Carvin X100 Schematic Verification and Ultra Hot Cement Power Resistors

[syscokid]

This old 1980 to 1982 amp is not mine, but it is on my bench right now, and I noticed that this 47k/2w resistor on the schematic is actually 470k in the amp (red circle #1). The amp already was modified with a filter cap replacement that looks a bit ugly, and was wondering if the previous "tech" had replaced the resistor with the wrong value? Or maybe the schematic has an error.

Carvin_x100a1981_POWER.jpg

Also, the two 200 ohm/5w cement resistors (red circle #3) get extremely hot and only have about 1/16" clearance between the resistor and circuit board. The board is very darkly discolored around those resistors. I was thinking of replacing them with new ones, but elevating them about 1/2" above the board. I assume that I should not increase their wattage rating. Yes? No?

RenderedImage.jpg

71426520023__28699CFE-750E-4E84-8A0F-65B23B53456A.JPEG

And just for the Hell of it, schem for the preamp section:

Carvin_x100a1981_PREAMP.jpg

[maxkracht]

470k should be fine. Unless I am misunderstanding something, it's just there to discharge the filter caps when you turn the amp off. Larger value = slower discharge.

Raising the 200 ohm resistors off the board is a good idea, bigger might be better if you find something that fits, but it has worked so fair. Air circulation will help. Also good to raise the 470 ohm resistors off of the board. You can put a kink in the resistor leads or balance them on some heat shrink tube or something so there is some strain relief on the solder joint with the elevated components.

Make sure the board hasn't started to become conductive where it is dark. Probably wise to reflow all of the solder joints around the power tubes. Once that board gets too dark you probably need a new board...

[xtian]

^ +1 agree.

[R.G.]

Maybe I'm missing something too. I question whether the actual value of that resistor should ever be 47K.
If I'm reading the schemo right, it shows 450V for the B+ to the OT. The screens get 450V through a 350R resistor to point B. Then this connects to point C through a 470K resistor paralleled by a 10K and a "60% Power" switch. The resistor in question then goes to ground. Point C is the plate supply to the phase inverter.
The use of a 47K means that if the 10K resistor is switched in, the plate supply to the phase inverter would be the point B voltage dropped by 47K/57K, or down to 82% of the screen voltage. OK so far. But when the switch is opened, the PI plate voltage would drop to 47K/517K, or 9% of the screen voltage. That seems way low, even if the screen voltage was, say, 400V or more.
Sure, 12AX7s can work at 40V on their plates, but it still strikes me as drastic. 470K for the resistor in question seems more realistic.

I've dealt with Carvin before, and sometimes there are "mistakes" in the schematics. I'm a suspicious sort, and I think that the "mistakes" can be there to prevent copying sometimes. Just suspicion.

On the 200R resistors: I have fixed Carvins before with a similar setup for providing the +/- 14V supply for the opamps. Sure, it can work, but it's a lot of dissipation for those resistors to carry away as heat, and it calls on the zener diode "regulators" to eat a lot of power too. I'm philosophically opposed to this approach. A couple of 7815/7915 regulators could do the job with less waste heat and more reliability. In zener regulated supplies, the resistors have to let through the maximum current the load will ever need, worst case. The zeners have to "eat" all the current that the loads don't use from that maximum. The excess above actual needs is burned in the resistors and zeners as heat. Series regulators like the 7800/7900 series let through only the needed current, and in general the waste heat is lower. I (think I) know why they did it. They run the zener supply from the bias tap that can produce well over 50-60Vdc on each polarity, and the series devices are only rated to 40V max. Instead of using a series zener before a linear regulator to drop the max input voltage, they just burned the maximum heat in resistors.

But I digress. If you are not up for a re-thinking of the +/- 14Vdc supplies: If you have room, swap in 200R resistors with a higher rating. This wastes the same amount of heat, but the resistor surface temp will drop 20C or so. Cement resistors are happy running at over 150C surface temps, but human skin touching this surface gets instant burns. Do space them up off the board for better cooling of the resistors. You can do something similar by replacing each 200R with two 100R 5W resistors in series if you can mount them securely.
These resistors are in a low voltage section of the PCB, so you don't need to worry much about their voltage rating. They only have maybe 60-70V max across them, probably less. Do check for the PCB being blackened/carbonized under them, but don't sweat "discolored" as opposed to "charred black".

[xtian]

On the 200R resistors:

Same deal on the famous Fender Hot Rod series. That's what keeps us techs in business!

[syscokid]

You can put a kink in the resistor leads or balance them on some heat shrink tube or something so there is some strain relief on the solder joint with the elevated components.

Can anyone please direct me to what “kink in the resistor leads” or “heat shrink tube so there is strain relief” looks like?

[xtian]

like dis:

IMG_2105 Large.jpeg

[syscokid]

I appreciate all the good info here. Thanks to y'all again!

At Mouser, I found some Vishay/Draloric 7 watt/200 ohm resistors that are only 3mm longer than those 5 watt cement resistors.

Also found a vendor who seems to have a bunch of the original Sprague TVLU 2925 multi caps that Carvin used on the early X100's... for $6 each!! Amp needs two of them.

Same amp I'm working on:

IMG_4108.JPG

Capture.JPG

The right-side filter is the Sprague TVLU. It is not connected and sits there like a dummy. Originally the amp had two of these side by side. Previous "tech" disconnected the Sprague on the right side and replaced it with two 40u/500v. I think they are F&T's. The "tech" also completely eliminated the original Sprague cap on the left-side and replaced it with a single blue Vishay/BC 100u/500v cap.

I just want to clean this amp a little bit and present the owner something a little more reliable.

[syscokid]

like dis:

IMG_2105 Large.jpeg

Those are some very cute kinks…

[maxkracht]

Others might disagree with me, but I do not trust old electrolytics, even if they are NOS. They degrade with age, and you don't know what environment they have been stored in. There are new production multi section cans, or you can install radial caps on a terminal strip in the amp. What the previous tech did may have been slightly sloppy, and they should have done a better job with strain relief, but you want new caps in there.

[R.G.]

Max is right. Unless there is an overwhelming reason favoring originality (as in museum-like collections) over actually functioning, replace the electro caps. Electrolytic caps simply do degrade over time. It's integral to how they are made and work. Amps intended to actually function need electrolytic caps replaced with fresh ones at intervals. Amps that simply have to sit there looking original can get by without cap replacement.

[syscokid]

… but you want new caps in there.

… replace the electro caps.

Of course, you’re both correct. My thinking is to make an attempt at forming, or reforming, the NOS caps. If they settle in at the proper parameters, and show no signs of excessive heat through the forming process, then they should be good… Yes? No? And if they are good, it will cost the amp owner only $17 in those parts.

Plan B: If the NOS caps are garbage, I’ll head over to Tubesnmore.com and buy a single CE Manufacturing 40/40/40/40uF 525v multicap at about $50.

IMG_0839.png

[R.G.]

My thinking is to make an attempt at forming, or reforming, the NOS caps. If they settle in at the proper parameters, and show no signs of excessive heat through the forming process, then they should be good… Yes? No?

It is possible to re-form electros to some extent. It is not clear what kind of lifetime you could expect from re-formed caps as opposed to new ones. In the dim reaches of history when electro caps were first used, the electrolyte for forming the oxides and the electrolytes in the finished cap were similar. In today's world, aluminum strips are run through forming tanks from roll to roll. The electrolyte is optimized for forming oxide quickly and with few defects. The electrolyte in the finished caps is optimized for low ESR, and contains very little water, etc., so they're optimized for different stuff. I haven't found any info on how long a re-formed cap is likely to last.

I'd call it a firm maybe. $17 instead of $50 is a good deal, but not if it fails soon after. If I were doing this, I'd offer the owner the (informed) choice. It would also make sense to include the different cost of your labor in each case.

[Phil_S]

IMHO, there is nothing about this amp that shouts out for originality. This looks to me like a clear case of substance over form. Skip the expensive can caps. If you insist on using can caps, JJ makes a 500V 40-20-20-20 for around $20. Leave on 20uf unused and find a 40uf radial cap that works in the other hole. Really, though, you can buy good quality e-caps in axial or radial configuration for much less and then you can figure out how best to mount them, probably with the assistance of a terminal strip. Your customer deserves good quality at lowest reasonable cost. Did he ask to keep it as "original as possible?" If he didn't why are you headed in that direction?

[maxkracht]

What Phil and RG said. If you already purchased the NOS cap, it is perfectly reasonable to put it in place for aesthetics, but I would personally go the radial caps and terminal strip route. Sometimes this allows you to improve the ground scheme as well. Add a dab of silicone under and between each cap for strain relief so the leads don't snap. It is, in my opinion, the least expensive, most reliable, and easiest for the next person to repair once those caps go bad. Most people will care more about reliability than originality and you can get very high quality radial caps for $2-3 each. No one is going to look in the amp except the next tech, and if you do your job well, that won't be for a while.