Racal RA-17: 6F33, S6F33 and 6AS6

[Skywave]

All the RA-17s that I have ever worked on have used a S6F33. But apparently, the 6F33 can be met occasionally instead: the electrical specs. look very similar. Are they any advantages in choosing one over the other ? (I doubt it - but just thought I'd ask ).

But, more importantly, this Q. I am aware that in the North American versions the S6F33 / 6F33 is replaced by a 6AS6. Recently, whilst overhauling a RA-17, I replaced the 6F33 with a 6AS6 (swapping over the two grid connections in the valve socket wiring, pins 6 and 7), but the resultant drive to the 37.5 MHz BPF was substantially down. (Yes, I did do a re-tune). So, if such a substitution is made, should there be accompanying changes in various passive components to compensate? (I have a good reason to ask: I do not have any 6F33 in stock, but do have a few 6AS6 - and one day . . . ).

Al. / Sept. 30, '14 //

[amornummi]

In my experience , all RA17 s are a bit marginal in the 37.5 mixer injection.You need a few volts at TP3 , V9 grid.
Check or just replace the comb generator V2 100k g2 feed.
This 37.5 Mc/s is noisy , I can see sideband noise on HP141 ! . Due to partition noise in 6f33 then loads of gain.
Had thought of trying a "Pullen Mixer" with a 6j6 to see if any better.

[Skywave]

Originally Posted by amornummi

In my experience , all RA17 s are a bit marginal in the 37.5 mixer injection. You need a few volts at TP3, V9 grid.

Which has been my experience too.

Originally Posted by amornummi

Check or just replace the comb generator V2 100k g2 feed.

Which I always do as a matter of routine.

Thanks for your comments.
Al.

[G8HQP Dave]

I believe the 6F33 is a modified EF91, and the 6AS6 is a modified 6AK5/EF95. If so, that might suggest that the former has greater gain but the latter maintains its performance to higher frequencies? I seem to recall the the EF95 has particularly high input impedance so adds less damping to any grid tuned circuits.

[QWENIX]

@ skywave

I bought a lot of S6F33 and 6F33 brand ITT's or STC's. I guess all used .... If you want one let me know. But I have to test them before ... one goes away..... Difficult to find are the test parameters data for AVO CT160 and TV7D/U.

best regards Qwenix.

[Skywave]

Qwenix: you have a P.M.

Al. / Sept. 30, '14 //

[G8BBZ]

The Racal Modification Instruction (No.M.17/17 of Jan 1978) which covered the replacement of the 6F33 by the 6AS6 requires the replacement of two resistors.
R29 (1k0) to be replaced with 4k7.
R22 (180R) to be replaced with 470R.
This is in addition to the wiring change to transpose the leads on pins 6 and 7.
There is no requirement for any retuning or adjustment.

I understand that the S6F33 is the ruggedised version of the 6F33 (equivalent to a Mullard SQ) and has the same electrical characteristics - albeit probably more tightly controlled.

Both the 6F33 and 6AS6 are dual control valves i.e. though they are represented as conventional pentodes on the circuit diagram,the "suppressor" grid is in effect a second control grid with a similar ability to control the flow of anode current. Hence the use of the valve as a mixer.
More info at:
www.r-type.org
It seems unlikely that either valve would have been derived from a standard type, given the fundamentally different mode of operation.
cheers
Peter G8BBZ

[G8HQP Dave]

Originally Posted by G8BBZ

It seems unlikely that either valve would have been derived from a standard type, given the fundamentally different mode of operation.

A closer grid pitch for g3 is needed. If you compare a 6AS6 with a 6AK5 they look remarkably similar apart from g3. Given the limited market for dual control valves basing one on an existing RF pentode seems a sensible way to reduce development and tooling costs.

[Skywave]

Peter (G8BBZ) - your post #7 answers my Qs. perfectly: thank you.
I have made a note of those remarks and added them to my RA-17 documentation for future reference. I'm sure that there are others here whom will find them just as useful.

Al./ Oct. 6 //

[QWENIX]

I first checked the Babani encyclopedia (third edition) and was able to reconstruct the data settings for the s6f33. Later on I discovered the settings in the 20th version of the avo valve data manual. So now I am able to test these valves and their (cv) equivalents on the CT160. Next thing I have to do is to convert these settings to my other tester TV7D/U.

@ topicstarter did you manage to buy a good s6f33 and tell me if you are satisfied with its performance ?

[Skywave]

QWENIX: you have asked me that Q. in a P.M., to which I have replied.

Al. / Skywave //