someone upped this on "another" forum...worth a look!!!!!

[pinkjimiphoton]

One thing to watch is that since all the wires in the originals simply slide on/off the PCBs, it's also very easy to get them replaced onto the wrong slot.

I figured that out after thinking about what *I'd* do, faced with slide-on wires like this. I would put them in a random mixture of right and wrong places, then "enjoy" a pleasant day or two trying to figure what exactly I'd done. 

you're a sick man, r.g.

i like that. :thu:

[pinkjimiphoton]

hmmmm...if all the trop fish are sh**ty, and are being used as blocking caps, i wonder if that's where my hum problem is coming from?

was the schem right with all the values? if so, any suggestions for replacements?

deano, yah bro,  i think that may indeed affect the fuzz...as i remember in mine originally, the fuzz was so damn loud and fuzzy it was practically unusable.

you know...

perfect!!

[R.G.]

hmmmm...if all the trop fish are sh**ty, and are being used as blocking caps, i wonder if that's where my hum problem is coming from?

Given the way that the smaller film caps are used, it's unlikely that they are behind the hum problem. They could well be a cause of odd tone qualities.

Generally hum filtering is done by bigger electrolytic capacitors.

was the schem right with all the values? if so, any suggestions for replacements?

It would be very interesting to see whether the two additional units in captivity match on small cap values.

[pinkjimiphoton]

wellp, i posted the values i had.

but remember, i changed every single electro in the unit, even trippled the one in the first stage of the power supply.

and the hum improved, but didn't disappear.


makes me wonder...in tube amps, a hum that just won't go away can be a leaky coupling cap. wonder if i'd be better off just to take a chance and change 'em all, maybe go with better quality ones? is there a brand or type you'd reccomend, R.G.?

i usually use orange drops in amps, but in this i'd imagine there's probably a better...and cheaper...choice.

[R.G.]

but remember, i changed every single electro in the unit, even trippled the one in the first stage of the power supply.
and the hum improved, but didn't disappear.

Hum comes from many places. DC power supply filter caps are only one of them. The actual path of current down a ground wire can cause hum. So can jacks and controls not making good contact with the metal housing. So can radiation from a transformer.

makes me wonder...in tube amps, a hum that just won't go away can be a leaky coupling cap.

Not exactly. That may be what makes you hear the hum, but the hum starts when the leaky cap lets the tube pull too much current, overloading its part of the power supply, and causing higher ripple. The cause *is* the leaky cap, but it's not the cap letting hum through that you hear.

wonder if i'd be better off just to take a chance and change 'em all, maybe go with better quality ones? is there a brand or type you'd reccomend, R.G.?
i usually use orange drops in amps, but in this i'd imagine there's probably a better...and cheaper...choice.

Before you set off on a difficult quest to replace all the caps, first make sure your grounding is pristine. Take every single wire that says ground on it, and trace it back to the input signal jack. While you're at it, remove all the jacks, clean the bushings and the area where they contact the metal housing. Ideally, put a star washer on the inside of the housing on each jack to bite into the metal housing for good contact. Get five 9V batteries and plug them up in series, then clip the resulting 45Vdc (!) to the console board where the transformer ct and one secondary wire goes in. This gives you at least a short time of running purely on batteries. If it hums with this as power, the power supply wasn't the cause. If there is no hum, then the AC power supply was causing it in one way or another.

I can direct you to film caps that are perfectly good for the 50V/60V environment in the Ludwig, but I hate for you to go to all that trouble and still have it hum.

[Keppy]

FWIW, we didn't find any cap values last week that differed from the schem. I built mine with regular schem values, using plain vanilla 50v film caps (mix of ECQ-B & Topmay). I don't think you need any special caps for this, as none of them get the supply voltage before it's regulated down to 35v. Well, none of the film caps anyway.

I finally get a day off tomorrow, so maybe I can set the trimmers and do gutshots for the clone (I've got the other photos already). And maybe, just maybe a video. Maybe. Wish me luck!

[pinkjimiphoton]

i'll go thru and trace every ground connection then, first. it's entirely possible just from the corrosion on the footboard that that could have a crappy ground. but there's only so many things that can cause that...i mean, a bad cap, transformer, or diode, right? i mean, if it isn't the ground.

[R.G.]

i'll go thru and trace every ground connection then, first. it's entirely possible just from the corrosion on the footboard that that could have a crappy ground. but there's only so many things that can cause that...i mean, a bad cap, transformer, or diode, right? i mean, if it isn't the ground.

Right. Any one of them may cause it. You have to track down and kill them *all* to get low hum. There isn't any short cut. Mother Nature insists that it *all* be correct before She'll give you freedom from hum.

I made a point of the corrosion, bad contacts, and so on because it's something people will never think of until they run into it.

[pinkjimiphoton]

a really good point, too.. i have screwed myself out of a lot of repair money by telling peeps how to fix their amps...often the switching jacks in effect loops are never used by people, and over time the corrosion builds up to the point where contact is lost...boom! dead amp.
i tell peeps to put a patch cord across the send and return, and if the problem goes away, get some plastic-safe contact cleaner, spray it in the holes and stick a plug in 10-12 times.

so far, none of the amps i've diagnosed made it onto my workbench.

thanks bro.

[R.G.]

thanks bro.

I have all these scars to remind me of little lessons that Mother Nature has taught me when I wasn't paying sufficient attention. 

[pinkjimiphoton]

me too!!! 

[Mr Bill]

Well, I started replacing the tropical fish after work, tonight.  Got seven left to do on the lower board and then I will replace the electrolytics.  I've saved the fish and will then sweep them to see how many actually are bad.     

[Mr Bill]

All tropical fish and electrolytics replaced,  It appears I only had trouble with three of the .1uF, the most prevalent of the tropicals cap.  It's important to point out when you use a cap as an interstage coupler/dc block they can breakdown/leak in circuit due to the difference in collector voltage of the previous stage and the base bias voltage on the next.  When I placed an equal value cap across these the low end flattened back out. When measured with the tiny voltage on the cap meter, they looked fine. They also looked fine sweeping a signal without the DC differential being supplied.  That being said, I would hold off changing them out.  I think my failure was unique.

Failure found to date, were the three caps, the bad series-pass transistor in the power supply, both the low Z and Hi Z input transistors were blown.  I may get the transistor designators wrong here because I don't have the schematic right in front of me but I was getting no collector voltage on Q5, Q6 due to Q21 being open.  That being repaired I need to get into the upper deck to find out why I can't get Q21 enabled.  I also found the 2N4401 transistor on the little add-on board was also blown open. Both devices were changed.  All other voltages on the lower deck transistors appear normal.

While replacing the 2N4401 transistor on the add-on circuit, I found a 3rd resistor, a 68K which was wired between the base and collector.  This wasn't wired on on the terminal lugs proper, but through the holes that stake the lugs to the bakelite block.  Does anybody else have 3rd resistor?  This "may" actually be a mod I did to bring up the bias on that transistor.  I remember telling Jimi I couldn't remember if I had built that little circuit, as it turned out I didn't, but that resistor may be the vague memory of modifying it 37 years ago.   

My treadle range is a piddly 0-7.2VDC.  I'm going to readjust the throw once I get the rest of the pedal trouble free (I have 12VDC available on the highside of the pot).

So, tomorrow it's time to attack the upper deck! 

[R.G.]

The 2N4401 is rated for 40V BVceo. I am mildly surprised that they'd use those in a device with a 35V supply. The supply voltage was one reason I recommended the 2N5551 and 2N5401 to Keppy for all the signal transistors. They're rated for 160Vdc, and are not likely to go mad even if the pass transistor fails - which I consider likely.  It's under a lot of power and voltage stress.

We did find that hidden 68K resistor on the add-on amplifer on another unit; good to know that it was a consistent add-on. It seems to work on the clone board.

[Mr Bill]

R.G. - I agree with you on the 40V rating!  Right now I'm trying just to get the signal paths up and running.  I actually am using  a 2N5551 in the series-pass just to get the 35V up so I could continue. That's grossly inadequate with the 12 volt differential I have across it and will be measuring the current tomorrow to see the dissipation, as I attack the top deck.
  One thing that could be interesting here; float a stock, fixed three terminal TO-220 reg in between the 48V/36V to take the brunt of the dissipation . This would also have the advantage of pre-regging the series pass reg! A stock value would just have to accommodate for the voltage drop and still leave a little so that the 36VDC series pass had a little range available to do it's job, which would consist of "laying in a hammock" with the smaller differential across it and could easily go back to TO-92, etc!  Obviously a variable 317 device could be used with the 240 resistor and other calculated value to dial it in...  Just a thought!  Ripple rejection certainly would be enhanced with this simple, cascaded design..I think I'm going to try it.

[Keppy]

FWIW, my clone uses a TIP112 Darlington (TO-220) for Q6 on the console board. That's worked great to regulate my 48v DC supply (which is a regulated switching supply) down to 35v. It's an ECB pinout, though, so it might be an awkward fit. It's still better in my mind than floating an additional regulator feeding a TO-92. I'm not sure if it would perform any more/less capably regulating the AC off the transformer, though. In my unit, the plastic of the TO-220 housing heats up, but not to the point I can't keep my finger on it. I can't keep my finger on the metal back, but it won't be melting any insulation. 

[Keppy]

Crap, that reminds me I was supposed to measure the current draw for R.G. 

[R.G.]

One thing that could be interesting here; float a stock, fixed three terminal TO-220 reg in between the 48V/36V to take the brunt of the dissipation . This would also have the advantage of pre-regging the series pass reg! A stock value would just have to accommodate for the voltage drop and still leave a little so that the 36VDC series pass had a little range available to do it's job, which would consist of "laying in a hammock" with the smaller differential across it and could easily go back to TO-92, etc!  Obviously a variable 317 device could be used with the 240 resistor and other calculated value to dial it in...  Just a thought!  Ripple rejection certainly would be enhanced with this simple, cascaded design..I think I'm going to try it.

Give it a try. The LM317 is one of my favorites for anything not bog-standard, and it's not limited to 40V max like the 78xx series. I thought about just adapting a 317 to the layout I just did for the controls panel, but finally just went with the stock emitter follower. The 317 has no voltage limit as long as you can keep transients from putting more than 40V across it. I've used an LM317 for regulating a tube power supply at 400V. 

I went through several options on what to do with the power before deciding to leave it as it is. Sloth wins again. I thought about the LM317, boosting a three terminal, power zener, and a discrete regulator. All of them seemed to be a lot of effort for not much improvement.

For the benefit of the readers who don't have a cadaver to work on, I put dual pad spaces on the console board, so it takes either a TO-92, a TO-126 or a TO-220. The 220 is really the only answer for any kind of reliabilty if you feed this either a high voltage or a stiff preregulated supply. Even then, you only get about 2W dissipation with no sink.

The thing I actually favor is running it from a 24Vac wall wart, full wave rectified. Most 200-300ma 24Vac wall warts have an open circuit voltage of 27-30V, and rectify to around 37-39V. This puts minimal stress on the emitter follower, since it's being fed just enough voltage to regulate with. It'll stay much cooler.

Another option you might want to consider is putting in a 10V/1W zener where that 100 ohm resistor is before the pass transistor. It'll waste ten of the raw volts and move much of the heating out of the pass device.

Keppy - no biggie on measuring current. It's only necessary for the minimal wall wart kind of supplies. Actually, with ten LEDs (!) on the front panel, all pulling about 5ma, they could easily pull more current than the whole rest of the circuits. On the new console layout, I ran all the LEDs from 48Vdc directly, through three 1/4 W resistors each in series to get the regulated current in the pass device down. It's more reliable to waste power in resistors than in transistors. 

[Mr Bill]

Keppy - I'm going to be in the upper deck today so I can measure the current for R.G. and post it.

[Keppy]

Thanks Bill, but I think we already measured Dino's. It's the clone we're after.