advertisement


PYE PF91 / HF25

simonpyeman

pfm Member
I’m currently restoring a pair of PYE PF91’s and wondered if there is anyone out there who has specific knowledge of them? They appear to be an evolution of the Leak TL12.1 but with some interesting differences. It seems to be very difficult to find much about them on the web, however. The HF25 is a near identical circuit but with ultra linear KT66 pp instead of triode connection. I have the circuit diagrams for both and the complete manual for the HF25 but only an incomplete one for the PF91. Any information or interest gratefully received.
 

Electronics Old And New (M Caldeira) dis a very nice rebuild on a HFS20M a few years back and IIRC found some very interesting things out about the circuit. It is a great YouTube channel, I’ve learned a lot from it.

PS I don’t know the Pye stuff well at all so no idea if there are many/any similarities between models.
 
Sorry, but I can't be of much help with the technical side. However, I have owned a pair of completely rebuilt pair of HF25s for around 15 years used with Quad ESL57s and I have also listened on several occasions to the PF91 models . Not a lot to choose between them in terms of sound quality but a contemporary review reckoned the the ultra-linear amps performed better with transients on music peaks and also had marginally better bass. From memory , as might be expected , the output transformer specifications are different.

Incidentally, many years back, i had a conversation with a retired Pye employee who worked on both these amplifiers in the 1950s. Apparently, the transformers were not wound in-house but manufactured by a satellite supplier to the company. Both amps were ahead on their time in several respects in terms of performance and design features and remain competitive with many current / modern valve power amplifiers. I'm sure that you'll be delighted with your pair once you have overhauled them. I would strongly recommend that you replace all the top-mounted H.T. smoothing capacitors b.t.w. ideally by rebuilding the cans internally with new components to preserve the original appearance of the amplifier.

Good luck with the restoration.
 
Many thanks for the replies so far!
The main technical problem is that one of the OPT primaries has gone open circuit. I have found that Majestic have the winding details for the HF25 but I can’t find out anything about the PF91 OPT’s. The number of primary and secondary sections and impedance settings appear to be the same, though - so could it be reasonable to assume that the only difference between the two would be the UL taps? The plate-to-plate impedance seems to be the same for both triode and UL operation? (Mr Snooks suggested that I could install a switch system to a able the use of both modes in future if I use the HF25 spec opt’s.)
Another point of interest is the feedback systems, ie. 26dB GNF and positive feedback. This seems to be a most unusual arrangement which I don’t fully understand yet...! I’m also wondering if new opt’s will require changes to the feedback resistor bypass caps.
The more I find out about these things, the more there is to know and the more interesting it gets....
Thanks for your comments, brecklandboy, just the kind of thing I was hoping for. I’ve had these amps many years driving Lancasters with 15” Reds and have never heard anything easier to live with - different, yes, but nothing better overall. I think the real cause of the problems was that one of the mains voltage selection plugs corroded and caused intermittent connection, but because it was intermittent I failed to find it until the damage was done. I have since hard-wired the selection so it cannot happen again! When stripping down the amps, it never ceases to amaze me how they kept working so long with the drift achieved by some of the components. Having said that most of the really critical ones have been replaced by me over the years.
 
I do have a photocopy of the PF91 factory manual . Some of the pages are not ideal in terms of clarity. However, I can tell you that the original Pye part nos. for the mains and output transformers are 771197 and 771198 respectively. The weight of the PF91 power amplifier is given as 25 lbs which I believe is around 2 lb. less than the HF25. To a large degree, I suspect the extra weight of the '25 can be attributed to differences in the transformer(s) used for each amplifier.
 
Thanks for that. As a matter of interest, what is the core height of your opt’s? The PF91’s are 2.25” and external core dimensions 4.5 x 3.75”. That should give us an idea of where any weight change has come from.
 
The positive feedback, in fact positive current derived feedback, is a way to try and obtain very high (theoretically infinite with correct adjustment) damping factor. It was a bit of a craze at the time and IIRC originated by RCA.
Great in theory and with purely resistive loads but less so with typical real speakers...

The OPT is a difficult one (obviously!). For the right price I'm sure companies like Majestic or Sowters would disassemble the faulty transformer and then rewind it in exactly the same way.
 
Thanks for that. As a matter of interest, what is the core height of your opt’s? The PF91’s are 2.25” and external core dimensions 4.5 x 3.75”. That should give us an idea of where any weight change has come from.

Measurements taken from my HF25 output transformer correlate with the ones you have supplied from your PF91. although internet images would suggest that the former is bigger. If you check the Pye part number given in your HF25 manual, I suspect that they would differ.
 
Drifting slightly off topic... IMHO the Mozart power amp deserves more attention than it gets. A very interesting and clever design! I rebuilt one with all modern parts for a customer about 14 ish years ago (along with matching pre amp and tuner) and was most pleasantly surprised by its sound quality through the KEF R105.3's I was using then. Tight bass and an "exciting" but slightly "brassy" sound from recollection. The customer rang me with an unrelated enquiry just the other week and said they were still giving sterling service:)
 
Low output EL34 amplifiers can sound very 'sweet' and surprisingly detailed. I love the copper front panels featured on some of the Pye integrated models of the late 1950s / early 1960s which echoed the styling of the pre-amplifiers / control units that partnered their larger and more upmarket power amplifiers.
 
Measurements taken from my HF25 output transformer correlate with the ones you have supplied from your PF91. although internet images would suggest that the former is bigger. If you check the Pye part number given in your HF25 manual, I suspect that they would differ.

The part numbers are indeed different, but they would be since the PF91 has no 43% taps. What is very interesting is that the stack heights seem to be the same, as I agree with you that pictures make the HF25 look taller...
Thanks for taking the trouble to measure it - very helpful.

Thanks also Tony and Arkless: the Mozart is indeed an ingenious design and the linked pages illuminating. Could not be more different to the earlier amps, though, except perhaps in build standard. I’ve never heard one.
 
I’ve had a request for photos...so just to prove that an old dog can learn new computer tricks (thanks Alan!) and I didn’t even need a child to make it work for me!
Starts off like this:

U
nderneath, the main 8uF 750V filter cap had detonated liberally coating everything with wax and shredded paper. The previous owner had bodged it back together without cleaning:

After the first round of tidying up - the LCR’s have been there a while and are known ‘safe’ for the moment until the amps are working again:
 
There’s been a couple of pairs of HF-25’s up for sale recently but not in this country...if anybody knows someone who’s thinking of selling..........
 
By way of an update, I have just received the OPT back from rewind by Ed Dinning. He seems to have done a nice job and rebuilt it using the original bobbin and tag board, so cosmetically very little different from before. Also a pleasant and helpful character. Without altering the chassis or original wiring looms in any way, I have rebuilt both amps with new carbon resistors and film smoothing caps, etc. so that they are a properly matched pair. The speaker terminals are also new and input changed to RCA to make them easier to swap in/out.
Now for recommissioning....
 
Had a bit of fun with the hum pots: these amps don’t tag the heater 6.3V to earth like Leak do, but to one of the KT66 cathodes - presumably to lower the potential difference between the heater and cathodes of the splitter ECC33 which are at 63V. The pots were carbon track types which had disintegrated internally but the centre shaft pokes through the chassis and is used for adjusting the voltage centring. Visually it is quite unlike anything I could find, so I installed a 100R LIN 3W ceramic pot into the original hole in the backing plate and shortened the spindle, boring it out to be a friction fit on the new spindle. Once reassembled, the pot is now visually identical to how it was beforehand from above.


 
So, OPT back in and connected up. Powered up, all the voltage points seem believable if not precisely what was stated on the circuit diagram, but presumably a bit of drift to be expected due to valve aging. Connected to a test speaker, first impression is how much difference the new passives have made to noise: there isn’t any - so quiet I thought they weren’t working at first! Onwards....

 


advertisement


Back
Top