Wanted 5687 valve

Looking for a 5687 for my new pre-amp
Anyone got anything interesting or esoteric?

I might have Kev but I am off to Marrakesh today but I will look when I get back.
Just had a quick look and I have GE 5687 5 stars and Raytheon 5687WA.
2 off each.
I also have some equivalents.

Thanks Dave,
Will have a chat when you get back.
Enjoy your trip :grin:

Anyone else?

What is in there at the moment, Kevin?

At the moment there is a Phillips ECG (Sylvania) 5687.
Which i am reliably informed is not great.

Mine had a Raytheon 5687WA when I first got it, which sounded great but was horribly microphonic. I replaced it with a Tung Sol but I can’t remember what vintage, I’ll have to check.

The Tung Sol sounds fine, but I’m keen to hear what you think of other alternatives. @browellm changed his a while ago, but I’m not sure what he bought. Maybe Mark can advise?

I’ve been looking for 5687 or equivalents for the past few weeks.

I found an old thread on the Wam and Phil (papa lazarou) used to extol the virtues of Tung-Sol in his Kondo.

However there are loads of different versions and ages of Tung-Sol, and prices form £30 a pair to £300 a pair.


I don’t know what mine is. I must check. I think I bought a pair at the time.

Don’t know if there is any information here, that might help.
Look at the bottom of the page.

Coming at valve choice from a slightly different angle - I’ve not found that particular manufacturers are in any way consistent, even allowing for things like authenticity, age, amount-or-otherwise of use, etc. It seems to be the stuff of myth and legend more than of any usefully consistent guide to quality, tho’ Mullard and Telefunken fangirls will naturally disagree…

Instead, I tend to look for mil-spec / ruggedised versions, as - in theory (and often in practice) - they tend to be longer-lived, more consistent across batches and less microphonic unless FUBAR.

In this instance I believe that’ll be 5687W, WA and WB, plus CV-prefixed versions.

Keep an eye on datasheet specs mind, cos some claimed equivalences (especially among thiefBay vendors who are all lying, thieving POS) are bullshit.

I’m about to buy a couple from Definitive, mainly because 5687s are horribly inconsistent and as Sodders pointed about ebay is just too unreliable.

At least I can return them if there is something wrong with them.

Dunno if they have any ATM, but always worth giving Langrex a shout too.

What are you using now?

Can’t remember, either some form of Tung-sol or a Sylvania possibly.

1 Like

As if by magic, look what postie has just delivered

Shitty phone pic, sorry

By way of a bizarre corollary I’m currently working on a push-pull amp which uses four 12AU7As. Each one is configured as a mu-follower driver for one of the 6550 output valves. For some reason in each channel one of the 12AU7As is one of these

i.e. an old stock Sylvania, and the other is one of these

(this is a stock pic so not necessaarily this date code) i.e. a nearly new Electro-Harmonix, made in a very different country at a very different time.

Given that each 12AU7A is directly coupled to just one output valve and that the beneficial effects of good matching between output valve gains is very well known, I suspected that this gross mismatch might well be compromising the amp’s performance (although, to be fair, it was pretty decent). Anyway I swapped two valves over to put the pair of Sylvanias in one channel and the pair of EHs in the other. The performance of both channels got objectively worse. So I’ve swapped them back.

What do we know eh ?


1 Like

:open_mouth: . . .

In my case just barely enough that it is actually worse than nothing…

Tho’… presumably you tested them to see how each measures?

If I’d had the time then I should have done. Sadly I have to hand this amp back to its owner at Audiojumble on Sunday. All I can say is that the valves’ measured DC performances are close to one another.

One feature of the mu-follower is that it delivers a gain of pretty much mu, and of a valve’s three characteristic AC parameters (ra, gm and mu) this is the one which varies the least between different valves of the same type and also valves of different ages. So, to summarise, this circuit probably ‘cares the least’ about valve variations because they should all have a similar mu.


1 Like

I’ve got 2x Raytheon JAN-5687’s you’re welcome to borrow, think they are 1960’s ones.

Checked them, they are both 1967 ones

Nice one Chris.
Will send you a PM