Scratching my head with a warmed up amp blowing fuses

[moonrock]

Just finished a hrm project that uses Mercury Magnetics JTM45 PT/OTs, 3 12ax7s, 1 12AT7, 2 KT66s, GZ34, and an extra 12v radio shack transformer to drive the DC heaters of 2 12ax7s and the 12AT7. The KT66s and PI are heated using the AC. The 12AT7 is used for a tube buffered effects loop. It runs quiet and sounds great. It has one problem: I loaned it to my brother-in-law. He played it in a band setting for a couple of hours. He went to move the amp, so he unplugged it from the wall (without turning it off), moved it, plugged it back in the wall and the fuse blew. I'm using a 3A slo blow. I've repeated this several times (plugging into the wall with the power and SB switches on) and the fuse does blow when the amp is warmed up. However, It won't blow if I turn the amp on in the correct order (Power then SB). Funny thing is, I can put a 1A fuze in there when the amp and tubes are cold and repeat the plug in test and the fuse doesn't blow. Do the tubes draw more current when they are hot? I realize one shouldn't be plugging things in with the power on, but I'd like the amp to be more robust than what it currently is. Thanks!
Bruce

[heisthl]

If you wired your standby correctly so the first filter gets charged on standby there may be something amiss with that first filter (marginal cap, bleeder resistor too small or series caps with no bleeders/etc) that only shows up when the downstream caps still have some residual charge. Filter caps don't like it when you exceed their voltage rating. Any chance you are using series filter caps and one is in there backwards?

[moonrock]

heistl -
Thanks for the thoughts. I am using series caps for the first 2 B+ nodes. The caps are 47uF/450v with 470k/2w balancing resistors. B+ is 461v so there is 230v across each cap. I've confirmed this with the DVM and the caps are installed properly. What really confuses me is that it only happens when the amp is warm. Also, if powered up in the correct order it doesn't happen (power then SB). So I was thinking it is some in-rush current issue. The difference between the cold start vs the warm amp start is the heat in the various components and the tubes. I've looked around inside the chassis and I don't notice any hot resistors, caps, or diods, so I was wondering about tube performance when hot. Also, I'm using a radio shack 12v/1.5 amp transformer to power the 5v relays and 6v preamp heaters. I basically rectify the 12v supply, then split it to 5v and 6v regulators, then filter. I have a 5 amp fuse, so I'll try that today to see if it is a matter of not having enough head room for the current spike transient. Thanks again.
Bruce

[heisthl]

heistl -
I have a 5 amp fuse, so I'll try that today to see if it is a matter of not having enough head room for the current spike transient. Thanks again.
Bruce

I wouldn't - safer to build a light bulb test fixture (limits current to +/- 1 amp) and use test clips to find the offending power hog by one at a time lifting B+ nodes,heaters, add on trnasformers,etc and observing the bulb brightness - it will be maximum bright when your wierd conditions occur. When you get it right a 1 amp fuse should hold even when the amp is warm.

[moonrock]

Well, feeling kind of silly, but I discovered I had a fast-blo fuse in. Put in a slo-blo and everything is good so far. I did the light bulb test and didn’t see anything funny. The bulb is bright on power up but then dims. Light gets slightly brighter when the SB switch is thrown. Hopefully it was just the wrong fuse.