KiT 6550 ECF 80 Pin 6 Voltage

[Richard]

Hi Tim,

The power supply is common to both channels so I wouldn’t expect it to affect hum differently between channels when the volume is changed. Have you checked different input selector positions (if you have them) and using different sources, and any earths?

Re. the psu, parts I’d check are;
Choke L1
R5 5k6
C5 120uF 500V

C5 is close to its voltage limit so change it out if it’s not new.
R5 should be checked cold and hot after running a while for any change in correct resistance.
Choke L1 should be checked for inductance 2.5H and resistance approx 28R cold then hot after running a while.

I had a similar choke in a different amp which shorted its turns after running for about a minute when it warmed up and a hum kicked in on both channels (it was still conducting but lost its inductance so didn’t smooth the DC). I found the problem with the cold/hot resistance check when the resistance measured much lower hot. Be very careful metering there as voltage is very high. The 10V drop across the choke can be checked running carefully with a probe clipped on each choke terminal before powering up. The choke DC resistance is best checked powered off and cold, then soon after running with it powered off again when the voltage has dropped but the choke is still warm.

[TimN]

Originally Posted by Richard

Hi Tim,

The power supply is common to both channels so I wouldn’t expect it to affect hum differently between channels when the volume is changed. Have you checked different input selector positions (if you have them) and using different sources, and any earths?

Re. the psu, parts I’d check are;
Choke L1
R5 5k6
C5 120uF 500V

C5 is close to its voltage limit so change it out if it’s not new.
R5 should be checked cold and hot after running a while for any change in correct resistance.
Choke L1 should be checked for inductance 2.5H and resistance approx 28R cold then hot after running a while.

I had a similar choke in a different amp which shorted its turns after running for about a minute when it warmed up and a hum kicked in on both channels (it was still conducting but lost its inductance so didn’t smooth the DC). I found the problem with the cold/hot resistance check when the resistance measured much lower hot. Be very careful metering there as voltage is very high. The 10V drop across the choke can be checked running carefully with a probe clipped on each choke terminal before powering up. The choke DC resistance is best checked powered off and cold, then soon after running with it powered off again when the voltage has dropped but the choke is still warm.

Thanks very much for your help, Richard.

You mention a 10V drop across the choke but there is no context. I think you might have missed a bit.

I have had suspicions about the choke for a while. How critical is the rating of it? I can't find an off the shelf replacement for it. All I can find is 5H 400mA 42R that Sowter has. Is there somewhere to look for the design criteria of the power supply and choke in particular? A quick look in Morgan Jones couldn't find anything.

Tim

[Richard]

Hi Tim,

Sorry yes, context is looking at the circuit there is 475V going into the choke and 465V coming out. The choke passes all the valves’ HT which is a few mA for the small valves and 4x 88mA for the 6550s (found at their cathodes I=V/R so 44V divided by 500R) = 352mA or thereabouts in total. Choke says 350mA rating on the circuit which agrees.

So, the choke drops 10V due to its resistance which isn’t on the circuit but V=IR so re-arranging, R=V/I (also useful are I=V/R as previous and W=IsquaredR)

We know V is 10V from the circuit and I is 352mA from working out the 6550 at the cathodes so,

R=10/0.352 = 28 Ohms

Now we can check the choke cold and hot either power off for resistance 28R or power on for voltage drop in that circuit 10V. If it is going short or already shorted internally you may find no voltage drop across it or the drop varies as it warms up.

I would go back first though to check the inputs aren’t the source of the hum. Try different sources or short the inputs and see if that stops the hum. If it is generated by the psu it will still hum.

[TimN]

Hi Richard,

L1 resistance measures 38R cold. After running for a few minutes and switching off, it seems to drop to around 10R or lower before returning to 38R within around 10-12 secs. It's difficult to be precise because it takes the DMM a second or two to catch up with the change from V to Ohms. I might get a better picture of what is going on with a scope.

The interesting thing is voltage drop measures 12V DC and 13.7V AC at 99Hz. I tried it on two different DMMs. Does that mean it isn't inducting?

R6 is well within tolerance. I changed it out when I rebuilt the power supply board. It's rated at 3W.

C5 measures around 120uF which is not bad because it's rated at 100uF 500V. It's around 2 years old.

Tim

[TimN]

Of course, you can't measure resistance with a scope - I don't think. It's been a long day!

[Richard]

38R cold may be about right. My calculation of 28R for the choke resistance was based on the original circuit diagram values and there were some changes made in practice to the 6550 cathode R values (what are yours?).

If the drop is 12v that would tally with the higher 38R and the current through it may be different again with different cathode resistors.

The 13.7VAC you measure is probably AC ripple across the choke which then goes onto the second caps for more smoothing. You could meter further along across R5 looking for the 95V drop shown on the circuit and also see if your meter will lock onto a low AC ripple content at that point.

So the choke looks ok cold but a noticable lowering of resistance when warm is what mine was doing as described earlier. It's not the actual resistance that matters (otherwise we could just add a resistor!) but, that it indicates the turns may be shorted and hence inductance and smoothing lost.

I'd contact Matthew first to see if he has a suitable replacement as they may have been similar in other amps such as Kit34.

[bob orbell]

Hi Tim/ Richard, been following this, so can I add somthing, I see from your pictures that you have replaced C1 to 4, I cant see how you have configured the capacitors, but C1-2 should be in series as are C3-4, with R1-2 and R3-4 for equilizing the voltage, the reson for the series conection is so the 5U4G see's 50uF, and not 200uF, it might be worth checking, also disconect one terimal when taking resistance readings from the choke. Sorry if you already know. Bob

[Richard]

Hi Bob, yes good comment.

My hum (starting soon after switch on) took a while to find out the cause.

After bypassing various caps with others across them, checking a few resistor values, swapping in other valves etc I focussed on the choke and noticed a change in the voltage drop across it when the hum kicked in (the amp was upside down in the living room plugged into the system and the fault was a repeatable hum which started a minute or two after switching on).

Clearly it could have been several things, valves, caps, bad resistors etc and the 10H choke was the last I looked at. When I did though this was probably more obvious in my case (Glasshouse 300BSE) as the circuit drop is -35V across the choke and the drop changed when the hum started.

John Caswell kindly posted me a suitable value choke which I stood alongside the upturned amp (it was huge and wouldn’t fit the chassis which still had the original fitted) and connected using the honourable art of solder blobbing to check our diagnosis.

By now over a week later it was very satisfying to hear silence and no hum. I played it that way for a good while as the choke was no longer available from HFC and no commercial one would fit. It took a few weeks more to have 2 wound by Amethyst (£100 minimum order!) and when one was finally fitted it was bliss – job done.

I know I did the calcs and checked the Ohms choke resistance with a meter hot and cold and concluded turns were shorting inside and that would have been with the wire blobbed on and off as you point out.

Sorry if this a bit long winded, I write it in support really to say yes these things sometimes are a pain in the butt and take a while but you will get there...

[TimN]

Hi Bob & Richard,

Yes C1-4 are all oriented correctly. I've tried taking one terminal off the choke and measured the resistance. Why should I do that? It dropped to 37R.

The 6550s are Tung-Sol.

I measured the voltage across C5. There was 360V DC. The DMM still measured around 13V AC but I couldn't pick up any Hz. Does that indicate anything?

I've tried re-routing the cabling but that hasn't made any difference.

As you say, I think it's going to be a matter of picking through it and swapping things around until I find the problem (hopefully). As you say, it is indeed a pain in the butt!

But thanks to your help, I have learned something in the process. Thanks you very much.

[bob orbell]

Well Tim, your question - I've tried taking one terminal off the choke and measured the resistance. Why should I do that? - the reson is that the choke is connected to R1 2 3 4 5 6a and 6b plus C1 2 3 4 and 5, the resistors and choke are connected in a combination of series and parallel, you could calculate this, look at the power supply circuit, the capacitors my have a small charge in them and this is what makes the meter fluctuate, plus your meter injects a small voltage into all of these components, so, you are reading the sum of all the resistors and choke and what ever the capacitors are left with, although they should settle down to infinity, but this may take some time. Bob