CL-90 installation ?

[dodeca12]

I'm useing the chong cap board board and after three months my JJ rectifier tube flashed over. I ordered a new one as well as a cl-90 from digikey. I want to verify where I install it. Does it go in series with the black transformer lead? i.e. from the power switch? Richard

[TomMcNally]

Yep the current limiter goes in series with the primary of the transformer. I don't think that's gonna help with the JJ rectifier tubes ... they aren't the most robust thing in the world.

[dcriner]

I install an in-rush limiter for all tube-type equipment.

Since the ST70 draws 190W, I used a CL-80 rather than a CL-90. (The CL-80 is rated for higher amps than the CL-90.) Digi-Key sells both.

My theory is that a soft start will save wear and tear on the tube heaters. But, the B+ will be slowed down too, so maybe that would also help prevent flashover?

I think that rectifier flashover most often happens when the filter caps are over-sized. Does your cap board result in oversized filters?

Just because a rectifier flashes over doesn't necessarily mean it's ruined.

[erichayes]

Hi All,

I went out to the lab earlier this eve (before the power went out) and did a couple of tests on 5AR4s. The results confirmed my memories of my favorite high vacuum rectifier.

5AR4 #1 is a Mullard, labelled Dynaco, in good health. Warm-up time to initial emission was 11 seconds; full emission occured at 18 seconds.

5AR4 #2 is a JJ new stock. Warm-up time to initial emission was 20 seconds, with full emission occuring at 24 seconds.

My point: The 5AR4 is a great rectifier, regardless of manufacturer (JJ spent two years testing their version before releasing it), but it's not perfect. It should never have been used solo in the ST-70; there should've been two of them. Heater warm-up time, delayed or soft start B+ just don't enter into the equation. In a ST-70, the 5AR4 is getting the snot kicked out of it, and it will fail with annoying regularity.

I've been dealing with ST-70s since they were invented, and the biggest bitch owners have is the failure of the 5AR4 in the middle of a listening session-- they can and do spontaneously combust. And it is permanent unless you retire the tube to preamp PS duty, where the draw is 20 mA or less--and even then, all bets are off.

[sorenj07]

well, if there's room, just stick a couple 6DN3's in some novar sockets for fullwave rectification. with each damper's tolerance of 2.1A, you'll probably be arc-free Yellow_Light_Colorz_PDT_12

[Shannon Parks]

I know this is sacrilege, especially on this forum...
Sheldon
(waiting to be burned at the stake)

Welcome, Sheldon! Nahh, we won't burn you at the stake - we'll leave that for other forums. We do like sacred cows around here, though. They make the best burgers.

[dcriner]

Hi, Sheldon. A couple of concerns here.

1. Solid-state diodes have less voltage drop than a rectifier tube, so B+ will go up, for sure, which may not be good. You can add a dropping resistor that should resolve this issue.

2. A tube rectifier takes several seconds for the filament to warm up and start emission. This delays achieving full B+, which is a good thing -- might help avoid "cathode stripping" of the other tubes caused by applying plate voltage before their filaments warm up. Solid-state diodes will start charging the filter caps immediatedly, and B+ will come up faster. Maybe a CL-90 inrush limiter might help this?

Doug Criner

[sorenj07]

Or just slap a switch on the transformer's CT that connects to ground, and wait a few seconds before activating. I do this in my 6SN7 preamp with minimal problems. Also, try getting a bleeder resistor in there, because after you turn the amp off, the capacitors still put voltage on the now-cold cathodes

[erichayes]

Hi All,

As I've mentioned in other threads, cathode stripping is a very real problem with output tubes, but all we were talking about was output tubes. Well, guess what . . .

It is an issue with rectifier tubes; we just don't pay that much attention to it because rectifiers either burn out or blow up before they have a chance to wear out. But when they do wear out, it's because the cathode--whether directly or indirectly heated--doesn't have sufficient emissive material left to supply the current needed by the rest of the circuit. Generally, with filament type rectifiers such as the 5U4, the filament becomes so weak physically that it sags and touches a point on the plate structure, causing a short. 5AR4s, when they live long enough, will exhibit the same shiny plating phenomenon on the inside of the envelope that output tubes show.

But what, you say, about the 12X4s in PAS preamps, or 6X5s and 5Y3s in the little integrated amps? I have a wall of Berlant Concertone record/play preamps built in the '50s and used professionally for God knows how many hours, and every one has its original Berlant branded 6X5. In record mode, total current draw is around 40 mA, which is about half of what the 6X5's capable of handling. But the more meaningful parameter is the voltage on the input cap: 330VDC.

Cathode stripping is primarily an electrostatic phenomenon, rather than thermionic. 5U4s and 5AR4s typically operate in the 450~525VDC region, where electrostatic migration of particles just starts really waking up (although corona discharge has been observed at voltages as low as 90VDC).

Bottom line: If you use tube rectifiers, get comfortable with the fact that they fail, regardless of cause, and there's not a damned thing you can do about it.

[dcriner]

B+ will increase due to the solid-state diodes, and also increase some more if the values of the filter caps are increased.

The higher voltage on the filter caps is one issue, which can be resolved by substituting caps with higher voltage ratings. But another concern is putting a higher plate voltage on the tubes themselves. EL34s are rated at 450V. A stock ST70 puts 410V on the EL34 plates, correct? My preference would be to install a dropping resistor in series with the diode output to return the plate voltages to the "stock" Dynaco value. I think the output tubes would thank you

Doug Criner

[Uncle Ned]

It's not clear to me that cathode stripping is an issue with little tubes like these, and such low B+ voltages. Transmitter tubes, and way high b+ voltages may be an issue. If cathode stripping is an issue, why isn't it an issue with the rectifier tube itself?

I managed to cathode strip some 6550's by flipping the power switch on and off while tubes were warm. I know some Wurlitzer jukebox amps had that problem because the filament voltage was 10 volts when the amp was first turned on. I really don't think it's a problem with cold tubes that have a cathode resistor or bias voltage applied to them before warmup.

As far a transmitter tubes go, I think it depends on the tubes, mercury rectifiers, coated filament and indirectly heated tubes, I'd say, yes, you need a delay, but I used to know a broadcast engineer in DC who said you couldn't cathode strip thoriated tungsten filament tubes. Claimed he always threw the power straight on cold tubes of that type... note, if he was wrong, he'd have had some explaining to do, in the radio business, it's only OK for the DJ's and program directors to waste $5,000 at one flick of a switch...

[Uncle Ned]

But another concern is putting a higher plate voltage on the tubes themselves. EL34s are rated at 450V. A stock ST70 puts 410V on the EL34 plates, correct? Doug Criner

Philips rated EL34's at 500V wired ultralinear. The Mark II Dynaco schematic says the EL34's in that are at 465 volts, but I never saw a Mk3 or Mk2 with plate voltage that low, even with a 5U4 instead of a 5AR4. Throwing a resistor in there basically mimics what the tube rectifier does to the voltage, makes it vary with the current.

[Wharfcreek]

I'm one of the "new guys" around here....so my oppinion won't hold much value I'm sure. To be honest, I'm relatively new to Home Audio Tube Amplification....since last fall. However, I've been building Tube Guitar amps for about 20 years now. I tend to agree with a few basic comments I've read in this thread. First, tube rectifiers WILL fail. But then with a lot of use, every tube in the amp will eventually fail. The question of pre-mature failure vs. just "wearing out" is one you can debate forever. Did the tube fail because there was something wrong in the amp? Did it fail because it was "Defective"? Or maybe, is the design of the amp such that it pushes the tube pretty hard, thus a pre-disposition to failure at a rate, comparatively, more quickly than the same tube might fail in a different kind of amp? One thing I've not yet fully understood about a lot of these home audio amps is why, if "Stripping" is such an issue, did not the deisgners and builders install a "Standby" switch, as you see in virtually every tube guitar amp? Needless to say, when Dynaco designed their amps, "COST" was a big issue. My understanding is that the amp was designed to be at a cost level of UNDER $200 in kit form....a magic number, or so the marketing guru's thought. Well......no need for a stand-by switch at that price....it just adds another $1 to the cost of the amp. I happen to share the appreciation for solid-state rectification in tube amps. To my ear, it produces a sound quality that I like. It's a personal thing!! But one thing I can say for sure.....if you want to protect the life of the tubes in your amp....install a Stand-By switch......or,.... buy a transistor amp!!! (Oh, guess what......transistors fail too!!!)