• 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

Request for help on Kenwood KA 3500: no bias current

javajunkie

javajunkie

New Member
Hello. Admittedly first time poster here asking for help, but have been cruising the site for some time for entertainment and the wealth of info.

I currently have the amp sitting on a 40W bulb current limiter for testing.

The amp mentioned above developed two shorted outputs - Q21 and Q23. I don't think the ones in place were original...they are the D587A and B617A sets, both channels. They apparently are slightly lower output and lower voltage handling. The amp had been operating correctly for some time prior to s**tting itself. I obtained and fitted replacements for Q21,23 which I tested prior to installation, and replaced various shorted transistors with same or subs as per a good thread I found on this site. All transistors were tested using my Peak DCA75. I have lifted and tested all resistors and electros on that half of the board. The caps have all been changed recently and are pretty new.

The problem is that I get a voltage reading of virtually zero when checking bias on both channels. By that I mean about .3 of a milliamp. This is by using the standard method outlined in the manual. I am well familiar with this meter, and have used it to adjust the bias on this amp before, as well as checking the bias on my tube guitar amps. I have checked and re checked Qs 7, 8, 9, 11, 13, 15, 17, 19, 26, 27, and electrolytics.

Transistors I replaced according to this info from another thread which had a happy ending:

http://audiokarma.org/forums/index.php?threads/ka-3500-transistor-subs.728835/

qe1-4 2sa620wn > ksa992fbu
qe5,6 2sc1885 > ksc2383yta
qe7,8 2sa620wb > ksa992fbu
qe9,10 2sa733 > ksa733cyta
qe11-14 2sc945 > ksc945cyta
qe15,16 2sa733 > ksa733cyta
qe17,18 2sc1735 > ksc2690ays
qe19,20 2sa850 > ksa1220ays
qe25, 27 2sa673 > ksa1220ays
qe26 2sc1222 > ksc1845



Some test results for your perusal:

E C B
Q17 .43 38 .97
Q18 .42 38 .97
Q19 - .6 -38 -1.2

B C E
Q20 -1.2 -38 - .6
Q21 .43 38 - .1
Q23 - .63 -38 - .1

Most of my readings seem to be there or abouts. My reading at the nodes of R17,19 and 18,20 is only 24.2vdc where 26 is specified.

I have checked the .47ohm resistors for continuity.

Any help greatly appreciated, I'm fine with tubes but still coming up to speed with solid state. I can provide plenty more readings/measurements as req'd.

Cheers!

P.S. I think there may be a discrepancy with the pin arrangements of Q17,18 but will wait to hear anything before getting into it.
 
Last edited:
Register to hide this ad
40W DBT too small. Need a 100W, as long as you don't see any signs of high current draw.

I'd have NOT recommended a wholesale transistor swap. Way too many chances to screw something up. And how are you measuring the idle current? You say you are getting 0.3mA, and you should NOT be measuring current at all. In electronics work, we measure voltage across a known resistance whenever we can and calculate the current. It's 100x safer. On the 3500, that means connecting a voltmeter (set to read in millivolts) between the emitter of the NPN power transistor and the emitter of the PNP power transistor.
 
Hi and thank you for your reply.

To clarify, I am using a dc voltmeter in the 200mV range in order to obtain my 40mA, as per the manual instructions. Using ohms law I guess the current reading is calculated by reading volts across the two 0.47ohm resistors (essentially 1 ohm)?

I have just been doing some quick research. As I am not up with the lingo yet, can you please elaborate on 40W DBT?
 
As EchoWars says, 40 watt bulb is too small. You're probably running at much reduced line voltage, which often affects bias. Once you think the amp is working properly, set the bias without the dim bulb tester (DBT) at nominal line voltage. And don't short anything out! I typically plug them into a Variac with a meter and set exactly 120 VAC, but that's probably overkill.
 
Ok after a couple of train trips I have procured some 100 W bulbs. Unfortunately I still have a zero reading at the bias points. The bulb is off, apart from the usual brief inrush flare on start-up. I have checked and re-checked my settings and technique.

One thing: I am not confident of the orientation of Q17,18 and I'll tell you why. The originals were C1735, 'b e c' as tested by me (one was still readable albeit having hfe measurement of only 5). The manual clearly states that the pin orientation of a C1735 should be 'e c b'. When testing and replacing the transistors, I made detailed notes of each instance, incl diagrams. My notes clearly read that Q17,18 were installed with the collectors and emitters transposed - in contrast to every other trans on the board.
My question is, could mixing up e and c of either or both of these transistors result in my symptoms? What, if any, damage can result from incorrect orientation?

At the moment, I have the C2690s in those positions oriented as specified in the schematic. It even correlates with the little 'e, c, and b' indicated on the trace side.

I'm really scratching my bald head and paranoid about damaging these outputs.

Here is a link to a thread where the dude had his Q17 backwards, 'pparently:

http://audiokarma.org/forums/index.php?threads/kenwood-ka-3500-bias-wont-adjust.731245/
 
The original C1735 should be (looking at the flat), L to R, BCE. If all that checks out OK, look carefully at Qe26 and Qe27...that they are in the right place, and that their orientation is good. This is the muting circuit, and it will do pretty much as you describe if things are not right.
 
A transistor reading "5" in your Hfe tester is damaged, and probably fooling the meter.

Test the new transistors, and place them according to schematic (check the schematic makes sense and is not wrong)
 
Okey dokey thanks. I'm confident all transistors are oriented correctly now.

So, after various family commitments, M&M consumption and hair-pulling I haven't had any breakthroughs with this thing other than not blowing it up. Still no mV reading at the bias points. I dissected the 'muting circuit' and tested the values - even subbed the transistors - to no avail. I have taken some measurements and attached the photo in the hope that some eagle-eye spots something.

It seems there is no current flowing through the emitter resistors - the voltage potential is the same at each end of the test points. It's -95mV. So it seems to me that there's either an interruption in the circuit, or for some reason polarity is screwed up somewhere?

Test Points

Test Points

 
The output transistors must be biased at the base by the driver transistors, which should be conducting biased at their bases too.... and, supply voltage must be present at the collector so there is some electricity to let flow. I hope this can help you to trace back the fail.

Can you measure if the drivers are conducting? You should measure voltage across driver's emitter resistor. (Those 330 ohm) Voltage at one driver's base is a bit low. Perhaps that one is not conducting.

Expected voltage at the output's bases to make them to conduct is around 0.6 ~0.7 V
 
Last edited:
Ok,

Swallowing my pride and posting this so some may learn/be entertained.

Well I did say I was a noob at this solid state gear...

Coming from a background of tubes I am always in the habit of setting the bias trimmer(s) to the lowest setting when installing new tubes, not knowing how much they may pull, and gradually bringing things up. At this lowest setting, some idle current is (from my experience) always still present in the cathode. Moving on, I had my bias trimmers in the KA 3500 at 9 o'clock with zero voltage, assuming there was a fault. Of course eventually in desperation and running out of ideas I decide to start twiddling the trimmers and my eyes nearly pop out of my head when I see the voltage rise. I probably had the frickin' thing fixed a week ago :rolleyes:

And all this after poring over the "always check the basics" thread.

Thanks to those who posted especially elnado who put some of the transistor testing in to basic concepts for me.

P.S. Be wary of the transistor pinout info in the pictorial descriptions in the service manual - I'm not saying it's all wrong but at least one of them is.
 
Glad you solved it. Transistors need a minimum voltage at the base to start conducting. That is the purpose of the bias adjustment, to keep the transistor always beyond that threshold so it's always conducting and never turns "off".

Regarding the pinout, I've seen the same transistor number with different pinout. I always check them before soldering in place.
 
You must log in or register to reply here.
Back
Top Bottom