kel 84 gremlin!

[David Mccallum]

Hi folk's,I've got a low level whine comming from the valves on my kel84 integrated once it's been on a short time.A soft tap on the top case between the valves stops it ,it comes and goes from time to time.Other than this the amp works very well indead,can anyone help me sort this or is it normal for this amp to do this?.

[Richard]

Hi Davy

I can visualise(!!) the sound you mean and don't think it's exclusive to Kel84.

I'd try slackening, then retightening, all the pcb, choke and TX screws to see if it stop this sypathetic resonance. Also move the valves physically to different sockets to see if that stops them setting it off.

Rich

[David Mccallum]

Thank's for your reply Richard,I've tried moving the valves around to no avail.I will however try your sugestion about slackening then tightening the fastenings.Its a weird kind of problem,comes and goes seemingly of its own violation,the ECF80's seem to start a kind of whining resonance which at it's loudest pitch can be heard through the speakers during quiet moments of music. Have you come across this before in a valve amp?.I built the amp as the book tells it, apart from fitting rubber grommets in the fastening holes for the power tx, I had quite a noisey mains power transformer and this sorted it.I hope to resolve this as I,m otherwise very impressed with this little amp. Have tried a mains filter ( no good). Thanks again.

[Richard]

Davy

From your first post I was thinking along the lines of sypathetic resonance initiated from the power TX or choke. What you describe now could be acoustic feedback. I wonder if you have particularly microphonic ecf80s. It would be worth trying a different shelf/location for the amp and different valves if you have them. The supplied ones are usually fine to get up and running but are known to be easily bettered by the likes of Siemens and Mullard - not expensive and easy to get.

Have you just built it btw? There was a thread detailing a wrongly supplied value of feedback resistor a week or two ago - worth checking those. The circuit stability may also change a little as it heats up so we won't rule out oscillation caused by that or some other reason but let's try the easy things first.

Rich

[David Mccallum]

Hi Richard,im sorry i should have mentioned that i have changed both ecf80's
to mullard equivilent recently this seemed to help slightly but not cure the problem.I've been using this amp for a couple of months now and this whine seems to be it's only problem .It sometimes starts it when the amp has'nt heated up fully,but usually it's been on a while before i notice it.It started it
last night while amp was on but nothing playing,a soft tap on the right hand ecf80 with my fingernail stopped it instantly,however it builds back up in a few minutes.As to the resistor problem you mention, I was advised and sent two 22R resistors to fit in place of the originally supplied 220R in positions
R8/9 which i did. I checked the resistor values in circuit,some variation showed up but nothing too far from rated value except R38 where i had a problem getting a reading ,the meter(digital)took ages to settle on a reading of aprox 260k instead of 220k.This i surmised to be probably due to it being parallel to C3 though im no expert,your thoughts are very much appreciated,thanks again.

[JerryT]

David
You've identified your problem. Microphony in one or both ecf80s leading to a limited acoustic/electric feedback loop.

Take Richard's advice - move the amp (or the loudspeakers) temporarily - ideally put amp and speakers in different rooms. The problem should disappear. Experiment with positioning until you (and those who live with you) are happy.

If it hasn't there are spring clamps and/or spring loaded shielding cans that can be used to replicate the finger!

Finally you might try a 'valve damper' - Hi Fi News (ooooh - sharp intake of breath from loyal HFW readers) had an inconclusive article about them this month that may help.

Good luck.

Jerry

[David Mccallum]

Thank's Jerry,food for thought if such a thing exists. I thought microphony effects on a valve required sound waves hitting the glass(it resonates sometimes with nothing playing ).I am however interested in trying some kind of dampening device if available at reasonable cost.Your thoughts on this are most apreciated.

[JerryT]

David

You are right feedback requires a connection between output and input. Sound waves, vibrations through the chassis and electrical connections all provide a way to do this. Your fingernail experiment suggests (nothing ever tells) that the whistle is vibration driven. The orginating vibrations may be ultrasonic, at normal frequencies but at a very low amplitude initially or they may be transferred through the chassis of the amp. The slow on set of the audible whistle suggest to me that this is a natural vibration within the valve that has a limited efficiency feedback path.

If moving the amp solves the problem then the transfer is from the louspeakers. If it doesn't then the source is likely to be mechanical transmission through the chassis and into the valves through the pin/socket connection.

Springs, spring loaded shielding cans chnage the impedance of this interface and hence the transmission of the vibration. 'Valve dampers' try to change the impedance and energy flow in the valve casing and may also break the feedback loop.

Your flicking the valve will have disrupted the oscillation of the filament and other loosely suspended elements in the valve. Clamping/damping may repeat this.

A final - very speculative - thought is whether a well implemented DC filament supply will inhibit this effect by shunting the electrical signal associated with the vibration to ground. I'll leave this thought to the gods (James, Lord, Paul (post Damascus) are you listening? - no blasphemy intended).

Hope this helps a little .

Jerry

Hope that this helps

[David Mccallum]

Hi Jerry,i think you may be right about the source being vibration's from the power supply.I,m going to look into getting some vibration damper's for the ecf80s hopefully that will work,will let you know,thank's again for your help.

[David Mccallum]

I got my self a couple of dampening ring's (Golden dragon's),fitted them to the ecf80's and held my breath.Ok not quite a solution,they may have helped slightly though.They may have helped create a slightly better sound stage too,but this is speculative.I've ordered a russ andrews classic power cord in the hope that this may some how help but i think the mains transformer may be the source of this whistling noise.(It was noisey when first switched on and remains fairly noisey even after mounting it on rubber grommet's) i dont know if changing it will help or not so any advise on this will be welcome.I dont want to keep on shelling out money if it does'nt help,although i do like this amp in virtually all other respects.Meanwhile i will live with it although i would like to resolve this minor but sometimes annoying problem.Thanks for all the help so far folk's.