New find - Sunn Spectrum II

[gui_tarzan]

It's alive!!! Boy does it make noise without a guitar plugged in. With a guitar plugged in it's quiet and sounds really good.

All I had to change initially were the cap can (used the JJ can and mounted it on its side inside the chassis) and the two 50/150 bias caps. So I will drain the caps and start going through all the jacks and see if I have anymore noise. There was a lot of corrosion in the output jacks from sitting in the guy's garage for two decades. The pots will also be replaced and then anything else I find out of sorts will be changed.

A couple of things I'm not familiar with are the molded black caps, are they similar to the Fender blue molded caps? Should I change those as well? Also the Mallory caps with transparent cases, what is the replacement suggestion for those? I attached a picture of those circled with red and blue respectively as well as the caps I replaced circled in brown.

Also I used a 5AR4C and a matched pair of 6550s. It came with a KT88 and a 6550. It seems like the voltage is a touch high (I'll post those readings tomorrow night), would a GZ34 help that (even though they're the same)? My understanding is the 5Y3GT/5U4 isn't a suitable replacement.

Once I get everything worked out it will be for sale, it's really loud and I don't need another really loud amp.

[soundmasterg]

The molded black caps are probably Black Cats, which are paper and polyester caps....they sound good and if they aren't leaking too much DC past themselves, I would leave them. They are generally pretty reliable, but not as reliable as the blue molded Mallory's that come in blackface Fenders. The other caps you mentioned are typically ceramic caps in those values, and they don't often go bad. You can check any of these caps by lifting the low voltage side and putting your meter on the flying lead, then turn the amp on. Your meter will spike at a higher voltage and then settle down into a lower voltage, and this lower voltage is the leakage in DC voltage. A cap should block DC and only allow AC through, but no cap is perfect and there is always some leakage. Having too much leakage can throw off the bias of the following stage and cause a different operating point. Depending on the stage, as little as 0.3V can cause a problem.

The 5AR4/GZ34 are both the same tube, so assuming the one you have is a real one rather than some other tube type masquerading as a 5AR4/GZ34, then the voltages won't change much if at all by changing to a different one. Your B+ should be somewhere around 560V. A 5U4GB could be used, but it isn't the proper tube for the circuit. B+ would be lower than it should be and the amp would also have more sag than it should, and most importantly the bias would have to be readjusted for use with a different rectifier tube type like a 5U4GB. A 5Y3 can not be used in this circuit.

These vintage Sunn amps are very loud amps...with most 2x6550/KT88 amps making 70 watts or more.

Greg

[gui_tarzan]

Thanks Greg, that's good info. This is a new Ruby 5AR4C so I'm pretty sure it's genuine. A lot of the older radios and such I've had for parts had 5Y3GTs in them. Is that because they were single-ended amps?

[soundmasterg]

Could be. Designers back then would use whatever tube rectifier would do the job and met the requirements. The 5Y3 can't handle the current that the GZ34 can handle and it also has a higher voltage drop. It is also a directly heated cathode instead of an indirect one like the GZ34. They use a 5Y3 in a Fender tweed deluxe model 5E3 and in that circuit with those power tubes and current requirements, it performs well. In a Sunn circuit, it can't handle the current requirements and will self destruct.

Greg

[gui_tarzan]

So the lowest bias reading I can get right now is -51v, the schematic says it should be -55v with KT88s. I was thinking it would be in the mid-high -30s or low -40s with 5881s in it, would that be right or will the Sovtek 5881WXTs handle it?

Plate voltage is 530v
Pin 4 = 528v
Pin 5 = -48v

[gui_tarzan]

Based on a comment from another forum I'm going to wait until I get some KT88s and not try to make these 5881s work.

[soundmasterg]

The Sovtek 5881's are very robust tubes and they could take it. They were used in the Sovtek Mig 50H amp with over 500V with no problems. I believe kg used them in his BAGA amp too at very high voltages with no problems, though perhaps in his amp the screens were lower. The proper tube to use would be 6550 or KT88's however. With the higher wall voltages today, the B+ is higher, and as a result you need more negative bias voltage to keep up with tubes that are being driven harder. Are you saying that the max negative voltage that you can get in the bias supply is only -51V? I would check the series resistor and the resistor to ground in the bias circuit too. If they have changed in value, then it could affect the voltage out. I have personally seen a couple Sunns with an incorrect resistor in place of the 18k shown on the schematic, and the bias voltage was off as a result no matter how the pot was adjusted. Knowing what the negative voltage is for the bias doesn't really help you as it doesn't tell you what current the tube is drawing....all it tells you is what the negative voltage is that is being injected into the power tube grids and since every tube is different and will draw different amounts of current at that same negative voltage, you really need to check the current the tube draws to find out what the bias is. You can do this by putting in 1 ohm resistors between the cathode and ground and measuring the DC voltage across the resistor with a DMM. You can also do this with a Weber Bias Rite, which essentially does the same thing with you having to modify the amp. The Bias Rite plugs into the socket and the tube plugs into the Bias Rite. It reads the cathode current, and the plate voltage on a meter. You adjust the bias to a safe range for whatever tube type, and then you take the Bias Rite out and put the tubes back in before you play as if you don't the Bias Rite could blow. You can see them at Weber Speakers - Making the world a bit louder each day..

Greg

[gui_tarzan]

Thanks Greg, I have a bias socket to test those, I just hadn't gotten around to doing that part yet. The plate voltages seemed really high but this is the first Sunn I've worked on. Yeah, -51v is the lowest it would go. I checked all the resistors in the unit since there aren't all that many and they were all either on the money or within 10%. I will check the draw though, I am curious to see what they are.

[g1]

If you do determine that there is no fault in the bias circuit, some people find that the bias winding in some Sunn models does not provide enough bias range. In these cases alternate bias circuits are sometimes used, taking a feed from the HV winding through a resistor, rather than using the bias tap. (see Marshall JTM45 for example)