Asher Guitars WD Music Products Amplified Parts Mod Kits DIY Nordstarnd Pickups Warmoth.com
Asher Guitars WD Music Products Amplified Parts Mod Kits DIY Nordstarnd Pickups Warmoth.com
Asher Guitars WD Music Products Amplified Parts Mod Kits DIY Nordstarnd Pickups Warmoth.com

Valco clone

Discussion in 'Shock Brother's DIY Amps' started by Prophetsnake, Sep 17, 2017.

  1. Prophetsnake

    Prophetsnake Tele-Meister

    134
    Aug 2, 2016
    ireland
    OK. A little difficult to get in there, but here's what it looks like at the moment. The control panel is remote and in any case Ive found the problems there.


    It's playing now, but the the volume is pretty pathetic - not quite as loud as an acoustic. The tone, what there is of it, is outstanding though, and it does get a little gritty if a chord is played aggressively.

    I think this thing is going to be great when I eventually get it going.

    There's hum. It's fairly loud and almost certainly 50 hz. It increases with volume and there's none with the volume all the way down.

    I may have to rethink my heater wiring. At the moment it goes from the transformer to the V4 and then to the remainder of the tubes in the normal way. But from the V4 I have it backtracking to the same hole in the chassis and then about 12 inches of wire to the pilot light.

    I detected no change at all while chop-sticking. If I place my hand near the pre-amp tube, it gets a bit quieter. If I touch it, it gets a little louder again, and the tone of the hum changes. Wiggling the other tubes seems to have little effect



    IMG_1555.JPG
    IMG_8690.JPG
     
    Last edited: Oct 20, 2017

  2. Nickfl

    Nickfl Tele-Holic

    686
    May 24, 2016
    Florida

    I think your lead dress is probably at least part of the issue. Try to make sure your signal leads are seperated from other wires (esp heaters) as much as possible. Keep the heaters away from everything else and try to tuck them back up in the corner of the chassis wherever you can. Also make sure what whenever any leads do cross they do so at a right angle to each other as much as possible. Doing all of the above will probably help with the hum

    Edit: I also see that you had another thread asking about the pin 1 shield on your (presumably) metal shell 6sj7 preamp tube. What did you end up doing with that? It should be grounded, if you left it floating it might be causing some of your noise issues.
     
    Last edited: Oct 21, 2017
    Prophetsnake likes this.

  3. Nickfl

    Nickfl Tele-Holic

    686
    May 24, 2016
    Florida
    As for the volume, are you playing it plugged into the lightbulb limiter or plugged straight into the wall? Also, have you taken voltage readings and calculated your bias?
     
    Prophetsnake likes this.

  4. tubeswell

    tubeswell Friend of Leo's

    Jul 1, 2008
    NZ
    Is that a 470R cathode resistor?

    What are your idle voltages (B+, plates, screens and cathodes) for the 6V6s?

    The stock circuit uses a shared 225R resistor. With a shared 470R, you will be getting double the cathode voltage and substantially reducing the plate-to-cathode voltage (and hence the available gain) of the output tubes.
     

  5. Prophetsnake

    Prophetsnake Tele-Meister

    134
    Aug 2, 2016
    ireland
    Thanks Tubeswell. Yes, that's correct, I did the 470 on advice because of concerns of excessive voltage. I can see that that's not an issue, so I'll swap the original back.

    So far I've only measured the B+ and they're 378 at the Outbut tubes, 126 at the phase inverter and 122 at the pre-amp. I really wasn't expecting that sort of voltage on the far side of the choke, which was inserted between the OT centre tap and the screens.

    Nick, that's right, the shield is ungrounded, but that's easily fixed!

    I'll do all of this and report back in a couple of days. Gotta go make a living first.
     

  6. tubeswell

    tubeswell Friend of Leo's

    Jul 1, 2008
    NZ
    With the original spec cathode resistor, you should find that the increased tube current pulls the plate idle voltage down 20V or so.

    Screen supply decoupling and screen voltage also play an important role in determining output power and
    the quality of the signal when driven hard. The lower the filter corner frequency, the tighter the amp response will be. The lower the screen voltage, the lower the output power - all other things being equal. Also, with lower screen voltage you need to have lower bias voltage all other things being equal, to get the tubes to 'turn-on' effectively.

    The Valco schematic runs a field
    coil speaker, which makes that speaker more efficient in this circuit and filters the screen supply. It also has the screens at nearly the same voltage as the plates. If you aren't running a field coil speaker, you should probably decouple the screen supply, either with a Choke (in a CLC filter) or a resistor (in a CRC Pii filter), otherwise the screen response will be fuzzy and weak with lots of screen current feedback lowering the output tube power. (By comparison, a tweed deluxe 5D3 has a paraphase inverter and runs a 10k screen dropper and puts out about 12W. A 5E3 has a 5k screen supply dropper, and a cathodyne inverter - which is a more evenly balanced inverter, and puts out about 18W).

    Similarly, 126V at the phase inverter supply is probably too low to enable the PI to drive the output tube grids with a good strong signal. You want the PI supply to be about 250-300 to get a decent output signal swing to drive the output tube grids. Substantially decreasing the resistance of the supply resistor between the screen supply and the PI is probably called for if the reduced output tube cathode resistor doesn't go far enough for you. This messes around with the values in the Valco schematic, but if you are not running a field coil speaker exactly like the Valco did, then the amp won't behave like the Valco.
     

  7. Prophetsnake

    Prophetsnake Tele-Meister

    134
    Aug 2, 2016
    ireland
    Thanks Tubeswell. In fact, there is a choke in place of the field coil. It's attached to the OT centre tap and goes to the screen on that schematic. I suppose that's what is dragging the voltage down as far as it is.

    I've also just realised that I made a major error in the build. There's no centre tap on either secondary from the PT, and I have forgotten to include an artificial centre tap!
    Doh.

    I'll be able to sort that when I get home on Monday.
     

  8. tubeswell

    tubeswell Friend of Leo's

    Jul 1, 2008
    NZ
    It would be handy if you could upload an as-built schematic of what we are dealing with here so we can get to the bottom of whatever the problem is.
     
    Nickfl likes this.

  9. Prophetsnake

    Prophetsnake Tele-Meister

    134
    Aug 2, 2016
    ireland
    Sure. I'll do that when I get home. I can't account for what's going on at the mo. I certainly can't account for the huge drop in voltage. The only thing stepping it down is the choke.
     

  10. Prophetsnake

    Prophetsnake Tele-Meister

    134
    Aug 2, 2016
    ireland
    Here's the schematic with all of my mods included. I've added cathode bypass caps to the phase inverter and output stages, used a bridge rectifier, the choke, and of course deleted the field coil. IMG_7600.JPG

    This is the choke : https://www.amplifiedparts.com/products/filter-choke-50-ma-fender-deluxe-vibrolux

    The voltage at each of the three filter caps is 353 - 351 - 89.

    I'm not certain why I am getting such a sharp drop between the second and third filter caps and I'm thinking that that is an issue. Perhaps I have the wrong value resistor there.

    I haven't added the centre taps yet, nor earthed the shield on the 6SJ7. I'll do that a little later and report back.
     

  11. tubeswell

    tubeswell Friend of Leo's

    Jul 1, 2008
    NZ
    Yes 47k will drop a lot of voltage. Compare that to the pre-amp supply resistor values in typical tweed Fender schematics. The voltage drop of 262V is occurring across 47000R; 262/47000 = 5.5mA. This seems to be about what one would expect at this point in the supply, with two triodes and a small signal pentode being supplied through this sized dropping resistance. It you want to raise the pre-amp supply voltage then decrease this resistor to something like 22k or 10k.

    However there is something else wrong with your voltage readings.

    The 1st voltage drop in the supply is 2V, which doesn't stack up across 1500R; (2V/1500R = 1.3mA). The current in the choke should be the sum of the current for the screens and the current through the 47k resistor to the rest of the pre-amp. A typical 6V6 screen at idle is about 2mA, so with both 6V6s and the pre-amp tubes, you should have a drop of 1500R x 9.5mA = 14V. The DC resistance of the choke should be less than that. (Actually 1500R is the open circuit impedance, which is different from DC resistance)

    How many times did you take these VDC readings with your meter? Does your meter have a fresh battery in it? Do you have another meter you can get some readings with just for comparison? (Earlier in the thread you report 378V at the B+ and 122 and 126 in the preamp?)

    Have you actually measured the DC resistance through the choke? Earlier in your thread, you report fairly loud hum at 50Hz. The choke could be shorted. (What is the history of this choke? It is only DC rated for 50mA, so should not be used in any situation where the peak current exceeds 50mA.) If the choke is shorted, this could explain the 50Hz hum. But check the batteries in your meter first.

    Also (and this is just a comment on the drawing) the way the 'rectifier' is drawn on your schematic doesn't make sense. A Full-wave Bridge rectifier for a high tension supply should be wired to the PT's high tension winding (not the rectifier winding) like the one in this link:

    [​IMG]

    (Although you must actually have it wired from the HT winding to get the voltages you are getting)
     
    Last edited: Oct 23, 2017

  12. Prophetsnake

    Prophetsnake Tele-Meister

    134
    Aug 2, 2016
    ireland
    Thanks for that, Tubeswell.

    Yeah, you can ignore the rectifier. Consider it 'representitive' ie, there is one there and it's doing its job ok. My wife was pestering me for her tipex back.

    I just measured the choke and it's showing 156 ohms, which is obviously wrong. So it's either faulty, or I have something wired in parallel. I suppose a short would do that, though, eh?
    I bought the choke new, and to be honest, never tested it, simply installed it. It is where it is to avoid having the output tube plates load it up. Do you reckon a large resistor would be better there?

    I took the readings several times, from several different spots at each juncture to ensure I was getting the correct reading. As a matter of interest, I checked the plate pin at the 6SJ7 and it was reading 4.8 Volts. I checked it several times and it was the same, but I reckon that can't be or the amp wouldn't play at all - it does, albeit quietly from the guitar with the hum rather loud. I'm hoping a lot of the hum will disappear when I install the artificial centre tap resistors. Speaking of which, do you recommend that these be connected to the cathode on one (or both) of the output tubes?
     

  13. tubeswell

    tubeswell Friend of Leo's

    Jul 1, 2008
    NZ
    Well 156R would account for the 2V drop more, which equates to 13mA, which now makes the rest of it look more sensible. However, this would seem to indicate that the 6V6 screens are running quite hot (3.8mA each), which they would be if the screen voltage is only sitting 2V below the plate voltage. In these type of amps where you are in push-pull Class A most of the time, its better to keep the screens about about 50V below the plates. So you could try a 10k resistor in place of the choke, and another 10k resistor in place of the 47k resistor in the preamp (which is typical of an early era (wide panel) tweed deluxe or other wide panel tweed amp running a paraphase inverter and cathode biased output stage). Another option could be 5k to the screens and 22k to the pre-amp (like narrow panel 5E3 tweed deluxe supply rail).

    Also, check that you have a fresh battery in your meter and/or compare readings with another meter.
     
    Last edited: Oct 23, 2017

  14. Prophetsnake

    Prophetsnake Tele-Meister

    134
    Aug 2, 2016
    ireland
    Yes, I'm with you, I think. Having said that, every time I get a notion that I'm beginning to grasp this stuff, I discover I'm more out of the loop than ever I was.
    Looking at it through the lens of your post, and poring over some early Tweed Deluxe diagrams with the same or similar tubes, I can't understand what National was thinking with this design. Unless that speaker field coil drained a lot more than the figures would seem to indicate, I can't see how it would do anything but what it does - and how the voltages to the respective tubes could be anything near what they ought be.

    I think the multimeter is ok, but I'll check the battery. I'm pretty sure my spare meter is fuxored, though.
     

  15. tubeswell

    tubeswell Friend of Leo's

    Jul 1, 2008
    NZ
    Also, if your AC heater supply isn't balanced, the amp will tend to be hummier. So if there is no centre tap for the heater winding, then wire the heater filaments to each end of the 6V winding using twisted pairs and run 2 x 100R resistors (one from each winding end) to ground. Or for even quieter operation, instead of running these 2 x 100R to ground, run them to the 6V6 cathode (cathode end of the shared 6V6 cathode resistor) for a bit of DC heater elevation.
     

  16. Prophetsnake

    Prophetsnake Tele-Meister

    134
    Aug 2, 2016
    ireland
    Yes, I had intended to do that from the get go and just plain forgot. I've built two 5f1 s with bridge rectifiers, but connected them to earth. I think I'll try the 6V6 cathode in this case and see how that works. I'm planning on doing that tonight.
    Meanwhile, I have to get me a few 10K resistors, and I'm not holding out a lot of hope that the local hipster manned electronics superstore is going to have them, no matter what their web site says.

    Out of curiosity, do you have any idea what National were up to with that original design? It seems to me that no matter which way it's sliced, there's no way all of those tubes are going to get anything like an ideal voltage. Valco_ann_1212.jpg
     

  17. tubeswell

    tubeswell Friend of Leo's

    Jul 1, 2008
    NZ
    Would be good to see the overall plate, screen and cathode voltages for all tubes (as well as the B+ and high-tension voltages). I'm curious to see what the 6SJ7 is doing.
     

  18. Prophetsnake

    Prophetsnake Tele-Meister

    134
    Aug 2, 2016
    ireland
    Sure. I took the tubes out to get a baseline, and there is definitely a problem or two, even to my eyes

    Tubes out:
    First power node: 383V
    Second : 382 -so that choke is doing virtually nothing, or I have something wired incorrectly.
    third: 129V

    The V1 socket (6SJ7)
    Pin 1, which has no wires connected to it at all, is showing 75 volts.
    Pin 2 heater 6.5
    Pin 3 Suppressor 4.7
    Pin4 grid one 7.9
    Pin 5 Cathode 3.7
    Pin 6 Grid number two 128
    Pin 7 heater 6.5
    Pin 8 Plate 128


    V2 6SN7 GT
    Pin 1 Grid -section two !39V
    Pin 2 Plate section two 130
    Pin 3 Cathode section two 65
    Pin 4 grid Section one 35
    Pin 5 Plate 127
    Pin 6 Cathode 68
    7+8 are heaters

    V3 6V6

    Pin 3 Plate 382
    Pin 4 Grid 2 382
    Pin 5 grid 1 114

    pin 8 Cathode 0

    V4

    Pin 3 Plate 382
    Pin4 Grid 2 382
    Pin 5 Grid 1 0
    Pin 8 Cathode


    So, something wrong there, surely. One of the grids on V4 is getting DC power and shouldn't be. And pin 1 on V1 is a bit wonky as well. I can get a loan of another Meter tomorrow and double check, but at least this gives me a few points to investigate.

    Fun!
     

  19. tubeswell

    tubeswell Friend of Leo's

    Jul 1, 2008
    NZ
    Pin 1 on the 6SJ7 is the outer metal case. Be careful, you don't want to zap yourself by touching the case. For safety reasons, it would be a good idea to ground Pin 1. The reason that is it probably 'reading' 75V is due to induction.

    There is something fishy going on with your readings. The pre-amp supply node shouldn't be sitting at the same voltage as your plates in your pre-amp tubes. This indicates that the plate resistors aren't in series with the load. Look at this first. Make sure you have the circuit wired up correctly.

    Another thing that is likely to be pulling the pre-amp voltage down a bit is the screen supply voltage divider for the 6SJ7. (The 100k + 22k). A more conventional screen supply would see the screen supplied via a 1M to 2M2 Rg2 (instead of the 100k) with a 0.1uF screen bypass cap going to ground (i.e. no 22k). The way the voltage divider has been designed in the Valco is a crude attempt at a shunt stabiliser (whereby the overall resistance of the voltage divider has to be low enough to ensure that the actual screen current is only a fraction of the shunt current being sunk through the 100k + 22k (so that you don't get much proportional change under signal conditions).

    So, in addition to dumping the 47k supply resistor, I'd change the screen supply for the 6SJ7 if it were me. YMMV.

    If you decide to go that route, take a look at the Fender Champ 5C1 for an alternative circuit. (The 6SJ7 in the 5C1 is grid-leak biased, but you could do a cathode biased version with a smaller (say 470k) grid leak resistor and a cathode resistor.) You want the cathode voltage to be around 1V to 2V to emulate the bias voltage in the 5C1 6SJ7 stage. So size the cathode resistor to suit. (I suspect the 4k7 cathode resistance in the Valco is designed to get 1V or so with the very low current that the 6SJ7 in the Valco is seeing, but this will probably result in an unsuitably cold bias if you rearrange the 6SJ7 for higher tube current). Note: The reason you are probably seeing 3.7V on the 6SJ7 cathode at present is more to do with the way you might have the plate end of the circuit mis-wired (see my initial comment in this post)
     
    Last edited: Oct 23, 2017

  20. Prophetsnake

    Prophetsnake Tele-Meister

    134
    Aug 2, 2016
    ireland
    A bit of progress.

    I found a wiring error on the output tubes and corrected it. I grounded the 6SJ7 shield and installed the artificial centre tap using the cathode of V4. The amp is much much quieter now, though there is still some hum, which I haven't yet investigated.

    It's still not producing a lot of noise. I had a quick look around at the voltages and they are ridiculously low on the V1 and V2 tubes. Plate voltage of 4.8 on V1, for instance. I'm amazed that it's making any sound at all.

    I was getting a bit tired, so I'll do all of the readings tomorrow.

    I do have a few resistors of sufficient wattage to split the voltage, but not what you recommended, but I think I might try a 4,7K in place of the choke, then a 11.8K in place of the 47K that's in it now and see how that goes.
    Also, the cathode bypass resistor shared by the 6V6 tubes; do you still think it's a good idea to go back to the original 225 ohms?

    I had a look at the 5C1 circuit and I can see how it does the job in a much simpler way. Tell me this, though; if I were to go that route, should I include that entire Fender circuit? IE the input coupling cap and grid resistor, or leave the Valco input as is?


    It seems impossible that National could have sold an amp that performed like this. The only thing I can imagine that can be at the heart of the problem is that my transformer - 275 V secondary, has a much lower voltage than the original. The other major alteration from the original is the speaker and OT transformer, but since they're both designed with a pair of 6V6 in mind, I doubt that very much.
     

IMPORTANT: Treat everyone here with respect, no matter how difficult!
No sex, drug, political, religion or hate discussion permitted here.