SVT 2 Pro Fault Troubleshooting

[Zackdude]

I was given a busted SVT 2 Pro and I am currently hunting down an issue of it tripping the fault circuit with 3/4 volume. I have replaced D29, D30, D31, D39, D40, D41 as they were shorted as well as performed the resistor replacement & diode removal called out in TIB003 from SLM Electronics. It now seems a bit better in that it only trips when the drive is set high. What would be my next steps? I did see in another thread about the bias control op amps being replaced as well. Also does anyone have a good understanding of the power fault circuit in this amp?

[nevetslab]

First off, Welcome to the forum!

I service Ampeg SVT's all the time, including the SVT2-Pro. The Power Amp section is virtually the same as the SVT-CL, using the same AC mains, Output and Main Power Amp PCB assemblies. When I have an SVT (not SVT-VR, which has NO protection circuit other than a fuse), I pull it all apart, first removing the 6 power tubes and labeling them as to their position. Top rear view, has the tubes in two triangles. rear row, left-right is V4, V2, V3, and front row, left-right is V5, V6 & V1.

The protection circuit is measuring voltage drop across the six 10 ohm cathode resistors of the power tubes. Each cathode has a circuit that looks at the current flow thru each, with feeds to the window comparator that drives the Red and Grn LED Bias metering, which covers the range of around 14.7mA to 29mA, if memory serves (got it written down someplace). The other circuit for each sums up and feeds the Fault Circuit. I don't have my notes handy as to the threshold of where IC2B flips state when the summed potential exceeds the limits, but the voltage at the buss input to IC2B inverting input at R47 is where that takes place. If you exceed the trip level, it flips state and fires the -14V level to turn the relay off that shuts down the HT supply and activates the RED fault light.

Before I deal with the six power tubes (unless there is an obvious bad tube present, white inside, cracked glass bottle, etc), I'm opening up the amp, first getting the power tube PCB removed for inspection, checking for open screen resistors (as well as seeing if it has 220 ohm 1/2-1W resistors instead of the older 22 ohm resistors with diodes across them). I mark down any failed resistors to mate up with the associated tube as suspect). I'm then looking at the PCB for solder fractures thruout that board. Then, looking at the solder joints on the main PCB, particularly the rear supporting components..the two bias pots (often source of solder fractures), the Preamp and Heater Transformer header terminals (usually having fractures), all the driver and input tubes of the power amp (often have fractures), feedback terminals on the output edge, ribbon connector terminals, etc. With the sheer weight of these amps, if they travel constantly to earn a living, solder fractures develop. I constantly find them and cure them before I'll even power the amp up.

After having thoroughly inspected all the power amp boards, I'll also look at the preamp board for the same thing....solder fractures. And, while I have that board out, I"m addressing the loose front panel handles, normally replacing the # 10-24 PHMS with socket cap screws, flat and split lock washers, driving them firmly home. Usually find fracture on several pots, as well as the input jacks.

The support standoffs for the front-facing edge of the main PCB all too often are loose. the one on the output transformer side is ALSO the System Ground path to chassis. That MUST BE TIGHT, or the amp will have hum that will drive you crazy trying to find where it's coming from. That PCB mtg screw has to be tight as does the standoff with it's M3.5 screws securing the standoff.

Also...if the mounting screws for both the power transformer and output transformers are loose (and most likely are), you have to lift the main PCB up to get a the two screws hidden underneath it for the output transformer. PITA, but, that's the drill. Tightening up the core bolts on the SVT-2 Pro amp is nearly impossible...you can only get at the top two bolts.

With all this busy work out of the way, I"m re-assembling the amp, so I can begin inspecting the power tubes. I stand the amp up on the power transformer end, so I can get at the six cathode resistors of the power tubes from the bottom of the PCB.



Four of the six cathode resistors are along the bottom edge, positioned as V3, V1, V2, V4, with V6 & V4 rotated 90 deg and up from V5's cathode resistor. Look at the PCB layout for the main power amp for details on that. I thought I had a photo showing which is which...it's probably on my shop computer, and I'll look for that in the morning and edit this post.

At any rate, I check one pair of power tubes at a time.....V1/V6, V2/V5, V3/V4. If you encounter a bad tube, it will usually flash on you and immediately fire the protection circuit. I"m also looking to see what the plate current balance is for the set of tubes. I'll arbitrarily re-set the two bias pots for 22mA on the upper and lower tube, then as I swap out the tubes, I"m noting what the current is on the remainder of the tubes. (you're measuring DCV across 10 ohms, so 220mV = 22mA) I have a 4 ohm dummy load attached, running the amp on 120VAC thru a metered' variac, and watch the ammeter closely each time I switch out of Standby. I let each pair of tubes warm up for a few minutes, and will watch the plate current level for a minute or so to see if there's any trend in sudden increase. Invariably, I find one or two power tubes at fault, that often have an associated open screen resistor. I also check the six cathode resistors on the man PCB.

If I have one or more bad power tubes, then I"m at the decision step. Do I have any 'pulls' of the same mfgr 6550 or KT88 tubes that I could try to fill out the remainder of an otherwise working set of power tubes. Or, bite the bullet and buy a new matched sextet of power tubes. Even with a new fresh set of power tubes, I'll still go thru this current matching step, and group the six tubes for the best grouping, and do a final biasing, setting them up for a nominal 22mA plate current. Once I have all six tubes matched and grouped well, I load them all in and re-check the plate current, letting them idle for a bit. I normally have a fan on them at that point, as is normal in operation. If all is stable, I begin re-assembling the amp.

I also install hi temp silicon tubing on the hold-down clamps to protect them during transit.



The tubing over the spring/top hat joint is probably the most important sleeve, as I've had numerous tubes fail from that spring joint digging into the glass of the tube.

From here, it's back to re-assembling the amp and checking the results once all is back together.

[glebert]

Responses like that are why I love MEF! That is years (maybe decades) of experience and wisdom given out purely to help someone else out.

[Zackdude]

Thank you Nevetslab! This is a huge help. With such a thorough reply, I thought i would be thorough with my endeavor. I started the tare down, I found that the bias pots solder joints seemed inadequate and the xlr connector definitely had a bad solder joint (attached picture). There was also a missing screw from the output transformer as well as a locking nut on smaller transformer (filament transformer?). Also was missing some locking nuts for the power tube sockets. I am surprised the whole power tube board is held to the chassis with just those tube sockets! Preamp & Graphic EQ boards seem to have no problems. Did find a burn mark on a connector for the heater wiring to the power tube board. Other than that though everything seems to check out visually. How about some of the electrolytics on the preamp board? I feel like for a 20 year old amp it wouldn't hurt to replace a few now that I have it completely open. I also was able to figure out the layout for the cathode resistors. I marked up an attached picture showing which tube has the associated cathode resistor terminals on the power pcb. Next is fixing these few small items, cleaning the dust out, and on to checking the bias on all the tubes. Thanks again.

[nevetslab]

Yup.....you defiintely found the solder fractures. I didn't specifically mention the XLR connector, but it's nearly always plagued with the radiused solder fractures. As are the bias pots. Can't afford any failure on the bias pot connections!

I found the photo I had marked up for identifying the cathode resistors from the foil side of the PCB. V4 is actually paired up with V6, and V5 is the one at the end of the four along the edge. As seen below




I get really nervous when I can't located missing screws, nuts or washers in any of the SVT's. Having keps nuts missing from the power tube socket mounts puts them into the heart of High Voltage land! There aren't any escape routes to the outside world normally, which means they're well hidden but lurking in dark hallways. On the recent Ampeg issues, the standoff's used between the power tube sockets and the PCB, they've been using loctite on the top panel hex drive binder head screws, which strip out easily when trying to remove that assembly, or, the screw will just spin, and unthread from the bottom screw, now that they've eliminated the stud-standoffs and gone with screws at both ends. I had to cut some screw heads off recently to get that power tubes PCB out!

Surprisingly, I 've only had to re-cap one SVT over the years. The Snap-in caps being used do hold up remarkably well.

I hadn't seen burn marks on the power tube heater wire's rt angle fast-off before. They now are soldering these two wires to the PCB, just to slow down disassembly for inspection.

At any rate, happy to be of service. I keep a daily Service Notes journal on all maintenance and repairs, so I have a good service history/database on all the rental inventory I maintain, as well as on anything that I've never seen before. Comes in real handy over the years.

[Zackdude]

Ok. I got it all back together. Found another bad solder joint for the center frequency switch. I followed your method of checking the tubes and cathode resistors in pairs. At roughly the 5 minute marker after turning on is where I decided to take my readings.

V1 - 230 mV
V2 - 209 mV
V3 - 212 mV

v4 - 223 mV
v5 - 193 mV
v6 - 220 mV

I noticed the readings were increasing at about 1-3mV/minute even at 5 minutes of warming up. Is this normal, should I have maybe waited longer? Also how do I group/match the tubes?

[Enzo]

I like to let an amp idle for 20 minutes before making adjustments. I do check in case it wants to run away in the meantime, but no permanent adjustments until it is completely warmed up.

[nevetslab]

Ok. I got it all back together. Found another bad solder joint for the center frequency switch. I followed your method of checking the tubes and cathode resistors in pairs. At roughly the 5 minute marker after turning on is where I decided to take my readings.

V1 - 230 mV
V2 - 209 mV
V3 - 212 mV

v4 - 223 mV
v5 - 193 mV
v6 - 220 mV

I noticed the readings were increasing at about 1-3mV/minute even at 5 minutes of warming up. Is this normal, should I have maybe waited longer? Also how do I group/match the tubes?

The 1 to 3mV change you’re seeing is 0.1 to 0.3mA, which is quite low, and is perfectly normal within a 5 min period from a cold start. As Enzo points out, making permanent bias adjustments shouldn’t be done until the amp has been running at least 20 minutes. I’ll look at the trend when I begin grouping a set of unknown power tubes.

The readings you’ve recorded indicates a close spread thus far. It’s feasible that you could get it tighter by grouping V5 with V2 & V3, and V1 with V4 & V6. Sometimes it doesn’t re-balance as expected, but I’d give it a try. If it gets even closer, then reset both sets for 22mA, and let it fully warm up before making final adjustments.

Interesting to note that Ampeg’s Grn/Red bias light circuit doesn’t’ work until all six tubes are installed. That’s one reason I begin with one pair at a time, setting them to a nominal 22mA on each half, and then see where the other four tubes fall in. It looks like, in your case, this will work out.

When I’m letting it ‘burn in’, I have a fan on the tubes. You could put the cover back into place, since the fan is built into the SVT2 Pro’s chassis.

[Zackdude]

got it all matched and biased today. Plugged in and starting playing while increasing master with gain set very low. Once i started playing with the gain, i get a lot of distortion . Further increasing the gain i blew the fuse. Back to square one.

[nevetslab]

At this point, you at least have the rear panel bias LED's to assist, since your bias settings should have parked the power tubes in the middle of the bias range. You do have the Power Amp input jack available to help isolate where the trouble is. I have had one of the 12AU7 driver tubes in the output stage go bad, and totally upset the bias level, throwing one side of the output stage into high current. But, blowing the 10A fuse....I hadn't had that problem.....it just went into protect.

If the amp is now pulling high current as soon as you switch out of S/B, you may have one or more failed rectifier diodes in the power supplies. I had an SVT-CL with that problem just recently, found it was one of the HV supplies' rectifiers....forget which one. Shorted diode in there will take out a mains fuse immediately. This may be a bit tricky to locate the source. Usually, issues in the preamp section won't take out the mains fuse, unless there's power supply issues that don't show up until running with signal. The Fault Protection circuit concerns itself with the output stage. Have you a fresh pair of 12AU7's and a 12AX7 to re-tube the power amp circuit?

[nevetslab]

got it all matched and biased today. Plugged in and starting playing while increasing master with gain set very low. Once i started playing with the gain, i get a lot of distortion . Further increasing the gain i blew the fuse. Back to square one.

Any further development on the problem? You still may have one or more power tubes that go bad under drive. You can operate the amp with one or two pairs of output tubes installed, for troubleshooting purposes.

[Zackdude]

Update, there was only a 6A fuse in there. Replaced it with a 10amp. Reset the bias as close as I could to 220mA per tube, let it warm up for 20 minutes. Will not go into power fault any longer on any setting. Cleaned up the pots and sliders, replaced the knobs with the correct stock ones and cleaned up some duct tape gunk on the chassis. Runs great now! Thank you so much for your help.

[Zackdude]

Also is there a way to remove a previous post? I dont see that option. Wanted to remove the picture I attached of the incorrect layout of of the cathode resistor terminals

[Enzo]

You cannot alter posts after a period of time for editing. If we allowed unlimited editing, threads would get rewritten all the time and responses would no longer make any sense. And in the case of occasional arguments someone could post something, argue about it, then change the original and deny having said it.

220ma per tube????? with 660v on the plates, that would mean the tube was dissipating 145 watts at idle. I don't think so.

[Enzo]

You cannot alter posts after a period of time for editing. If we allowed unlimited editing, threads would get rewritten all the time and responses would no longer make any sense. And in the case of occasional arguments someone could post something, argue about it, then change the original and deny having said it.

220ma per tube????? with 660v on the plates, that would mean the tube was dissipating 145 watts at idle. I don't think so.