Recommended valve retailers.

tekbow

aforementioned email from Bill

"Simon,

that amp does have the XL mod. It is easy to remove. I have attached a picture. There is a place where in place of a 39K resistor there are a 15K, a 1Meg and a .0047uf cap connected together above the circuit board. Remove these and restore the original single 39K resistor.

We originally used Philips 6L6GC in our amps We switched to the Sovtek 5881 when the Philips were discontinued. They were the only decent thing available. We still use them because they are very reliable. They produce slightly less power and have a softer edge than the 6L6GC. You can use any 6L6 or KT66. KT66 may give you a little more power and headroom and a smother breakup at full power. At lower power levels the power tubes have less effect on the tone.

Bill Sundt"

@grappagreen

Yeah, i know, right?

ICBM

tekbow said:

I'm sceptical about how much difference the type of output Tubes make to the tone at lower volumes.

Yes, it made almost no difference at low volume, and I almost wondered why I'd bothered until I took it to rehearsal and it was astonishingly different - completely the opposite of the "you'll never hear it when the bass and drums kick in" idea, it was far *more* noticeable in the mix when it had something to cut through.

tekbow said:

How old wete the tubes being replaced for example?

A couple of months. I had the amp from new, and the first set of valves went rattly so I had them replaced by the shop - I thought why do it yourself when you can have it done for free? - then the replacement set did as well, so I thought why have them replaced with more crap even for free! In those days NOS power valves weren't as difficult to get or expensive as they are now, I think it cost me £25 each for a pair of JAN Philips/Sylvania 7581As - actually the same valve as the late Sylvania 6L6GC with a military part number. Those are really expensive now if you can even find them at all.

tekbow

Chinese valves mate... actually were they? The rattly ones?

nevermind, i've got scorn to deliver on current chinese tubes and i'm gona deliver regardless. Thats why you have guys like TAD selecting above and beyond factory standards and why they end up being quite pricey. I wonder how many tubes they have to go thru to get an RT001 for instance? or is it a 010? Anyways, grrrr modern tubes.

Course we'd all be ok if SED =C= were still in business. They made a great 6L6GC and EL34 by all accounts.

grappagreen

I’ve got winged C’s in my EL34 and 6L6 amps.... shame they’re not made any more..

tekbow

@grappagreen

They can still be had, and aren't uncommon yet, might be worth buying a stock of them though.. must get some for my amp to try.

need to get a decent bias meter first. don't fancy poking around with a multimeter.

There's a UK based supplier called valvetubeguitaramps.com who seem to be releasing their own soon. Shows plate voltage, bias current and calculated power with tight tolerence resistors for low reading error. Will apparently be made in the UK.

Haven't bought anything from them yet, could be another potential retailer, they seem to get NOS stuff in and don't go nuts about price. Thy had some Brimar ECC83's up for sale a while back for £40 quid a pop which was decent.

ecc83

Tekbow. "Poking about with a multimeter is how it has been done for decades!

If you already have a decent digital meter, buy two more cheap ones. These can then be crocked onto screen and anode pins while amp is cold and disconnected. Best meter goes on the cathode current sense resistor. Switch on.

You now have pretty much all the information you need. Assuming a 1R sense resistor mV=mA. Cathode current of course but if you know the screen feed R value you can calculate Isc and get the anode current. Not really necessary though, Ik is what most of us look at and if you calculate Pa on that basis at least you err on the safe side.

Precision resistors? Ok, so swap the cathode R for a  1% tolerance jobby, cost about a quid. Again, not a lot of point. The first time you blow the bits off the amp the current will increase and settle to a new value. If you fit brand new bottles after an indeterminate time the grid current will reduce and shift the bias point a bit. Then, how stable is your mains voltage?

F me T! We did this with 4% @FSD accurate Avo 8s or worse!

Dave.

tekbow

@ECC83

These are very fair points, and my counterpoint would be just because something's been done a certain way for years doesn't mean it remains the best way of doing it. And if we can have a specialist tool that's not too expensive, eliminates multiple connections and meters, then I'd rather use that.

The fewer steps in the process, the less there is to go wrong.

Power supply is pretty stable here, new build, asked the sparky last time he was in fitting smoke alarms, and yes variances in voltage will shift the bias point on what would be a continuous basis (actually thought about buying a UPS or power conditioner like we had at work) but we do the best with the reading we have at the time.

It's not that the process worries me, or I'm scared of poking around, it just isn't as necessary as it used to be.

Wish I'd have nabbed a fluke meter from work before I got laid off. Too honest I guess

ecc83

I am the first to concede we should be as accurate as possible in all our doings but there are limits!

If you have ever hung a meter on a current sense R and left the amp running, no signal input, you will see the value drift slowly over time. Usually (with 'our' amps) the bias current dropped as things warmed up but in other designs the relationship between HT variation (with mains) and its relation to bias volts could be such that the drift is the other way.

We is talking bloody poor, bloody hot DC coupled amplifiers here! Then, the reading will sometimes "blip" no doubt a spike on the mains, fridge kicked in or out.

The above is no signal conditions. Now run the amp to clipping for one minute. Ik will be very different and take quite some time to return to close to the original setting. Note "close to" Won't go back exactly.

In the end it really does not matter IMHO. There are those that say the valves should be set to 80% of rated Pa. Others reckon 50% is hot enough. "We" biased the A100 (EL34) to just under 50% at 25mA cathode current per valve. I NEVER had anyone say it was too cold or too hot! (did not STAY at 25mA all the time of course!)

IMHO a basic DMM (1% accurate?) and a 10% sense resistor is easily good enough for guitar amplifiers.

If that tester checks one valve at a time it must introduce SOME inaccuracy since for an amp with a common cathode R the valve under test now has two resistors in its cathode circuit.

I HAVE a Fluke! Mod 83. Gift from a now departed friend and I miss him.

Dave.

tekbow

Not actually sure how this one will work, they're in preproduction testing. From what they've said it'll be something similar to the compubias or Aiken Wombat (of which I keep an eye on ebay), so I'd like to think they're doing multiple (perhaps up to 4 tubes) simultaneously.

Pretty sure my 5881s were biased at 65%, the amp doesn't sound bad, so I'm assuming that'll do.

richardhomer

Just on the subject of Hotrox - and in the interests of balance - I’ve used them several times and found them to be very good.

grappagreen

ecc83 said:

Tekbow. "Poking about with a multimeter is how it has been done for decades!

If you already have a decent digital meter, buy two more cheap ones. These can then be crocked onto screen and anode pins while amp is cold and disconnected. Best meter goes on the cathode current sense resistor. Switch on.

You now have pretty much all the information you need. Assuming a 1R sense resistor mV=mA. Cathode current of course but if you know the screen feed R value you can calculate Isc and get the anode current. Not really necessary though, Ik is what most of us look at and if you calculate Pa on that basis at least you err on the safe side.

Precision resistors? Ok, so swap the cathode R for a  1% tolerance jobby, cost about a quid. Again, not a lot of point. The first time you blow the bits off the amp the current will increase and settle to a new value. If you fit brand new bottles after an indeterminate time the grid current will reduce and shift the bias point a bit. Then, how stable is your mains voltage?

F me T! We did this with 4% @FSD accurate Avo 8s or worse!

Dave.

I have to say that whilst I've built an amp and did it safely I purchased a BIAS meter on the basis that if I could avoid exposing myself to something that could kill me it was worth a few quid to do so!

When I built my 5E3 I had every confidence that I could work safely and understood how to mitigate/manage the risks.. I just don't do enough electronics work (or understand it well enough) to keep those risks in check after the passage of time.

Si

thermionic

tekbow said:

Wish I'd have nabbed a fluke meter from work before I got laid off. Too honest I guess

I often regret doing exactly the same! It would be of no use to the bastards after they'd made all the engineers redundant anyway.

I've now got a grey import (Chinese market only) Fluke 15B+ from ebay. I think it was just under £70 and it feels like a quality meter. I might try and get an Aneng for about £15 as well, they seem highly-regarded.

kt66

The best place to buy valves is an audio jumble fair like Tonbridge...........sshhhh keep it quiet. 

tekbow

@thermionic

when i get back in, I'm having one. i'll blame it on some anonymous site crew when they find it missing in the work container.

57Deluxe

I've had good deals off Thomann  in the past...