New find - Sunn Spectrum II

[gui_tarzan]

I just picked up this Spectrum II (the external pic is from a web search - I took it apart before taking pics) and thought I'd have a real mess on my hands when I pulled the chassis out. It's got a bad re-cover with the wrong vinyl and has been sitting in a guy's garage for twenty years according to him. The Sunn logo is missing from the front panel, the jacks and sockets are corroded but the inside of the chassis is spotless. Sparse compared to my Sentura II but absolutely spotless.

There isn't a date written on the chassis but the pots are dated the 34th week of 1969 and the initials "CAD" and an "OK KJ" are written in red pencil.

It's got a very old GEC KT88 and a Sylvania 6550, both Coke-bottle types. A GE 5U4GB, a Sylvania 12AX7A and a 6AN8A round out the tubes. It looks like it has never been touched inside.

Is there a source for the original style Tolex or at least something close?

[cooldude666]

looks like a similar circuit to the 200s and the other 3 knobbers sans the lo/hi boost. SUNN Gave a bunch of the same amps different names to signify which cab they came with. Awesome score!

[gui_tarzan]

Here's what it actually looks like right now on the outside. There is hand writing on the tube chart that says "Spectrum II".

[soundmasterg]

Those big yellow caps are not original....those are for the bias supply...I would check that they are the correct value. They do not even look like electrolytics actually, though they should be. If they are electrolytics, then the positive end should go to ground since it is in the bias supply.

Also, the rectifier tube should be a 5AR4/GZ34 and not a 5U4GB.

One more thing: the can cap should be replaced, especially if the amp sat for a long time unused. With today's wall voltages, with the correct rectifier tube the B+ will probably be around 560V or so...which is too much for that can cap anyway. The replacement can caps are also over-voltaged in this circuit. You can either leave the can in place and disconnect it internally and replace with discreet caps and terminal strips (my choice as it is cheapest and allows for future cap replacement flexibility) or you can leave the can in place and disconnect it internally and replace it with one of the SDS cap boards that Triode Electronics sells. Either way you would end up with a setup that will be able to deal with the higher wall voltages we have today than when the amp was made and will give you long service life.

Greg

[gui_tarzan]

The yellow Aerovox caps have a date code of 6620 on them so I'm thinking they might be original in this amp or at least very close-to-new replacements. They are definitely electrolytics.

My mindset is to replace all electrolytics in amps I have (or purchase) and resistors that are out of spec simply as preventative maintenance and leave the others alone unless they're obviously bad. I have the JJ replacement 40/20/20/20 can but it's about 3/8" larger in diameter than the original and won't fit in the hole. I think I might put it inside the chassis and leave the original in place, unhooked of course. I'm not going to pay almost forty bucks for a CE just to make it look the same. As far as the rectifier goes, it came with the 5U4 but that's not what's going back in it. I have several different rectifier tubes to try (5AR4/GZ34) to see which one I get the most acceptable voltages from.

These must have vibrated something awful back in the day, the transformer bells have several mounting tabs broken off. I'm getting replacement bells from Heyboer for them.

[loudthud]

The broken transformer tabs are from shipping the amp upside down. I have a couple of amps with the yellow caps in the bias supply. They are most likely original.

[gui_tarzan]

I have bell covers coming from Heyboer, should have them by Friday.

[gui_tarzan]

After discovering how hard it is to find 50/150 caps for the bias circuit, I need to understand why Sunn used those values in the bias circuit and if they can be replaced with more common caps. And also why they used 1000uf/35v caps coming off the Red/Blue primary wire on the PT to the V1 anodes and other places. It's easy to find bad parts and replace them, I'm still learning the theory part of the circuits so if it seems like a dumb question, at least I'm attempting to learn.

[J M Fahey]

50 x 150 = 47 x 160
1000 x 35 = 1000 x 50

[g1]

They will have used those caps because they were commonly available at the time. Modern equivalent would be 47uF at 160V or 200V.
As for your other question, looks like you found an error in the schematic . The point marked D is the low voltage supply for the transistors. That is why they have 1000uf 35V caps there, it is only a 16.5V supply. It should be marked "E" instead of "D".
Down to the left of supply "C" is a point marked 260V. It should be marked "D". It runs the anodes of V1 etc.
The points marked "D" which are running transistors should go to the low voltage supply I called "E".
Sorry if I didn't explain that well.

[g1]

Here's a revision with corrections, hopefully I got all the "e" points

[gui_tarzan]

I should have mentioned that, I also have a corrected version but thanks! So what exactly determines the value of the caps in those parts of the circuit?

[Enzo]

Cap values back then had really wide tolerances, so they were given to nice round numbers like 50uf. But even back then there was an official number sequence for parts, and resistors were already using it. 50 was not a standard value in the sequence, but 47 was and is. That is why we always see resistors with values like 33k and 47k instead of 20k and 40k.

Part of the reason for that is the tolerances. If the normal range of a part extends a certain amount higher and lower, then the next higher and lower standard values ought to be spaced far enough apart that the ends of those ranges just overlap. There is little point in having extra values between the main ones if the overlap covers them. When one NEEDS more precision, then one moves to 1% parts or something.


Blah blah, so the 50uf cap of yesterday is the 47uf cap of today. There is no decision of one or the other. 47uf is a common standard value, and absolutely no benefit would come for hunting up a "50uf" cap somewhere. Same thing with voltages. Nowdays 63v is a common standard voltage for a cap. Not back 50 years when this amp came along. You can always use higher voltage caps than called for. But why did they pick 150v? Look at your schematic. The bias supply needs to put -55v more or less on the power tubes. The bias supply starts at the diode from the bias tap of the power tranny. Voltages are not specified, but we can calculate them. The diode charges the lowr 50uf cap directly. Across that cap are three resistors, two 10k resistors and the 10k pot. That -55v comes out of more or less the center of that group, which means the voltage across all of them ought to be roughly twice the bias voltage, or 110v. So if I have 100-110v, I want a cap with a little headroom, so 150v there seems ideal. They could have used a 200v cap, but that costs more. Now the other 50uf part because of the action of those resistors can never see more than 2/3 of that 110v. Now matbe they could have saved a little with 80v caps there, but it also costs money to stock two different caps, so just using another of the 150v ones means just one type cap to stock there.


There is absolutely nothing critical about this circuit, it is just a bias supply. ALmost no current is drawn from it other than the voltage divider itself, about 3-4ma. But it is half wave, so they needed enough capacitance to smooth that out, and 50uf was enough. I, as do many others, usually install 100uf caps on bias supplies, partly because I have a large supply of them. Not a lot of decision making ever went into this.


Sounds convincing, huh? Then we turn to the 16v supplies and immediately have to wonder this: If the bias winding makes about 100v on the negative, how then will the same winding make only like 24v when turned the other direction? And I surely do not see it. Or if the 24v is there, how do we get close to 100 on the negatve? But ignoring that large mystery, the same decision went into the 100uf caps. Only a few milliamps are drawn from that low voltage circuit, so we need enough filtration to keep it clean. But also the supply powers the bulb in the trem bug, so ANY ripple on that supply will be injected right into the signal path as hum, so that explains the two stage filter with largish 1000uf caps. COuld they have used 1500uf? Sure. Would 470 be enough? Maybe

This exact same power supply circuit for the two supplies is used in several models, so I am inclined to think we don;t have a typo, but unless I am overlooking something obvious, the question remains for me.

[gui_tarzan]

Thanks for taking the time to explain it in detail, Enzo, that helps me understand it in real, practical terms.

[gui_tarzan]

And here is the upgraded 3-wire power cord and new transformer bells installed. I have a clamp for the new JJ cap can and the bias caps ordered as well.