Rogers Cadet III - excessive hum?

[Martin Cox]

I'm currently working on a Rogers Cadet III. It's one of the later models with with the preamplifier and power amplifier on the same chassis. It's not an amp I've previously encountered.

The original fault was no output from either the Radio or Phono inputs but the Tape input was fine. This was due to the 220k anode resistors on the first ECC807 being open circuit. Replacing these brought everything back to life.

However there is more 100Hz hum than I feel there ought to be. I'm aware the HT supply is a voltage doubler circuit and my understanding is this is more prone to 100Hz hum than a more conventional bi-phase or bridge rectifier supply. So my question is how much hum is normal?

Voltage readings taken with a 20,000 Ohm per Volt meter around the power amplifier largely agree with those on the published schematic.

I currently have an old pair of LS3/5As connected to the 15 ohm outputs for test purposes and I'm seeing a 100 Hz charge/discharge waveform with an amplitude of about 150 mV peak to peak across the loudspeakers which is clearly audible. Although my 'scope is supposedly good up to 400V on its input, I'm reluctant to stick it straight across the HT!

So far I've replaced the 2.2 Ohm resistor between the transformer and the diodes since it was measuring about 10 Ohms and replaced a rather distressed-looking 470 Ohm resistor. I've tried substituting new 100uF 400V capacitors for the existing reservoir capacitors. None of this has made any significant difference. I don't think there is excessive HT current being drawn.

I shoould welcome your thoughts.

Martin

[GrimJosef]

150mV p-p is north of 50mV RMS and that sounds like way too much, even into 15ohms.

Is the hum affected by the volume control ? If not then you needn't worry about the stages before that.

HT rail ripple should be cancelled substantially by the push-pull output stage. Do you have any worries about this ? Do the valves appear to be conducting more-or-less equally ? The quiescent current will be controlled significantly by the output valves' screen grid voltages and these have an additional HT smoothing stage after the HT reservoir. Does this smoothing stage seem to be working ?

Could there be some issue with the negative feedback ? If that's failed somehow then the hum and noise will be raised.

I take it this is a stereo amp. Are both channels humming equally ?

Cheers,

GJ

[Martin Cox]

Thanks. The hum is independant of the volume control.

The valves look as though they've had a fairly hard life. Unfortunately with a shared cathode resistor there's no easy way that I know of measuring the current in each individual valve. They do seem to be getting equally hot for what that's worth. The valves are certainly not above suspicion. The screen grid voltage appears to be correct and the voltage across the cathode resistors is around 9 volts for each channel.

I do have a working Avo Mark 4 Valve Characteristic Meter but the the ECL86 was a little too modern to be included in the test data for it. It's also in the loft and bloomin' heavy!

I'll follow up on your suggestions tomorrow and let you know how I get on.

Martin

[GrimJosef]

The common cathode resistor does make it hard to check that the valves are conducting equally. You might get some idea by measuring the DC voltages dropped across the two halves of the output transformer primary, perhaps accounting for their different DC resistances. Otherwise changing valves can sometimes give a bit more information.

But if both channels are humming equally then that rather points suspicion away from the output stages and towards the power supply. In that case you might want to look at grounding issues. If I remember rightly I once had a corroded ground connection in a Rogers HT psu and the 100Hz capacitor-charging current spikes which flowed through that caused a deal of trouble.

Cheers,

GJ

[Simon Gittins]

ECL86 is in the 23rd edition manual:
ECL86 517 238 148 0000 00 415 237 146 6.3 1.9 250 1.2 1.6 B9A TP
ECL86 517 238 148 0000 00 415 237 146 6.3 7 250 250 36 10 B9A TP
https://frank.pocnet.net/instruments/AVO/MF/AVO_VDM_23RD_Valve_Data-Manual-OCR-20150222.pdf
Not that it helps if it's in the loft!

[Valvepower]

I think GJ is on the money here as a while back I sorted Cadet III and the smoothing/reservoir capacitors were duff and the filter/decoupling capacitors were suspect. I replaced the whole darn lot.

Being a voltage doubler power supply circuit the smoothing/reservoir capacitors have hard life in this amplifier.

I’ve also had Cadet III with an open circuit output transformer primary winding, thankfully it was just a broken winding/coil wire going to the solder tag - it was right on the tag end.

Terry

[Martin Cox]

So here's the latest.

GJ was absolutely spot-on with the additional smoothing for the screen grids. C9,16uF, 350V was open circuit and C10 also 16uF 350V was distinctly sick too.

I've replaced both with 22uF 400V capacitors and the hum has vanished; I can't hear it which isn't necessarily the same thing but it's certainly now too low to measure with the kit I have.

Out of interest, the voltage drops across each section of the output transformer primary were very close so I don't think the valves are too far out of balance. I have advised the owner that if he comes across any decent ECL86s at a sensible price it woud be worth snapping them up if he intends to keep the amp. I was able to test the ECC807s and there's no cause for concern there at the moment.

I shall add the information on the ECL86 to my valve tester's handbook.

Thanks to all for their helpful suggestions.

Martin

[julie_m]

Ripple on G2 can certainly manifest at the anode -- I found that out for myself. It might even have been GJ himself who suggested to look there.

This reminds me, I have a Rogers HG88 that I haven't used for awhile and must get a look at sometime .....

[yestertech]

These are 'cute' little amps, apart from a lack of deepest bass due to the rather small output transformers. Nearly all the ones I've looked at had failed electrolytics and space for anything but upright mounting replacements is very limited !
The valve phono stage is impressively quiet, due mainly to the use of the ECC807 preamp stages. Luckily these rarely fail, as they're hard to find.
Sadly, the same can't be said of the ECL86s....


Andy

[Valvepower]

Hello,

Attached a few photos of the replacement capacitors I took whilst doing the work on the Cadet III which I have here.

I have stash of small can 100uF/450V 30mm Dia I use for jobs like this. Given the voltages I figure one could find smaller 22uF capacitor, however, I had some of the Rifa 22uF’s in the workshop and it made sense to use those.

I concur with Andy and I was also impressed by the phono stage. I feel the phono input overload is possibly compromised by taking the feedback from the low value cathode resistor; however in practice this won't cause any problems. From memory this is quite similar to a design in the late 60s Brimar Data book.

Going at a tangent, I did tweak a standalone Cadet II power amplifier to use PCL86 using a voltage doubler on the LT. This power amplifier was being used with simple IC based preamplifier so no LT was being taken from the power amplifier, the extra LT current usually taken by the preamplifier LT, was pressed into service to add a safety margin for the increased current drain due to the action of voltage doubler, even though the current drain of the PCL86 is half to that of the ECL86. This couldn’t be done to the Cadet III though.

Terry.

[Martin Cox]

That's a nice neat job and tidier than mine! In the end I left the existing components in place albeit disconnected and used wire leaded components added under the chassis. It does depend somewhat on what one has to hand or can get hold of in a reasonable time.

I have ensured that anything I've done can easily be undone and redone in the interests of greater 'authenticity' if desired. However the owner was simply happy to have it back working properly. He's now given me a Telefunken Concertino 2194 to look at but that's potentially material for another thread.

Martin

[kalee20]

Originally Posted by Valvepower

I had some of the Rifa 22uF’s in the workshop and it made sense to use those.

Your efforts to do a neat job fitting these should be amply rewarded - these are excellent capacitors and should last a very long time.

[Valvepower]

Hi Martin and kalee20,

Thanks for that… its appreciated

Its tough decision whether to keep original with the capability to go back – so to speak, or rework as in the Case of my Cadet.

Whilst I’m here, and talking about reworking, I did tweak the output transformer secondary (see pictures) as I’m using low impedance speakers (6 Ohm). I put the two secondary windings in parallel and to keep the same feedback conditions I reduced the feedback resistor to 3K3 and increased the capacitor to 780pF (100//680pF).

I measured a marginal increase in output, but not loads, although, I really should have measured the leakage inductance; however, off the top of my head I feel this should have been reduced very slightly.

Regards
Terry