[quote author="toby"]i know when people have had the 600 ohm input problem with the LA2 the have added a 2nd 10k input transformer.
toby[/quote]
I have the same problem with my LA2-based compressor. I should have listened to you people and built in a 1:1 (10k) transformer right away!
Instead I have this O.E.P. with 600R : 25k .
As long as connect a synth with low output impedance (Rhodes sound from my WSA-1), everything works fine. But when I connect my (real) Wurlitzer E-Piano, the LA2 input loads the Wurlie too much (dividing down its output voltage), and even though the transformer amplifies that voltage by factor 6.45, the overall noise level is much increased as well.
When I run the Wurlie thru a cheap guitar stomp box (Ken Multi Phaser (;->)), it's ok again - this can drive the 600R input. Still not optimal at low frequencies, I guess.
So sooner or later, I will put a 10k:10k in there. Maybe I'll use an extra tube gain stage before the optoelectronic voltage divider, to get a similar
gain as the original, but at a higher input impedance.
Generally, the more I think about building "pro" gear, the more I come to the conclusion that on the outside, the small power stage driving a 600R load is quite useful (behaviour of drivers and transformer at higher levels of power etc.), but on the input side, the 600R are quite annoying (unless your whole signal chain is designed for 600R, of course).
No wonder that you find a lot of recommendations for higher impedance / lower gain input transformers for cloned circuits.
Which leaves the question how to deal with the acompanying loss of input gain. Often this may not be a problem - but then again, the original designers certainly had their reason to implement some extra gain reserve. If I want to use one of these classic limiters just for that: limiting signal peaks, it won't be much of a problem (at least I hope so). But compressors can also be a wonderful creative tool when working at their GR limits (great for solo instruments complete with echo and reverb, all run thru maximum GR), so I should really consider an extra active gain stage when I huse a 1:1 input transformer.
JH.