Restoring 70's Kenwood integrated

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hi there,

I've acquired a Kenwood KA-3500 in almost perfect condition, the chassis of which I was going to adapt to DIY use. However, I can't bring myself to do it, after finding out that all that was wrong with the amp was that the Alps input selector pushbutton switches were gummed up.

I was thinking of recapping the boards (electrolytics), but wanted to know which other components degrade with age. It also uses what I think are mylar caps (the orange colour of ceramics but the shape of greencaps) and of course, ceramics. Do either of these degrade, or are they only worth changing if you wish to upgrade their quality?

I'm assuming carbon and metal film resistors don't degrade - is this correct?

Thanks for any help.

Stuey
 
take the transistors off the heatsink, clean all of the old heatsink grease off of them, and put new heatsink grease on them. if there are mica insulators on the transistors, make sure to take them off the transistors (or the heatsink they get stuck to both) and clean and regrease them also. be careful, the mica insulators (they're thin and clear, and look like they're made of glass, but they're a little flexible) can easily get damaged, and can be a bit brittle. if you take the screws out of the transistors and they won't budge, don't try to pry them off with a srewdriver or sharp probe. you want to keep the heatsink from getting gouged, and keep the mica insulators in one piece. i don't remember the KA-3500 specifically, but if the transistors are square or rectangular, you can usually get a pair of needlenose, and grip the transistor at the mounting hole, and pull it away from the heatsink that way. if the transistors are oval (TO-3), you have to unsolder the base and emitter leads. if the transistor still won't budge, get a pair of pliers, grip the "hat" of the transistor, and twist the transistor a little on it's axis, that usually breaks TO-3 transistors loose from the heatsink grease. remember, keep the heatsink smooth. i've seen gouges in the heatsink at the very least, keep the transistor from sitting flat (a transistor can't transfer heat if it isn't flat on the heatsink), and at the worst, poke holes in the mica insulator and short the transistor's collector to ground.

if the old heat sink grease is REALLY dried out, and it seemslike it will take blasting to get the transistors off the heatsink, soak the transistors in alcohol or solvent until you can get them loose.

another method of breaking transistors free is to take a flatblade screwdriver, place the tip on the edge of the transistor towards the center of the transistor, with the screwdriver shaft parallel to the heatsink surface)and give it a quick rap with the handle of another screwdriver. heatsink grease gets dried out and becomes almost as hard as cement. any regular solvent (alcohol works good) will soften up the residue, and can also be used to clean the heatsink, mica, and transistors.
 
Thanks Unclejed, that's great info. I'll do that.

The transistors are the flat type, 2SB618 and 2SC458 (or similar) style. I didn't know whether it was necessary to renew the grease to maintain heat transfer. Good to have it confirmed.

Does anyone have any ideas about the caps other than the electrolytics (mylar/ceramic)? I assume they'll remain stable, but I'm not sure.

Thanks again.

Stuey
 
as long as there's no thermal stress damage (splits, cracking, etc...) they should be fine. getting back to the grease, also check any devices mounted on small heat sinks.

reflow all of the soldering if you can, and any wire wrapped connections, i usually solder them as well (kenwood, poineer, sansui and others used a lot of wire wrapped connections between boards). if you have some liquid flux (liquid rosin flux), use a drop of it on any connections where the component lead won't take solder. you will probably find that the collector leads on a lot of transistors have "rings" in their solder connections. this is due to thermal stresses of the collector lead expanding and contracting, and it eventually weakens the solder connection.
 
Without the flux, you'll likely never get solder to stick to those wire-wrapped connections.

The KA-3500 has a very nice sounding amplifier section, but is held back by a rather mediocre preamp. I have one, and added an output relay and DC protection, as well as preamp-out/main-in jumpers and jacks. I use the amp section only, to power a pair of Heil AMT's in a biamp setup.
 
Ahhh, Echowars! I've just registered over at AK to ask you some questions, but haven't been cleared by the moderators yet.

Yes, I'd guessed as much about the flux...thanks.

I saw in the threads over at AK that you've modded yours quite a bit. Have you ever explained how to add the preamp-out/main-in jumpers and jacks anywhere on the web? I have now received the service manual, so I have the schematics etc.

Also, have you renewed the capacitors on yours?

What's the phono stage like? I assume replacing the caps on this would be quite critical, but I was also thinking of doing the resistors on the phono stage with metal films.

Incidentally, I intend to carefully dismantle the input selector pushbutton switches and clean them out, as they cause some crackling; have you experienced this? I saw another thread on a forum where someone had the same issue.

Thanks; apologies for all the questions.

Stuey
 
The pre-out/main-in is a lot of work. Although separating the preamp from the amp results in easily being able to drive the amp from any source, the preamp output is left with zilch for drive capability, and requires an opamp buffer. And since there's no good voltage for the opamp, you have to build a zener regulator (two of them, one for '+' voltage, one for '-') onto the board for the opamp. I really wanted to use the 3500 amp section to drive the Heil's, else I would not have insisted on making so many changes.

Yes, it has been fully recapped (except the large power supply caps, which I found no reasonable substitute for).

The 'break' between the preamp and the amp is just as the schematic shows...easy since Kenwood put the preamp on one board and the amp on the other.

Here's what the output of the preamp looks like (not that it's any startling revelation):
 

Attachments

  • 3500_preamp.pdf
    81.7 KB · Views: 185
i gather from the pdf that the original preamp output impedance was 15k. that explains why there's very little "drive". an op amp usually has a much lower output impedance (as low as 50 to 100 ohms for most of them). it's better to use a 7812 and 7912 for the op amp supply.... the power is much cleaner that way.....
 
Stuey said:
Why is the pre-amp so bad? Any ideas?

Cheers

Stu
It does an acceptable job, but it's just a bit too simplistic. Cap-coupled, running off a single supply, with NFB tone controls that cannot be bypassed. Gets the job done, but ain't what you'd call 'state-of-the-art'. If they'd used a simple opamp to drive the preamp it would likely have been an improvement (and the next step up in the series for that year, the KA-5500, does just that...using an ancient TA7136P opamp).

New caps help a lot. Bump that input cap on the preamp section up to 2.2µf and give it and the rest of the preamp signal path electrolytics a film bypass (a little .47µf metal polyester or some such) and it ought to perk that thing right up.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.