong; I'm checking up) power amp Q411,Q412,Q415,Q416 2SD415 (R) --> KSC2073 H2
power amp Q417,Q418 2SB541 (S or R) MJ15003G+MJ15004G (TO-3) or MJL21193G+MJL21194G (TO-264) or MJL4281AG+MJL4302AG (TO-264-3)
power amp Q419,Q420 2SD388 (S or R) MJ15003G+MJ15004G (TO-3) or MJL21193G+MJL21194G (TO-264) or MJL4281AG+MJL4302AG (TO-264-3)
Tone control and local rectifier Q501,Q502 2SA844 (E) signal --> KSA992 (F)
Tone control and local rectifier Q503-Q508 2SC1775 (E) signal --> KSC1845 (F)
Main rectifier Q6 2SD1162 (C) power --> ???
EQ (PHONO) Q701,Q702 2SA872 ( E ) ECB KSA992 (F)
EQ (PHONO) Q703,Q704,Q705,Q706 2SC1345 (E) BCE --> KSC1845 (F)

 

[Binkman]

I'm using KSC1845 in the Q401-Q406 positions in the hk730 and having excellent performance.

wow.. quite a list.. Did you ever do a complete replacement job using all ?
(I remember your affinity to be uhhh detailed.. do em all.

 

[leesonic]

I was making up a parts list today, even though I don't have the amp yet. I thought about going up to 6800uF 50V caps for the mains. @jheu02 do you have the diameter and height (not so important).

@dlucy I was thinking the same thing about replacing the small signal transistors as well.

 

[dlucy]

I was making up a parts list today, even though I don't have the amp yet. I thought about going up to 6800uF 50V caps for the mains. @jheu02 do you have the diameter and height (not so important).

@dlucy I was thinking the same thing about replacing the small signal transistors as well.

While I would not have recommended going this high before, the 730 that I'm working on now had been previously recapped with 10,000 uF reservoir caps... EACH. So, 4 x 10,000 uF.

And it's working fine.

 

[jheu02]

Reset bias to 35mV as per above and then gave a brief listen. some distortion and channel drop on R side due to dirty pots/switches as it cleaned up with moving them. So, not as extensive a listen as planned. Pulled the Tone board so I could get at the pots and volume knob with Faderlube, so decided to do the recap at the same time. Also replaced the input Rs to Dale RN60s. The Rs at he output side had already been made jumpers, so no change there.

Tone Pre

Tone Scat

Tone Post

 

[jheu02]

I was making up a parts list today, even though I don't have the amp yet. I thought about going up to 6800uF 50V caps for the mains. @jheu02 do you have the diameter and height (not so important).

@dlucy I was thinking the same thing about replacing the small signal transistors as well.

OEM main filters are 30mm x 70mm. You can maybe fit up to 35mm, but will be tight against other components. OEM are 3 leg (one dummy) and probably (though haven't measured) 22mm lead spacing or thereabouts. Haven't decided whether to go 6800 or keep it at 4700. SM lists +50%, -10%

 

[leesonic]

OEM main filters are 30mm x 70mm. You can maybe fit up to 35mm, but will be tight against other components. OEM are 3 leg (one dummy) and probably (though haven't measured) 22mm lead spacing or thereabouts. Haven't decided whether to go 6800 or keep it at 4700. SM lists +50%, -10%

I have four of these Mouser part number 647-UKW1H682MRD left over from a couple of NAD 3020B amps that I sold unrestored. They are relatively tiny at 22mm D x 50mm H, plus they are "high sound quality" according to the Nichicon data sheet. I've also used them in a Kenwood KA-7100 which has four of them, and that amp is rated at 60W per channel but tested higher. So I would think 6800uF would be find for this Harman Kardon.

These Nichicon KGs are 30mm D, but only 35mm high. 647-LKG1H682MESBAK. It's difficult with these old amps, trying to get them to look right with modern caps. Personally, it doesn't matter to me what they look like inside, but some people get all crazy about stuff like that.

Keep up the good work.

Lee.

 

[jheu02]

While I would not have recommended going this high before, the 730 that I'm working on now had been previously recapped with 10,000 uF reservoir caps... EACH. So, 4 x 10,000 uF.

And it's working fine.

Going that much higher can stress the rectifiers and the power switch from inrush current. Hopefully the 730 will CONTINUE to be ok, but there's some other mods you could do re: power switch triac and/or beefing up the rectifiers. That could then (I guess) stress the transformer....

So, I try to stay a bit closer to what's spec'd by the engineers. The 6800uF Lee is speaking of is at the upper end of the + 50%, but still within, and shouldn't require any other mods to be done safely. I was looking at the KWs (or other leaded cap) vs snap-in types I normally use just so I don't have to make pin extensions or drill new holes to accommodate. a 10mm lead spacing. Will revisit as I get my order together. I'm just using stock I have on hand currently. Two of the 220uF caps I replaced yesterday appear to me to be in the audio path, so I may order some UKLs for those (though the Panasonic FM series are really good caps, just not specifically low leakage). The Silmics I had and decided to use here vs UKL just to see if I want to go back to using some "audio" types vs low-leakage as I have been doing for audio path.

 

[jheu02]

@leesonic... I just measured lead spacing on the main filters... 12.5mm. The 63V KW has 12.5mm lead spacing. Pulled one of the caps. Due to the offset nature of the 2 primary pins from centerline, snap-ins might be difficult UNLESS you restrict the diameter to max 25mm. Will still require bending the leads, so a "standard" leaded cap will probably work best in this amp (though I was using a 35mm snap in LGU I have for another project to test fit). Will look tomorrow and see if any others I have are 30mm.

 

[dlucy]

Any reason not to drill new holes to accommodate whatever cap you choose?

 

[jheu02]

Not specifically, but the layout already has the holes close to the edges of the pads. So, wider lead spacing would make that a more viable option. One could drill a hole in a non trace area and then run a wire to the trace, but then you'll end up with stability issues unless some sort of glue is used to hold the cap in place. I'll photo it later and post.

EDIT to include pics...

Component side shows how a modern cap would need to shift towards rectifier and fuse in order to align centered pins vs OEM offset. I tried with a 35mm diameter cap last night, but no go. A 30mm might work as stated, but drilling new holes probably not so much. And with something like the UKW series having identical lead spacing on the 63V cap, I see no reason to work a bunch of modifications just to use a snap-in type (though those are usually the types I use for main filters).

 

[dlucy]

This is what I do on the 730s:

 

[jheu02]

Those look like UKWs and have standard leads vs snap-ins. Why drill holes when you could use the existing ones?

 

[dlucy]

Those look like UKWs and have standard leads vs snap-ins. Why drill holes when you could use the existing ones?

Their lead spacing is different.

 

[dlucy]

Well, I discovered that the SM instructions are WRONG! There's a 730 thread going and the SM for it says to connect to the collectors of the outputs in each channel. Voila! When I did that with this amp, I got readings close to 35mV...so adjusted the trimmers to maintain close to the 15mV as called for. Offset is still high, but at least I can test the sound before tearing things apart.

(so everything is more clear to future readers)

For the hk730, the adjustment is to be made while the leads are one the collector of each of that channel's output transistors:

(backside of the power map board with the output transistors and transistors's sockets removed)


(view from front of opened unit to the closest of the output transistors)


when the power amp board is installed and the top is off the unit, the collectors are the topmost pin of the output transistor's three pins.