Ultralinear Switch

[Hugo]

Hi guys, I'm playing around with a home build based on a 50W Marshall circuit. I'm using an output transformer with UL taps, and I'm trying to figure out how to switch between standard and UL operation. I understand how it works, the problem is hardware. I can't find a switch that can handle 500V DC.

What is the best way to implement this safely?

I tried both UL and standard wiring without a switch and found something to like with either operation. I know many people think UL doesn't sound good for guitar, but I actually found it useable.

Thanks!

[Jazz P Bass]

A relay may work.
Or a reed switch: http://www.pickeringrelay.com/pdfs/104.pdf

[g1]

What kind of DC rating do you think they use for standby switches in amps that have B+ over 500VDC ?
Maybe an engineering type can explain the discrepancy.
Here's an example of the type of switch I'm used to seeing in standby circuits, yet they are calling them "for low voltage DC rated applications" : http://www.carlingtech.com/toggle-switches-f-series

[Leo_Gnardo]

What kind of DC rating do you think they use for standby switches in amps that have B+ over 500VDC ?
Maybe an engineering type can explain the discrepancy.
Here's an example of the type of switch I'm used to seeing in standby circuits, yet they are calling them "for low voltage DC rated applications" : F-Series | carlingtech.com

Good question. I've seen AC ratings typically 250VAC. And the old standard Carling SPST like Fender's been installing since the late 50's - I use them and no new ones have failed (yet). Heck it's even rare the old ones fail. But when I install a normal/triode DPDT switch, I pick the heaviest duty Carling toggles I can find. I don't think they're the F series though. I do like seeing 1000VAC rating on the F switches. I'll have to look into who carries 'em.

DC breakdown of air is generally taken as 8000V per inch, dry air at "sea level" pressure. Have to derate for altitude and humdiditty. That's 1000V per 1/8 inch (3mm for our metric friends). There's not all that much room inside a toggle switch; they must be carefully designed & built to keep charged metal parts well away from the presumably grounded frame.

[J M Fahey]

Well, those are good switches, I've seen far worse.

Of course , there's a difference between "Legal" rating (now I've grabbed you bay the balls and will sue you to death) and actual physical or Lab testable arc/flashover voltage.

Those Carlins state:

Certifications UL CSA
Ratings 10A 250VAC

Now here 250VAC does not mean plain "connect it to 250VAC (say, a transformer secondary) and it will switch reliably" but "it will stand anything possibly present at an (up to) 250VAC mains line" and that's where we open the can of worms.

250VAC is specified so it covers standard 220/230/240 VAC used by most Countries in the World, of course.

Living in Argentina I'm not familiar with UL and CSA (our norms follow European types such as old DIN or modern TÜV/SEMKO/IEC) but in general all agree on most aspects: basically electrical equipment must stand VAC peaks "plus lightning hitting the power lines out in the open".

Since power lines have overvoltage protectors such peaks get converted down at distribution transformers but are still very dangerous.

To write a practical, enforceable Law, they usually consider that equipment must stand "twice peak voltage + 1000V " as a safety factor.

So 110/120V equipment should stand some 2 * 1.4142 * 120VAC=340V + 1000V = 1340V peak and in practice gets tested at 1500V DC , and 220/240V equipment gets tested at twice that, or 3000VDC.

That's the "HiPot" testing label you'll often see attached to, say, Power Transformers or SMPS power supplies.

That's why on the modern version of the Death Cap which in fact is mandatory to reduce radiated noise you'll see "250VAC" rated caps which meet above stated requirements while old Blackface/Silverface era .05x600V rated ones do not ... although 600V DC is way above 120VAC peak voltage , some 170V .

So in a nutshell , 250V AC rated switches which meet safety agency ratings must in practice stand way more.

In fact Carlin guarantees, in writing:

Dielectric Strength 1,000 V RMS (minimum)

which means over 1400V peak so certainly they will flash over with at least twice that ... they would have a very stupid or careless Legal Department if they let them publish such specs without at least 2:1 margin .

Back to tube amps: if you switch , say, 500VDC , they should be well within ratings.

Another hint that they are very good: they are specified to switch big line powered electric motors, which are inductive as H*ll , we are talking lathe, saw or drill machine ones, not a desktop fan one:

Ratings 15A 125VAC 3/4HP

[pdf64]

'Another hint that they are very good: they are specified to switch big line powered electric motors, which are inductive as H*ll'

Won't switches intended for such ac loads be a 'slow make / break' type, whereas for dc, quick make / break is preferred?
See p5 of http://www.aeroelectric.com/Referenc...ing_Manual.pdf

[Loogie]

I've found something to like in ultralinear as well, especially in an open loop amp where it doesn't sound a whole lot different to me than negative feedback. I think much (not all) of the bias against it is from people who've heard bad things about an amp (I think it was a Fender Twin of some kind) that had lots of negative feedback and used ultralinear taps. I'll bet many people who don't like it have never heard it.

If you get desperate you might be able to use one of the plug switches, like Marshall used for speaker taps. You could also kill the power for a second or two and then flip a switch. I too wonder why all those switches rated for 250v are used for standby in amps with 400+ DC.

[pdf64]

I think it was the 2BL62 that I've used for a UL / pentode switch, on a 600V 2xKT88 amp http://www.carlingtech.com/sites/def...ls_%26_COS.pdf
Only to be used with the amp de-energised, definitely not whilst in use!

Suitable 250V switches are probably used due to their capability being well proven, and to price and size.
Even if 500Vdc switches can be sourced, they tend to be big and expensive.

I'm not sure that the late 70s Fender 'UL' amps had lots of negative feedback?
eg see http://www.thevintagesound.com/ffg/s..._135_schem.jpg the proportion of the 4 ohm output signal that gets back to the LTP is only a little higher than the regular Marshall (100/920 compared to 4k7/104k7) and the open loop gain is probably lower to start off with, due to the 12AT7 LTP with 47k plate loads feeding ~68k grid leaks.

[tedmich]

UL amps configurations do not make a big difference in guitar amps, IMHO. There are other tap configurations for max power, but few bother.

[g1]

Won't switches intended for such ac loads be a 'slow make / break' type, whereas for dc, quick make / break is preferred?

Whoops! Yes, I should have linked the Carling 110/216 series instead (quick make/break) :
110/216-Series | carlingtech.com

I was just going by the picture and picked the wrong one. Now how many people do you think make the same mistake and end up with slow make/break in their (or someone else's) amps?
Also, many switches do not have DC ratings on them at all, and way back when I don't think any did.
Haven't seen all that many bad standby switches though, so maybe they are most often up to the task.

[Alan0354]

I am not suggesting or condoning using a switch that is not rated to 400V. But I did give a lot of thought. I feel as long as you use a switch with big plastic body that has like 0.5 inch surface distance on the plastic between the contact to the metal housing that screw onto the ground, I just don't think you have any problem.

I think the rating is for voltage between the contacts. eg. if the switch is open, the two contact can only stand the rated voltage before any possibility of arc over, not from contact to ground. In UL switching, you are switching between +B(fixed screen) to UL tap. UL tap is usually 40%, as CT is at 450V, the lowest voltage the plate can go is say 100V. Then the UL tap will never swing below about 220V or so. So the difference between the contacts is not more than 450-220V=230V!!! The voltage gradient is not that high.

Ha ha, I don't have anything to say about the standby switch. That actually put the full +B across the contacts. Well, old Fenders are about the most reliable amp, I personally never seen a standby switch blown.

All that said, I just use a big switch myself. Do it at your own risk!!!!

[g1]

Yes, if the arc over is just between the switch contacts, then no big deal. Maybe that's how they can get safety approvals, it's not really a safety issue if the standby switch arcs closed.
Kind of similar to voltage ratings on fuses, if a blown fuse arcs over, only to blow again, we don't really care.

[Alan0354]

Side step a little. If you are going to put a switch for UL, put another switch to go between that and triode connection!!! It's in the same area, you'll have the option of lowering the power from 50 down to say 15W. I had that in my former Marshall, I don't particular like it, but others do. Just a thought.

[Enzo]

The thing we need to worry about is switching. When the contacts open, will it then strike and maintain an arc across. we are not as much concerned that something will just by itself arc over to the frame.