ST-120 Bob Latino Kit Fuse Issue

Discussion in 'Audio Hardware' started by Jay Kobelin, Apr 19, 2015.

Thread Status:
Not open for further replies.
  1. Jay Kobelin

    Jay Kobelin New Member Thread Starter

    Just completed the ST-120 this week (April 18, 2015) and is OK except it blows its fuse under any volume condition and NOT on turn on. It plays fine for 10 minutes, sometimes for 30 minutes and BANGO a fuse blows

    We are using the 5 A slow-blow fuse

    Very odd....rectifier tube shorting? Kind stumped by this

    pcb4u at att dot net

    Thanks in advance
    Jay
     
  2. smctigue

    smctigue Forum Resident

    At the risk of stating the obvious, you should contact Bob.
     
    thegage likes this.
  3. Jay Kobelin

    Jay Kobelin New Member Thread Starter

    Well, we did and the email seems to indicate that the 5AR4 is shot but a web search points more to a systemic problem in the apparent weakness in the 5AR4 and who makes them. This indicates a rather "sticky" power supply design based on element spacing in a tube. Is the GZ34 better? Also noted was the Mullard was a better manufacturer of an 5AR4.
     
  4. chervokas

    chervokas Senior Member

    Well, I've seen a lot of modern Chinese 5AR4's fail early in spectacular ways, haven't used any of the Russian New Sensor-made 5AR4s but if you're going to use tube rectification in an amp NOS or even tested-good used Mullards or Amperexes or Sylvania, etc may be essential, of course a NOS Mullard 5AR4 these days could cost you $200, but it will last a long, long time -- that is presumping the problem with the amp in question isn't related to the build. If a newly build kit amp is blowing fuses and rectifier tubes on first turn-on, the first place I'd look for a problem is the build -- I'd go over all parts and make sure everything is wired correctly (paying particular attention to the polarity of any electrolytics and diodes), make sure there are no cold solder joints, make sure nothing's shorted to the chassis, etc. Is it blowing the fuse and rectifier tube right away on turn on with no output tubes in place? I'd also start the testing not w/ pricey NOS 5AR4s.
     
  5. skriefal

    skriefal Senior Member

    Location:
    SLC, Utah
    5AR4 = GZ34. They're the same tube. A vintage Mullard GZ34 will do better, but I'd recommend instead that you use a Weber Copper Cap WZ68. This is a solid state rectifier replacement with twice the current capability of a GZ34, and they sell for about $20 from Weber. Or you might try a vintage Mullard GZ33 or GZ37.
     
  6. Jay Kobelin

    Jay Kobelin New Member Thread Starter

    The unit turns on fine ...can run for 10 minutes, sometimes an hour and then the fuse blows. The 5AR4 arcs then apparently takes out the fuse when we see an arc in it. So I assume the kit is wired fine. This area of the amp appears to be its weak spot as other blogs talk about the 5AR4 weaknesses and now the seller of the kit says our line voltage is too high....it is high from what we had in CA which was dead in 120 but here I have measured here in Texas 124 volts. Now if a 4 volts increase in a approx 4:1 step up transformer is going to throw a tube into arcing, then I'm convinced the design/component choice is pretty marginal.
     
  7. Jay Kobelin

    Jay Kobelin New Member Thread Starter

    Does this device have a built in timer so we don't hit the tubes with B+ with cold filaments?
     
  8. chervokas

    chervokas Senior Member

    Well, yeah, though you would think a modern variation on the ST-70 would take into account the high and fluctuating voltage in terms of the way the power supply is designed, I mean obviously the original ST-70 design was based on 115-117 VAC, and even then the power transformer was barely up to the task, but I would have assumed any modern redesign based on it would have appropriate regulation since we know we encounter much higher line voltages today. I'd still think about getting the voltage to 120 w/ a variac and testing all the voltages on the amp to make sure everything's where it's supposed to be before throwing a $150 NOS rectifier into a situation where it might blow (I mean, are those over voltages also cranking up the amp's heater supply too much?). Then, personally, I'd think about going with solid state rectification, and/or some kind of bucking transfomer or something to get the 124 VAC wall supply down.
     
  9. Jay Kobelin

    Jay Kobelin New Member Thread Starter

    I called the electric company and they said all houses should have 120 or 220 at the home..I knew I was in trouble :) I said its 120/240 and if can't be be just one isolated home. It has to be the distribution transformer for the area or their generators are running too high.

    Back to the amp, sad that a 21st century kit could be sensitive to a 3% increase on line voltage. It it the 21st century and the advent of new components why design a transformer to mate up with marginal rectifier tubes than with the famous 5U4GB or go from day one and design around solid state rectifiers. Two 10 cent 1N4007s will do the job just fine, no rectifier heaters to deal with...just build a good filter and that's it. It's the DC doing all the work. One can stack caps to get some nice filters .. They need not be in one can .
     
    Last edited: Apr 20, 2015
  10. chervokas

    chervokas Senior Member

    Well, if that kit is relatively faithful to the ST-70 source, you have crummy new tubes, and a circuit where the rectifier was already operating on the edge in terms of the current demanded for a 35-watt per side EL34 based circuit. I assume there must have been some changes made in the st-120, but even with the higher voltages and bigger power transformer there's still just the single 5AR4 right? That seems like an amp that would probably operated better with good solid state rectification.
     
    Last edited: Apr 20, 2015
  11. Jay Kobelin

    Jay Kobelin New Member Thread Starter

    I would go with the Weber device but the quad cap can is rated at 525 and not too sure (without checking) whether the resulting increase in DC voltage would be too high for the remaining components. My opinion is this new kit is just rebuilding the the older style with some beefier stuff but did not take into account the ranges of allowable AC power in the USA. The ANSI standard has the range near 125 so it's all so questionable.
     
  12. thegage

    thegage Forum Currency Nerd

    With the warm up of a 5AR4/GZ34 the B+ comes up slowly, too, so no timer is necessary.

    Then maybe you should do some more investigation before forming an opinion. Bob's kits are not simply the same circuit with beefier parts, and they clearly are designed with modern voltages in mind. His kits are very well regarded and generally bullet-proof, and he says that the Weber is an option in some of his amps and a standard part in others. There is some concern whether the Copper Caps come on too fast and hit the filaments too hard, but others have said the Copper Caps have enough of a delay to not be an issue. If you're concerned about this you can add a timer to his amps.

    I've used an NOS Mullard rectifier, Chinese, Soviet, and Weber. The Weber certainly is worry-free, but to my ears it may add a bit more bite, or edge than I prefer. It's a low-cost test to see if it solves the issue.

    John K.
     
  13. googlymoogly

    googlymoogly Forum Resident

    I agree re: the Weber copper cap. I've been running a NOS Mullard GZ37, which can readily handle the current the ST120 puts out. I've had no problems at all during the 5 years I've been using the amp. OP, you may or may not have seen this on Bob Latino's website: http://dynacotubeaudio.forumotion.com/t565-gz34-gz33-or-gz37
     
  14. Burt

    Burt Forum Resident

    Location:
    Kirkwood, MO

    Not a real big fan of the Weber, it's just two diodes, no current limiting, no nothing. There was some discussion of them failing en masse in Collins transmitter power supplies and there was a lot of discussion on this.

    IMO it behooves the modern supplier of units to design around the tubes available now and if he is saying that only NOS rectifiers are good enough that is something we need to discuss if I am buying the stuff!

    I feel as though he should have determined that the normally available new production tube would work or have used a solid state rectifier.

    One good fix would be to construct a solid state rectifier plug in with seriesed bypassed diodes and dropping resistors that have some nice cooling airflow by being on a plug in base and sticking up just like the tube.
     
  15. Jay Kobelin

    Jay Kobelin New Member Thread Starter

    I never cared for quad caps and that goes back to the 60s....one section down, the entire can comes out or one needs to find someplace to hide a new cap. Secondly, the transformer could have been for a lower secondary voltage, no 5 volt winding needed and use a couple of 1n4007s and that would be that. For filtering, I would have stacked 400V caps with equalizing resistors to get 800V capability and the value capacitance needed which would have coasted at a B+ of about 480....this is just what would have been my approach .. trying to keep the ole look is great but when it comes to on the edge components, well....

    We have for now opted for a Mullard GZ34 and see what happens. It appears from local shops that have tubes (guitar, music, repair) that the tube we got was a cheap Chinese make.....so we will see what this new tube leads us.

    The use of a variac in today's electronics is really, IMHO, a way to work around a situation that should not exist.

    ANSI power standards for the USA I believe allows line voltage to ride up to 125 volts and all equipment built and sold today should take these documents into account.

    ..... Just my opinion....
     
  16. WntrMute2

    WntrMute2 Forum Resident

    I would try the Weber Copper cap and see if it solves your problem. Then go from there. I have the same amp and just love it! Yes it is a little temperamental in the rectifier department and MAYBE the designer could have done better but I tell you the sound is great and the cost reasonable. Put the Weber in, see if it solves the problem. I have mine working well with a Sovtek rectifier tube in combination with the "diode mod" as well as the delay timer. It did blow some fuses and eat a few rectifiers on the process.
     
  17. skriefal

    skriefal Senior Member

    Location:
    SLC, Utah
    Re: Weber Copper Cap WZ68

    Yes, it has a short delay (~5 seconds). But it really doesn't matter -- the cathode stripping fears are overstated. Latino/VTA themselves recommend the WZ68.
     
  18. Jay Kobelin

    Jay Kobelin New Member Thread Starter

    We got a Mullard GZ34 locally and the amp is running fine .. Ran it one hour not an issue. Pity the crappy tubes out there...but still believe the power supply should have been rethought for solid state .. I've been a ham for over 55 years since the advent of the power diode (from the days of mercury vapor rectifier tubes, selenium rectifiers, etc.) and I have never had to replace one diode.....relays, that's another issue :)
     
    Last edited: Apr 20, 2015
  19. Jay Kobelin

    Jay Kobelin New Member Thread Starter

    Update..wasn't a Mullard....we have that on order...got some other GZ34...restarted after shut down for an hour, couple of small arcs in this tube but didn't take out fuse, amp still running. Is there some type of forming required with this generation of marginal tubes like some tiny imperfections that will "flake off" during its burn in.....never saw such touchy stuff in tubes as we have witnessed here .. Makes owning one of these amps like walking on egg shells on each power up....
     
  20. Burt

    Burt Forum Resident

    Location:
    Kirkwood, MO
    Power companies are worried about voltage too low, because that makes electric motors run hot, not too high, unless it is way too high. This is why we have "voltage creep". As higher and higher portions of the load are switchmode supplies that have a 40 to 50 percent permissible spread this voltage creep will continue.

    Still, in the old days, you could run 130 to 135 volts with no short term issues on most consumer tube gear because there was a reserve built in on all components. Over time it would cause tubes to die early an cause the box to run hot but it would not cause quick failures.

    If tube rectifiers must be used on a 120 watt output Class AB tube amp there is no single rectifier in production today that will do that well. In the old days you had some pretty stout vacuum rectifiers plus mercury vapor and xenon ones. Not so much anymore. He should probably have used two. Also, as you will note by reading any good tube manual, there is a minimum DC resistance permissible in the anode circuit and a maximum capacitance acceptable for capacitor input filter tube rectifiers. The old ARRL Handbooks from the postwar thru JFK/MM era spell this out well too.

    Some people refuse to "read the f(antastic? fine? whatever....) manual". You can't make them, I can't make them, just isn't going to happen.

    BTW if you are using 1N4007s you want to use three of them in series, bypassed with a few pF of capacitance and a very high value resistance, on each leg of the transformer as per both the ARRL Handbook and the mid 70s Fender silverface guitar amps designed by actual engineers, you know, the ones that just keep working...
     
  21. skriefal

    skriefal Senior Member

    Location:
    SLC, Utah
    That would have lost them a lot of sales. A large percentage of the buyers of these amps want tubed rectifiers... or no sale. I do think that they should have redesigned the chassis to allow for dual 5AR4/GZ34 tubes, but I guess they were worried that doing so would drive away too many buyers because the amp no longer looks exactly like a vintage Dynaco ST-70.

    I do use a Weber WZ68 in my own ST-120. No problems yet.
     
  22. skriefal

    skriefal Senior Member

    Location:
    SLC, Utah
    The new Chinese (Shuguang) 5AR4/GZ34 rectifier tubes are generally reliable. As long as they're not driven to the ragged edge -- as they are in an ST-120.
     
  23. Jay Kobelin

    Jay Kobelin New Member Thread Starter

    Thanks for the reply ...good to read some points I have heard for years. The issue with 3 diodes in series I beleive goes back to the old days when the diodes first came out and equalizers were required to keep things equal. I haven't used equalizers in any designs with the 1N4 series in the last couple of decades. If I recall the PIV is 1000V for the 4007 so not much stress in this amps supply. If one wants to play it safe on the PIV, just use 2 in this case....I take it your a ham by your references.....great hobby, led me to engineering as a career for 40 years in aerospace.
     
  24. Jay Kobelin

    Jay Kobelin New Member Thread Starter

    We would go to solid state but our line voltage is a bit high and not knowing where the final B+ will result, we'll try this setup. Sure don't want to take out any other stuff in the amp. Amp sounds fine, now need to build the preamp and get the new turntable spinning. The preamp kit is a bit "scarier" as all the parts were dropped in one bag so each piece needs to be ID'd or measured whereas the 120 kit had the parts separated in bags and values noted on each bag.
     
  25. Burt

    Burt Forum Resident

    Location:
    Kirkwood, MO
    The buyers are a bit odd since the ST70 was and is the commonest and pretty much the most plebian looking amp ever built. A larger chassis would have been a big win, a lot easier to work on, run cooler and all.

    McIntosh used a pair of 5U4s on the MC60....a single channel 60 watt amp. That should be informative because Mc did not spend any more than they had to either.
     
Thread Status:
Not open for further replies.

Share This Page

molar-endocrine