Start-up Surge

[shaun]

Heyo. I know this has been covered to some extent elsewhere, usually in conjunction with the debate over whether a stand-by switch is necessary. I've searched the older posts but haven't found answers. So here's the dealio...

I built a 6L6 SE amp and used IN4007s for power rectification. I put my voltmeter on the B+ to see where the surge went to. It went to 650v. Kinda high. It is hard to catch because it is so brief, and then the amp immediately settled at 450v, and as it warmed up, settled at what I expected - around 413v on the first node. (The first node is a 47uF cap, followed by a 300R resistor - it is simply for power filtering and does not connect to the OT).

I thought, maybe the old debate about warming up the tubes to avoid surge might hold water, so I warmed up the tubes, shut off the amp, drained the caps, and rebooted. Still 650v surge. I suppose the tubes could have cooled to some extent within the 30 seconds or so between shut-off and start-up, but that 6L6 was still hella hot. I deduced that warming the tubes made no diff to surge conditions.

I added a 150R resistor before the 47uF node, but it did nothing to control the surge. If I go higher, I'll change the nature of the power filtering.

In my considerable research, I don't see quick fixes, just debate over stand-by switches. My primary concern is for the longevity of the power filtering caps. I'm using F&Ts - nice, but not cheap - and don't want them dying early. I am not 100% sure, but I think they are rated for 550v for brief surge, although they are standard 500v F&Ts.

Sorry if this is rehashing old news, and thanks for any guidance.

[AlNewman]

I believe thermistors are added in several amps to help control inrush current.  They are placed between the fuse and power transformer.  Resistance changes as heat increases.

Most fender amps have them.

[shooter]

I typically use 2 caps stacked in series as my 1st filter cap, using 350v X2 gives you 700v and 1/2 the capacitance

[dude]

So, what’s the answer to Shaun’s question?
Caps in series for higher surge..?
And thermistor, can one lead me to a Fender schematic with thermistor between fuse and PT..?
I’d like to know more too, about surge at start up.

[Esquirefreak]

I've seen some folks come up with slow-starting circuits with relays and transistors. Maybe worth searching the web and reading up? I think I searched for "solid state rectifier slow start" or something like that.

[TitaniumValhalla]

JJ and Mojotone both make 100uf/100uf multi-cap cans that would give you 50uf 1000v in series for about $10. That's about as cheap and elegant a solution as I think you're going to find. Standby switches are fun if you want to feel like you are performing the startup sequence of a rocket ship but otherwise I'm not terribly interested in them.

[AlNewman]

And thermistor, can one lead me to a Fender schematic with thermistor between fuse and PT..?
I’d like to know more too, about surge at start up.

https://el34world.com/charts/Schematics/files/Fender/Fender_hotrod_deville.pdf

In my super reverb it's placed between the plug and the fuse.  In my deluxe reverb, it's placed on the neutral side.

[tdvt]

I typically use 2 caps stacked in series as my 1st filter cap, using 350v X2 gives you 700v and 1/2 the capacitance

I have adopted this same method, along with the 220K balance resistors as Fender used in the bigger amps.


With the radial caps I have been using recently (2x82uF/500V) I have a large amount of headroom.

[PRR]

It takes real time to blow-up an e-cap. Last time I did it, at about 50% over-voltage, it took 20 minutes.

Are you really running 460V of AC? In a SE amp?? Seems mighty high.

[d95err]

JJ and Mojotone both make 100uf/100uf multi-cap cans that would give you 50uf 1000v in series for about $10. That's about as cheap and elegant a solution as I think you're going to find. Standby switches are fun if you want to feel like you are performing the startup sequence of a rocket ship but otherwise I'm not terribly interested in them.

Multi-cap cans typically have a single common negative pole. You can use the two halves in parallel, but not in series.

[shooter]

here's a fun start-step if you're really bored.

[pdf64]

That 650V surge is confusing me.
It may be something weird with the meter perhaps 
I’ll try to get around to scoping this on an amp or 2, with a 100:1 probe.

[shaun]

JJ and Mojotone both make 100uf/100uf multi-cap cans that would give you 50uf 1000v in series for about $10. That's about as cheap and elegant a solution as I think you're going to find. Standby switches are fun if you want to feel like you are performing the startup sequence of a rocket ship but otherwise I'm not terribly interested in them.

Thanks for this info, TV. I think that's the way I'm gonna go. Simply and effective.

For those interested, a vid popped up in my youtube feed (AI seems to read my mind in scary ways lately - those algorithms!) in which the tech gives a great explanation and possible solution to in-rush current when using 33,000uF (Yes, 33,000uF) filter caps in a hifi amp. He calls is a "soft start."

&t=1517s

Thanks all.

[shooter]

AI seems to read my mind in scary ways lately

you ain't seen nothing yet!!

[shaun]

here's a fun start-step if you're really bored.

Very interesting! Thanks Shooter.

[shaun]

I typically use 2 caps stacked in series as my 1st filter cap, using 350v X2 gives you 700v and 1/2 the capacitance

I have adopted this same method, along with the 220K balance resistors as Fender used in the bigger amps.


With the radial caps I have been using recently (2x82uF/500V) I have a large amount of headroom.

Yes! This looks like the way to go. Much appreciated.

[bmccowan]

It went to 650v. Kinda high. It is hard to catch because it is so brief, and then the amp immediately settled at 450v,

PRR:

It takes real time to blow-up an e-cap. Last time I did it, at about 50% over-voltage, it took 20 minutes.

So, are you worried about something that is not really a problem? If so, maybe worry about that algorithm. Wife and I were using Google maps for directions while driving. We were discussing a medical issue. Within an hour we were getting texts about related medicines. Yikes!

[shooter]

Yikes!

yep, i spent the winter detoxing from my phone, it now stays home, in one place. just like in the corded days of old, kinda refreshing knowing i'm clogging the data vacuum with dead air:)

[bmccowan]

I never fully warmed to my cell phone. The young engineers in my firm would say, "I tried to call you, but it just went to VM" I'd tell them that my phone was turned off. Shocked, they would ask why, and I'd say, "so you can't bother me." Half the time I do not know where the damn thing is.

[acheld]

Amphenol CL-70 thermistors work great as inrush limiters.   Mouser #: 527-CL70. 

[shaun]

Amphenol CL-70 thermistors work great as inrush limiters.   Mouser #: 527-CL70.

Thanks acheld. I wondered about thermistors. Has anyone here tried them and can report back? Any sag or anything of that nature?

[sluckey]

I have a C60-11 in my Reissue Twin Reverb. No sag. Most all the Fender reissue amps, from Princeton Reverb on up to Twin reverb use the same C60-11 thermistor.

[Soulfetish]

thermistors are a simple effective of limiting inrush current, which can happen when charging caps from "zero" during startup. But very high current spikes can happen for other reasons we don't necessarily have to get into here, because they would be random in most cases where the user flips the power switch arbitrarily.
Is the thought here that a thermistor would limit the current and slow the charging of the caps long enough that the secondary would charge to the semi-unloaded 450V? If so, I like it :) But it would be interesting to see if that was reliably consistent.
I'm wondering if what the OP is experiencing is transients in that winding caused by the leakage inductance and parasitic capacitance in the winding and the capacitance of the diodes PN junction during the first few cycles?
It's not uncommon to see snubber circuits consisting of a capacitor, or resistor/capacitor combination across a secondary winding to limit transient spikes (although not often seen in guitar amps). Some of the heavys can speak to this with more authority, so I'm curious about their thoughts on this.
Just throwing it out there.

[AlNewman]

I would think unless fender has stock in thermister manufacturers, they must do something.

[acheld]

These thermally variable resistors work really well at limiting inrush at startup.

However, unless the user allows enough time between "turning off" the amp for the resistor to cool, they are not so protective.  This thermal lag explains why they won't contribute (much) to sag.

This data sheet has lots of good info.   https://www.mouser.com/datasheet/2/18/AAS-920-325D-Thermometrics-NTC-Inrush-031814-web-1315885.pdf