SABA Freudenstadt Model 18

[Bluenose]

Hi Guys.. A bit young but I have one of these which tends to exhibit just a slight bit of distortion after a while (15minutes?) I've changed the ECLL800 output valves but the daunting hybrid circuitry leading up to these is a challenge. I suspect the capacitors feeding the valves but which ones the3 circuit diagram I have is in German...oops.

Having taken it out of the case and looked... then put it all back.. The distortion seems to have gone?! Puzzled.. Any ideas gents?

[Graham G3ZVT]

Distorted on one stereo channel or both?

[ms660]

Originally Posted by Bluenose

Hi Guys.. A bit young but I have one of these which tends to exhibit just a slight bit of distortion after a while (15minutes?) I've changed the ECLL800 output valves but the daunting hybrid circuitry leading up to these is a challenge. I suspect the capacitors feeding the valves but which ones the3 circuit diagram I have is in German

The schematic is quite easy to follow once you know the system.

The switch contacts in the schematic I'm looking at are shown with FM Mono selected.

In that schematic the output pentode's grid coupling capacitors are:

RH channel....C517 and C521

LH channel....C417 and C421

All the above are 22nF (0.022uF)

Looking at the schematic you could be forgiven for thinking that the signal amplitudes at the control grids of the pentodes in the ECLL800 would be substantially different, that can be explained away by the fact that the valves triode is operating at unity gain (the triode has a very very low amplification factor):

https://nvhrbiblio.nl/schema/Saba_I18.pdf

http://www.r-type.org/pdfs/ecll800.pdf

Lawrence.

[Chris55000]

His!

Distortion on this type of circuit on one channel is either due to the ECLL800 valves being defective on either of their two pentode sections, or the 150 k triode anode resistors R431/R531 going high in value – the triode is specially made to act as a direct–coupled phase–inverter stage for the second stage pentode, and it does this automatically when the specified value of triode anode resistor is used.

Swap the ECLL800 valves over between the channels – it's quite safe to try this as the valves get their negative grid bias dropped across a resistor R610 in the h.t. – lead – if the distortion changes over with the valve, the valve producing the distortion should be replaced.

The cathode currents of both ECLL800 valves should be approximately the same both channels, if the set is of PCB design you might be able to desolder the common cathode tag 7 from the print to get a meter prod on the tag with the other prod to chassis–earth.

Another possible cause, if the coupling capacitors C417/C517 and C421/C521 are o.k. and the triode anode resistors for the ECLL800 valves are o.k., is either a defective loudspeaker unit itself, or possibly distortion being generated in the preceding stage of the amplifier, half an ECC83 valve, voltage tests should pick this up!

Distortion on f.m. only may be due to leakage in the two germanium f.m. ratio detector diodes, or a fault in the ratio detector transformer – green–spot corrosion on the wires?

Chris Williams

[Chris55000]

Hi!

It's easy to check whether the two pentode halves of the ECLL800 valves are in balance or not – disconnect the red and blue wires from the PCB to the output transformer, then connect two well matched 100 ohm 2W resistors from the PCB tags the o/p transformer leads were connected to, (this need only be done for one channel at a time, the PCB drawing will identify these connections for you) and connect the mid–point of the two 100 ohm resistors to the + (h.t output) tag of the B250 C150 contact cooled bridge rectifier and leave the set to warm up for twenty minutes or so, then measure the d.c. voltage across each 100 ohm resistor with a d.v.m.

If one voltage is considerably more than the other, swap the ECLL800 valves over, if you still get the same unbalanced voltage drop across the 100 ohm test resistors, there is a circuit fault, but if the unbalanced drops are associated with only one valve, the affected valve should be replaced!

Alternatively, you can (more simply) add the 100 ohm resistors between the red and blue leads of the o/p transformer on one channel, and the PCB tags they originally connected to, and then measure the voltage–drop across the resistors that way!

The reason I don't say "simply measure the d.c. across the transformer primary" is because the resistance of each half of the primary is usually different enough to make a direct measurement inconclusive!

Don't forget to reconnect the o/p transformer leads back to the PCB tags where they came from, when you're done the test!

Chris Williams