build a dummy load

[Randall]

I did this one with the 40 HZ volt meter method. It was quick and easy.

[Leo_Gnardo]

I did this one with the 40 HZ volt meter method. It was quick and easy.

Ran your SVT into speakers? My seismograph registered a small quake - so that was you was it?

T'umbs up again, dawg!

[g1]

g-one, that PDF that you've listed in post # 26 is coming up blank.

Sorry, I forgot that we have to "test" all attachments now to see if they actually work .
Resaved to change file size and reattached so it works now, but it's the same as the 6146b gif that Enzo attached in post #30. The quality of the gif may be better than the pdf version.

[Randall]

Ran your SVT into speakers? My seismograph registered a small quake - so that was you was it?

T'umbs up again, dawg!

Yes, I ran into a 350 watt Eden cab. And no joke, when I turned around there was dust on everything from the rafters in my basement!

[bob p]

That 40 Hz sine wave test signal sounds like it would be brutal on your voice coils.

I once had to deliver a 30 Hz sine wave signal to a JBL woofer as part of the centering procedure for a re-foam/rebuild. I made the mistake of thinking that an old 100W / channel Japanese Kenwood integrated HiFi stereo amp that I used as a bench amp would be adequate for the job -- it wasn't. I didn't' come anywhere near the rated power of the amp, but that low power 30 Hz test signal was enough to pop the secondary / B+ fuses. That makes me wonder about how their power ratings are measured. Probably at 1kHz, definitely not at 30 Hz. There's a lot of energy under the curve of an LF sine wave.

[bob p]

Thanks for posting the schematics, guys. It's really a shame that nobody seems to have a decent scan of a schematic for an original SVT. You'd think that with a classic amp like this one, somebody somewhere would have an original schematic that could be re-scanned at decent resolution. Enzo, I'll bet that you've got an original in that warehouse of boxes of papers that you have in the back of your shop.

It's interesting that the original amps had a PI calibration procedure. Looking at the fuzzy schematic for the old SVT, I see in the legend that it uses 10% tolerance resistors. Wow. And I bet those were carbon comp resistors that are prone to drift, so you'd really need to rebalance the PI to maximize power output. Now that we live in the day of the cheap 1% MF resistor, that calibration procedure seems like it shouldn't even be necessary, and that PI output matching should be a thing of the past.

I'm looking at my 2-Pro, and it doesn't even have the variable tail resistor to ground that was used on the original SVT for PI balancing. For the past 20-30 years it seems that the use of tighter tolerance parts has eliminated the need for PI balancing altogether.

[Leo_Gnardo]

I'll often replace an old possibly drifted-value current sensing R with a new Dale 3W WW 1%. Just did one in a MusicMan 65W the other day. The 10 ohm CC measured 11.6 so out the door it went. In your newer SVT-CL's & similar, there's a 10 ohm just like my fave Dales on the cathode of each output tube, and a comparator for each, you know the drill I'm sure, to make those dummy lights glow. A new complicated way to do something simple. Recent comments & commiserations about wear & tear & mismatch of driver 12AU7's that can't be overcome easily by hi tolerance passive parts. Would be good to have a balance option but that's where Ampeg's left us, and pages and pages of hard-to-understand "treasure map" schematics for their newer products.

Still squinting at the mini schematix in my "Tube Amp Manual" from GT. Where's the magnifying glass?

[bob p]

> a new complicated way to do something simple.

I have to admit, I'm all in favor of the idiot light biasing system. Sure, a lot of people lament that it's for idiots, and the claim that the only "right" way to do things is to meter each tube individually, but that presents a major logistical problem on the newer amp designs. Maybe not a problem with the SVT-CL type amps, if they follow a more traditional SVT layout, but a real problem with the rack-style SVT-2 Pro.

That 2 Pro amp doesn't have the preamp board mounted high in the box like an old SVT. Instead it puts an EQ up there, and moves the preamp board down to the bottom. The bottom of hte amp has got a bunch of boards crammed in there, with hte power amp boards being stacked in the back. Only the solder side of the PCB is visible and there are no test points. Aargh. You have to disassemble the amp to service it, and reassemble the amp to test it. You can't test it while disassembled because of all of the short jumper wires that require the boards to be snugged into place. Metering bias without using the comparators would be an uber-PITB. Learning to love the idiot lights isn't all that hard to do...

[bob p]

> Recent comments & commiserations about wear & tear & mismatch of driver
> 12AU7's that can't be overcome easily by hi tolerance passive parts

My experience hasn't been that there's a mismatch problem, though I am only dealing with a sample size of 2 units. My experience has been that the problem is one of wear on the drivers, where both triodes suffer reduced output that causes the amp to lose power. That's an emission problem, not a balance problem, so I'm thinking that the answer is in replacing tubes, not in tweaking the values of surrounding passive components to effect rebalancing.

After seeing those original SVT schematics and finally taking a few minutes to look at them, I'm thinking that a conversion back to 12DW7 and 12BH7 is going to be the answer for my later design amps. I have a good supply of NOS RCA tubes and very minimal parts substitutions would be required to make the change.

[Leo_Gnardo]

> a new complicated way to do something simple.

I have to admit, I'm all in favor of the idiot light biasing system. Sure, a lot of people lament that it's for idiots, and the claim that the only "right" way to do things is to meter each tube individually, but that presents a major logistical problem on the newer amp designs.

It's taken me a long time to get used to - and would still rather have test points but I can see why Ampeg has done this - to try and prevent people who don't have meters & knowledge from tweaking their amps to death.

Having gotten used to the green - red - none light system at least I realize when I can't get a LED to light at all, or goes from none to red, there's something that needs replacing. Usually a 10 ohm sense resistor has opened because of a shorted output tube, or a 220 ohm screen grid resistor, same reason. Yes, a giant PIA to access the board, but now I'm used to it, so in the barrel I go for another turn. And keep a supply of those R's around for the next time I need some.

[bob p]

one thing that i actually like about the new bias system is that you don't have to follow the "rules" for biasing to make green "just turn on." i find that if you do that, you can still have some audible hum in the output. i prefer to bias both sides into the green range and then tweak by minimizing hum from the speaker -- something that you can't really do with a resistive load. i know this is stating the obvious, but since we're talking about putting loads on the amp when biasing i thought it worth mentioning. the SVT can really hum if you don't have good balance between the two sides.

[cyclone]

I bought some 8 ohm 150 watt resistors from weber and mounted them in a stripped car stereo chassis. I put dual banana (pomona plugs on it to switch impedances quickly, made up some jumpers. handles 300 watts at 4 or 16. i have a bigger blown up car stereo chassis I was going to mount 4 for 600 watts but I rarely see that stuff anymore.

[Enzo]

I think the idiot lights are a great thing. ANYONE can bias the amp by watching the lights and following the instructions. No need for e meter or to measure anything, no calculating dissipations, just turn the controls until the lights are right. Even a drummer could do it.

If you are just determined to make it rocket surgery, you can ignore the lights and get out meters and do the whole thing., but they made it so you don;t HAVE to.


Yes, push pull amps cancel power tube stage hum, so if it is unbalanced surely hum increases. That is hum from the bias settings balance. the PI balance is about centering the thing so it clips symmetrical. Thing is, when you are clipping these beats, chances are a little off-symmetry is probably the least of your concerns. In 1970, Ampeg was still striving for a hifi sound, so such detail mattered. Now they are thinking like guitar amps.

[bob p]

"... when you are clipping these beats, chances are a little off-symmetry is probably the least of your concerns. In 1970, Ampeg was still striving for a hifi sound, so such detail mattered. Now they are thinking like guitar amps."

I'm all for power stage biasing & balancing, but I can't get too enthusiastic about chasing the PI balance in an amp like the 2 Pro that's really clean/symmetrical into a dummy load. I think high tolerance parts in the later design have rendered the PI balance problem obsolete.

If someone were driving these beasts in to clipping they shouldn't be worrying about the PI balance -- they should be worrying about their health. Even when clean, these amps are loud enough to crush bone. That makes the problem self-limiting -- crushed bones tend to limit how hard you'll drive an amp.