Solderin' time!

EdGrip

Since my mate's log cabin recording studio is just about ready, I thought I'd better stop procrastinating and sort my SFPR out... IN THEORY, and if BLS Transformers have got their maths right this time, all that remains is for me to solder in the PT...

octatonic

Jesus!

I'm calling you 'sparky' from now on.

EdGrip

Hardly; I'm a hands-on DIY person who took stuff to bits to see how it worked as a kid for fun - I have relatively little idea of the theory and have to ask people who actually know stuff - like @ICBM - boring questions quite a lot. Primary side done; tedious job on a Euro-market amp with a voltage selector...

octatonic

Looks like a fun project, Sparky.

ICBM

EdGrip said:

Primary side done; tedious job on a Euro-market amp with a voltage selector…

While you're at it, do not connect that selector up fully operational. Connect the centre wire to the 240V terminal instead so that the selector can't be accidentally (or deliberately!) set to the wrong position. Don't jumper the centre to the 240 or move the 240 to the centre either, that will cause a short if it's set wrongly. I have seen quite a few old Fenders with blown mains transformers caused by these selectors.

I sometimes don't even connect the other taps to the selector if it's something like an old Marshall where the live pins could possibly be contacted by some idiot with a screwdriver or a paper clip trying to "find out what it does" (yes, I have come across this as well…). I just insulate and secure the extra wires somewhere.

EdGrip

Too late, it's done now. ^_^ I'm happy with it, unless you mean that the switches themselves have a history of failing in a dangerous way? I've actually had the transformer wound with a 250v primary, then 240, then 230, and the others as normal - because the line voltage here of an evening is quite often in excess of 250v (I've seen 257v for a whole evening... I thought my plug-in meter might be making some kind of RMS error, but Fluke gave the same reading) so I thought I'd give the switch an actual practical purpose - only because Princetons have a very high B+ as it is. 

As I see it, those three voltages take the place of a variac for people who want to "correct" the venue voltage - if/when I sell it I will have to mark this very explicitly on the amp somewhere, or more likely just bypass the whole thing as you describe and leave it at 240v.

I can hear you facepalming from here...

ICBM

No, that does actually make sense - the Fenders from the very late 70s and 80s have a 260V tap as well, and I leave that operational because it is useful as a Variac or in places with very high supply voltage.

But it's a really good idea to disconnect all the taps below 230V (or 220 if you're ever likely to go to Europe with it) - it's actually not the 100-127V range ones which are the most risky since they will usually blow the fuse immediately with no other damage, it's the 200 and 210… those are close enough to correct that the fuse won't usually blow straight away, but the excess secondary voltages can do extensive and expensive damage throughout the amp.

Speaking of Princetons I got a mid-70s one to work on a couple of years ago where the owner's teenage son had "experimented" with the voltage selector... and had fried both transformers, all the valves and even burned out the pilot lamp. (I suspect he'd also been 'feeding' it fuses too since it's hard to see how all that could have happened at once, even with it set to 200V.)

EdGrip

All done, with the redundant voltages disabled as you suggested. I'm seriously apprehensive about firing it up, after the headaches last time. Y'know the feeling? At the moment the amp is Schroedinger's cat - turning the amp on is opening the box.

EdGrip

Would you have a recommended testing sequence after a new PT is fitted that you can't help but be suspicious of due to an error in the manufacturer's previous PT? Like, transformer voltages with nothing fitted, then voltages with with just the rectifier fitted even though they'll be v. high with no load, then with all other valves fitted?

Misterg

While you're waiting for ICBM, have a look at a 'light bulb limiter' - it takes a lot of stress out of situations like this.

Loads of info on the 'net, example here:

http://www.vintage-radio.com/projects/lamp-limiter.html

FWIW, I would check it without the rectifier first (with the blub limiter), then intall everything, (assuming the amp was OK before) - still with the limiter. YMMV

ICBM

Do you have an old 60W or 100W incandescent light bulb and a holder you can attach wires to? If so, power the amp up with the bulb in series with it. (Temporarily solder it across the fuseholder with the fuse removed, if you don't have any easier way of connecting it.)

When you power up the bulb should flash brightly and then fade out. Then you can go around and check your circuit voltages are correct. If the bulb stays lit, you have a problem but the amp won't come to any harm since the bulb will absorb most of the power. Still keep a good look out for anything that looks or smells like smoke though!

EdGrip

(Going through emails with them to check that there's nothing I've missed: "Yes, I had problems with a Fluke meter a few years back. Probably why it ended up drop-kicked out of the back of my old factory. We use a Thirlby Thandar for our voltage testing and an Avo in series for current measuring. There's extra voltage on the HT to counter losses as well. It should be 352V AC between red and orange with no load connected.")

Another thing I noticed going through the emails is that I'd sent a pic of the original transformer in situ to show how much clearance there was next to the rectifier, as I'd asked for a bit more current capacity from the new one in case I ever felt like doing any of the common Princeton mods which are said to be hard on mains transformers. He suggested that he might build a taller stack, so I sent the photo to show the area he was working in. Oddly, the new transformer is noticeably shorter than the original:

Of course, I have about as much idea about transformer design as I do about NASA engineering, so I feel a bit bad even mentioning this on a public forum. I'll ask Simon about it later.

EdGrip

You will be amused to hear I am building me a "guard lamp" like the one described above. For some reason the idea of making a whole thing on a bit of wood was more appealing than dicking around with a bulb-holder and bits of wire and soldering stuff in "just to test".

EdGrip

Just found this: Some dude's Princeton Reverb kit build... :O

ICBM

It's a very useful thing to have if you do amp work. It makes powering-up an 'unknown' much less worrying and saves wasting fuses.

I have been hoarding 100W light bulbs in case they become unobtainable in the future. (You can't buy them new now.)

EdGrip

Interestingly I found that B&Q are still selling 100w filament bulbs - but only in the R88 reflector bulb form. I can only assume they are exempt for some reason. Unfortunately they don't fit in the bulkhead lamp I bought. But I think I know where I can get a 100w ES bulb tomorrow morning... 

Would a halogen bulb have the right properties for the task?

ICBM

EdGrip said:

Would a halogen bulb have the right properties for the task?

I think so, yes. They're essentially a filament bulb with slightly higher efficiency to get around the regulations .

Misterg

A 60 W bulb works OK, too... You won't be able to play the amp with either, but it will let enough juice through to get things warmed up and minimise the damage in the event of a major problem.

(Don't like that PR photo - it's unnecessary. Music doesn't like going around corners )

EdGrip

I was thinking when I saw it, how would I do it differently? I wouldn't do those hard right angles - a gentle curve would stress the wire less and make a shorter run. Speaking of shorter runs, I would allow diagonals too...

Misterg

FWIW, I think short runs are the way to go. Also, running wires against the chassis where possible helps minimise the amount of noise pick-up & reduces the chances of instability. (Unless you're building  something like a Trainwreck Express where a certain amount of instability seems to be part of the design).

ICBM

Misterg said:

A 60 W bulb works OK, too... You won't be able to play the amp with either

You will unless the amp has a *very* high idle current draw, or you're trying to play it loud. I often test amps like that.