Blowing B+ fuse

[trombone]

This is a two part question that concerns output current. My amp is not a commercial one, so there's no point in giving a model. It is built on a Harmon-Kardon Citation II chassis with an extra B+ transformer. Power is about 75w

B+ voltage is about 525 unregulated DC.
Tubes are KT-90s.

Unfortunately, I do not know how to figure the current coming out of those tubes. That's question 1. How do I figure that? I have a feeling that figuring idling current is one thing, but figuring with the amp running is a whole other matter.

For question two. There is a glass 1/4 amp slow-blow fuse in each B+ line. The right side is starting to blow. (I play only orchestral music, and it's possible there is more current going into the right than into the left channel.) It does so quietly, when the system is on (as opposed to turnon). The tubes are a few years old, but test well with no shorts. This has been a problem for a long time, but it's been very intermittent. It's happening more frequently recently.

I plan to switch the output tubes on the left and right and see if the problem follows the tubes.

Beyond that, I'm not sure what else to look for. I have a basic idea as to what would blow these fuses, but I'm sure others have more experience and knowledge in this matter, so I'd appreciate any ideas you might have.

 

[richwalters]

Unfortunately this sounds like duff KT90's..:- I had suspect ones so I tapped them gently with a eraser/rubber on the end of a pencil and lo behold some started flashing inside. KT90's ex Yugo Ei did go through a very poor quality phase. Swap with KT88's is a quick cure or try EH KT90 types.
The A/B swap test is a good idea but mark the tubes beforehand.

The other problem may be caused by the amp oscillating in the RF, the citation 2 has been known to do this. Disconnecting the LS can provoke many tube amps to "take off". A dummy load connected across the amp LS terminals with a scope probe is the preferred way, with perhaps a small loudspeaker fed with a 1K resistor. By doing this you won't destroy your Rosewood speakers and you can still hear a duff connection.

I tend to be rather a heavy handed person: when I come across a suspect operating amp, I slam it on the bench to wake it up, then connect it a soft isolated mains supply then wait..:-for a response from it, if any :-

richj

 

[kevinkr]

Hi Trombone,
I think you need to take my advice on the fuse issue, long ago I indicated I was having nuisance fuse blowing issues in other amplifiers based on this design. (I designed the amplifier Trombone is asking about.)

I currently recommend 375mA for the B+ fuse on each channel. The fuse is only there to protect against the spectacular tube failures that sometimes occur in these amplifiers.

Have you replaced the original bias pots? Yet???

Interestingly the high powered version with regulated supplies has never had this particular issue.

Old EI KT90 are suspect, and new JJ are fire crackers.. From experience the KT88EH seems to be durable and good sounding in these amplifiers, and I think the reissue Tungsol 6550 would be good as well.

Richj makes a good point and given the age of those output tubes I would recommend replacing them.

Incidentally I am not in favor of doing the old tube swap from one channel to the other. Why play Russian roulette? Swap for another set from your stash instead. Those OPTs are worth way more than any output tube you have sitting around.

Also why not just email me/call me with questions about this issue rather than taking a risk with getting a bum steer here?

 

[kevinkr]

Idle current in this design should be 50 - 60mA, measured at the cathodes with the standard cathode feedback networks this should be about 1.7V - 2.0V..

Maximum current consumption in the output stage is between 275mA -300mA depending on the version of OPT at full power.

B+ is 525V..

Maximum output power is around 70Wrms per channel.

 

[richwalters]

Kevinr , somethings up here. It seems alot of volts are being lost into quies dissipation with B+ at 525V , max o/p power I would be looking at round 100W per ch with p-p config esp with KT90's. Perhaps I've missed something ?

I'm gettin' away using a 315mA fast blow fuse in the c/t with a conventional p-p 100W o/p stage. However a hefty bass transient or duff input screen will blow it.

richj

 

[kevinkr]

Kevinr , somethings up here. It seems alot of volts are being lost into quies dissipation with B+ at 525V , max o/p power I would be looking at round 100W per ch with p-p config esp with KT90's. Perhaps I've missed something ?

I'm gettin' away using a 315mA fast blow fuse in the c/t with a conventional p-p 100W o/p stage. However a hefty bass transient or duff input screen will blow it.

richj

Plate load is about 3.4K, and the transformer is a Citation II output transformer. The Citation II ran a pair of KT88 at 100mA per tube and 450 - 460V and delivered 60Wrms output. B+ is about 60V higher than the stock amplifier would use.

I didn't design the amplifier originally to use KT90 so I cannot say for sure that the output power isn't higher with these, however the power supply really is not designed to deliver more than 400mA on a continuous basis.

The differential driver stages were designed for class AB operation only and cannot supply the grid current required for class AB2 operation.

Full power both channels driven is 140Wrms total, or 70Wrms per channel. For higher powers I always double up, and have a number of 120Wrms per channel amplifiers in the field with nary a failure in 10yrs.