Magnatone Lap Steel Amp
- Thread starter CJ
- Start date
Help Support GroupDIY Audio Forum:
found a few links, it is a 12SL7, s/b able to build a schemo from the web and this thing on the bench, 1947-52, Magnatone Leilani, Dickerson era,
weird rectifier, that 35Z5,
http://dennysguitars.homestead.com/MaggieBlueMOTSAmpBakeliteHandle1.html
http://www.ebay.com/itm/VINTAGE-1950-039-s-MAGNATONE-PEARLOID-MOTS-TUBE-AMP-AMPLIFIER-/151335692132
weird rectifier, that 35Z5,
http://dennysguitars.homestead.com/MaggieBlueMOTSAmpBakeliteHandle1.html
http://www.ebay.com/itm/VINTAGE-1950-039-s-MAGNATONE-PEARLOID-MOTS-TUBE-AMP-AMPLIFIER-/151335692132
Before you "restore", or even TOUCH it power-up, you oughta put in an Isolation Transformer.
We don't want to lose you.
The schematic is essentially a 5-tube radio without the radio bits. The audio may be twin-triode instead of single triode. However many of these used a pentode first-stage.
Should not be hard to trace (after you get over the line-cord going right to audio ground). The key design goal was very-low-cost, which means minimum parts. Would not be surprised if they used 5meg grid-leaks instead of cathode-bias and cathode caps.
We don't want to lose you.
The schematic is essentially a 5-tube radio without the radio bits. The audio may be twin-triode instead of single triode. However many of these used a pentode first-stage.
Should not be hard to trace (after you get over the line-cord going right to audio ground). The key design goal was very-low-cost, which means minimum parts. Would not be surprised if they used 5meg grid-leaks instead of cathode-bias and cathode caps.
> weird rectifier, that 35Z5
Just a half-wave rect.
Specialized for high current at low voltage, lamp-tap, and super low cost.
WIDELY used in home radios.
http://www.nostalgiaair.org/PagesByModel/990/M0016990.pdf
http://www.nostalgiaair.org/PagesByModel/243/M0017243.pdf
http://www.nostalgiaair.org/PagesByModel/946/M0015946.pdf
http://www.nostalgiaair.org/PagesByModel/948/M0015948.pdf
That last one uses the similar 35W4, but the schematic is clearer.
25Z4, 1935, is the prototype, scaled for 0.3A heater power.
35Z4, 1939, is the 0.15A version. Was VERY popular 1946-1950, then fades as the Miniature tubes took over.
Just a half-wave rect.
Specialized for high current at low voltage, lamp-tap, and super low cost.
WIDELY used in home radios.
http://www.nostalgiaair.org/PagesByModel/990/M0016990.pdf
http://www.nostalgiaair.org/PagesByModel/243/M0017243.pdf
http://www.nostalgiaair.org/PagesByModel/946/M0015946.pdf
http://www.nostalgiaair.org/PagesByModel/948/M0015948.pdf
That last one uses the similar 35W4, but the schematic is clearer.
25Z4, 1935, is the prototype, scaled for 0.3A heater power.
35Z4, 1939, is the 0.15A version. Was VERY popular 1946-1950, then fades as the Miniature tubes took over.
CJ said:
Hmmm, there is a Leilani amp in pink MOTS at the studio. I haven't been inside it yet since it was bought in working order. If you need chassis pix for reference let me know via PM.
![Electronics Soldering Iron Kit, [Upgraded] Soldering Iron 110V 90W LCD Digital Portable Soldering Kit 180-480℃(356-896℉), Welding Tool with ON/OFF Switch, Auto-sleep, Thermostatic Design](https://m.media-amazon.com/images/I/41gRDnlyfJS._SL500_.jpg)
wow, thought it was mis-wired but sho nuff, audio ground is one side of the AC line! 
now i know why they call the .005 Sangamo the Death Cap, if it shorts, then the input jack grounds are now on the power line which could turn into a Spiritual Awakening, ;D
getting it sorted, almost done with the schemo,
now i know why they call the .005 Sangamo the Death Cap, if it shorts, then the input jack grounds are now on the power line which could turn into a Spiritual Awakening, ;D
getting it sorted, almost done with the schemo,
ok here is the schematic, who ever reworked this thing put a cap from the input grid of the 12SL7 to cathode #2, i think thy were trying to convert the amp to grid leak, but the input resistors are too small at 50 K and he missed the correct ground point,
so the only resistor value i could not get from online pics is the input resistor from grid to ground, but 500K seems ok,
also, the cathode bypass cap on cathode #2 (Magnavox) goes to chassis ground, there is an online pic (see links above) that has the fiber insulator plate on it which suggests that it should go to the AC line, but this cap has the original rivets holding the cap down with a metal plate so maybe there are some different versions of this cap wiring out there,
this is a dangerous amp, no fuse, and the potential to have the git player hooked to 120 AC if the wires or caps short out,
Henries are very low on the puny OPT, so probably not much bass response, however, this was designed for lap steel, so...
new method of using tube base diagrams to combine circuit and chassis wiring into one confusing package,
so the only resistor value i could not get from online pics is the input resistor from grid to ground, but 500K seems ok,
also, the cathode bypass cap on cathode #2 (Magnavox) goes to chassis ground, there is an online pic (see links above) that has the fiber insulator plate on it which suggests that it should go to the AC line, but this cap has the original rivets holding the cap down with a metal plate so maybe there are some different versions of this cap wiring out there,
this is a dangerous amp, no fuse, and the potential to have the git player hooked to 120 AC if the wires or caps short out,
Henries are very low on the puny OPT, so probably not much bass response, however, this was designed for lap steel, so...
new method of using tube base diagrams to combine circuit and chassis wiring into one confusing package,
Attachments
looks like some negative feedback at 60 CPS being fed into the second stage cathode to cancel hum via the bypass cap, pretty strange,
also looks like a strange form of grid leak with the cap in the ground leg instead of the grid, pretty weird,
going to rework the OPT, looks like a Pri-Sec wind with 1752T : 65 T turns ratio,
also looks like a strange form of grid leak with the cap in the ground leg instead of the grid, pretty weird,
going to rework the OPT, looks like a Pri-Sec wind with 1752T : 65 T turns ratio,
Attachments
> circuit and chassis wiring into one confusing package
For this class of death-trap, we usually had separate symbols for "circuit ground" and "chassis ground". And usually an R-C network between them, to supply some ground-reference for audio with "safe" (stings!!) leakage of wall power.
However some early designs did not separate the two grounds; hard to do so for guitar amp.
> Henries are very low on the puny OPT,
Remember that 50L6 (no relation to 6L6) is a low voltage high current tube. So the optimum load is low, perhaps 2K. And 4H*82Hz*6.28= 2,060 Ohms. So it covers guitar. Really more-or-less depending on actual speaker impedance in bass. But it isn't lame, just not generous.
You may also find the DCRs are higher than a big-buck design would pick.
Your schematic seems reasonable. Could be clearer, but that would be obsessive.
I really wish you would buy (or wind) a power isolation transformer for it.
With safety under control, this could be a fine sweet amp within its ~~ 1 Watt limit. And a fine nasty amp pushed past the Watt.
For this class of death-trap, we usually had separate symbols for "circuit ground" and "chassis ground". And usually an R-C network between them, to supply some ground-reference for audio with "safe" (stings!!) leakage of wall power.
However some early designs did not separate the two grounds; hard to do so for guitar amp.
> Henries are very low on the puny OPT,
Remember that 50L6 (no relation to 6L6) is a low voltage high current tube. So the optimum load is low, perhaps 2K. And 4H*82Hz*6.28= 2,060 Ohms. So it covers guitar. Really more-or-less depending on actual speaker impedance in bass. But it isn't lame, just not generous.
You may also find the DCRs are higher than a big-buck design would pick.
Your schematic seems reasonable. Could be clearer, but that would be obsessive.
I really wish you would buy (or wind) a power isolation transformer for it.
With safety under control, this could be a fine sweet amp within its ~~ 1 Watt limit. And a fine nasty amp pushed past the Watt.
Here's a cardboard Kent amp, a decade later, mini tubes, but much the same.
Note the entire first stage has been "floated" out at chassis ground instead of circuit ground, returning through the 100K+0.05u shock-network. To make this work, the first stage runs obscenely low current. This trick may not have been around in 1950.
Note the entire first stage has been "floated" out at chassis ground instead of circuit ground, returning through the 100K+0.05u shock-network. To make this work, the first stage runs obscenely low current. This trick may not have been around in 1950.
Attachments
> freq plot on OPT, good enuff for a lap steel student i guess,
I guess. 0.11V midrange, times -3dB, is 0.077V. It looks to me like 0.077V happens at 9Hz. That would be a BIG lap!
I bet you are driving it constant-voltage. The power tube is more like constant current.
I suspect the speaker should be 4 Ohms. Even then the 726:1 Z-ratio gives 2.9K which is high for 50L6 at ~~110V on the plate.
So drive the primary through a 30K resistor. Load secondary with 4 Ohm resistor, or with the actual speaker.
I'm guessing numbers like -3dB at 118Hz for resistor load. Which makes complete sense: 60hz and 120Hz buzz will be large with the half-wave rectifier, an Eight can't be pushed hard below 150hz-120Hz without strain, so a ~~120Hz droop is all to the good.
The actual response with speaker will echo the speaker impedance curve. The low inductance may even fight the speaker's bass resonance, less slap.
Your "3.8W" figure is not appropriate. That takes 200V supply. On the face of it, a 117V wall-outlet peak-rectifies to 165V DC. However a low-price half-wave hollow-state rectifier like 35Z5 will be significantly less. And apparently less than the 35Z5 data-sheet shows, because they add a series resistor (often part of the 35Z5 heater) to limit the start-up surge and rectifier-abuse (no PT to soften the blow). The "2.1" Watt 2K condition is more likely.
I guess. 0.11V midrange, times -3dB, is 0.077V. It looks to me like 0.077V happens at 9Hz. That would be a BIG lap!
I bet you are driving it constant-voltage. The power tube is more like constant current.
I suspect the speaker should be 4 Ohms. Even then the 726:1 Z-ratio gives 2.9K which is high for 50L6 at ~~110V on the plate.
So drive the primary through a 30K resistor. Load secondary with 4 Ohm resistor, or with the actual speaker.
I'm guessing numbers like -3dB at 118Hz for resistor load. Which makes complete sense: 60hz and 120Hz buzz will be large with the half-wave rectifier, an Eight can't be pushed hard below 150hz-120Hz without strain, so a ~~120Hz droop is all to the good.
The actual response with speaker will echo the speaker impedance curve. The low inductance may even fight the speaker's bass resonance, less slap.
Your "3.8W" figure is not appropriate. That takes 200V supply. On the face of it, a 117V wall-outlet peak-rectifies to 165V DC. However a low-price half-wave hollow-state rectifier like 35Z5 will be significantly less. And apparently less than the 35Z5 data-sheet shows, because they add a series resistor (often part of the 35Z5 heater) to limit the start-up surge and rectifier-abuse (no PT to soften the blow). The "2.1" Watt 2K condition is more likely.
oh boy,
" Many early examples of the ‘All-American Five’ posed a shock hazard to their owners. Lacking a mains transformer, the chassis of the AA5 radio was directly connected to one side of the mains electric supply. The hazard was made worse because the on/off switch was often in the wire of the mains supply which was connected to the chassis, meaning that the chassis could be “hot” when the set was either ‘on’ or ‘off’ – depending on which way the plug was inserted in the power outlet. The metal chassis securing screws were often accessible from the outside of the Bakelite or wood case, and there were many examples of owners receiving a shock by making contact with these screws while handling a set.
After public outcry and several sensational press accounts[citation needed] of the problem, the hazard was eliminated from later sets by the use of an internal ground bus connected to the chassis by an isolation network. Underwriters Laboratories required the adoption of the floating chassis, as isolation from the mains (the exact circuit and component values were not specified although the leakage current allowed was specified) to limit the shock to a “safe” current level. The chassis was maintained at RF ground (for shielding) by a bypass capacitor (typically 0.05 µF to 0.2 µF) usually with a resistor connected across it (typically 220 kΩ to 470 kΩ , although values as small as 22 kΩ were sometimes used or the resistor was simply omitted). See here for a typical schematic of a 1948 model AC/DC radio with a 220K isolation resistor.
Over the years, these paper capacitors often become leaky, and may allow sufficient current flow to give the user a shock."
Playing barefoot out of the shower then is a no go? ;D
got a 12SL7 on the way for 6 bucks plus ship, will re-post schemo with voltage values and reduced wattage claims,
ok, finally found a very similar circuit diagram, just sub in a 12SL7 for the 12AX7 and you have it, note resistor across .05 uf iso cap , so if you sing into a mic you gonna feel something like steel wool on yo lips,
UL listed or approved?
note lack of grid resistor on first stage, and a .01 cap across the OPT,
" Many early examples of the ‘All-American Five’ posed a shock hazard to their owners. Lacking a mains transformer, the chassis of the AA5 radio was directly connected to one side of the mains electric supply. The hazard was made worse because the on/off switch was often in the wire of the mains supply which was connected to the chassis, meaning that the chassis could be “hot” when the set was either ‘on’ or ‘off’ – depending on which way the plug was inserted in the power outlet. The metal chassis securing screws were often accessible from the outside of the Bakelite or wood case, and there were many examples of owners receiving a shock by making contact with these screws while handling a set.
After public outcry and several sensational press accounts[citation needed] of the problem, the hazard was eliminated from later sets by the use of an internal ground bus connected to the chassis by an isolation network. Underwriters Laboratories required the adoption of the floating chassis, as isolation from the mains (the exact circuit and component values were not specified although the leakage current allowed was specified) to limit the shock to a “safe” current level. The chassis was maintained at RF ground (for shielding) by a bypass capacitor (typically 0.05 µF to 0.2 µF) usually with a resistor connected across it (typically 220 kΩ to 470 kΩ , although values as small as 22 kΩ were sometimes used or the resistor was simply omitted). See here for a typical schematic of a 1948 model AC/DC radio with a 220K isolation resistor.
Over the years, these paper capacitors often become leaky, and may allow sufficient current flow to give the user a shock."
Playing barefoot out of the shower then is a no go? ;D
got a 12SL7 on the way for 6 bucks plus ship, will re-post schemo with voltage values and reduced wattage claims,
ok, finally found a very similar circuit diagram, just sub in a 12SL7 for the 12AX7 and you have it, note resistor across .05 uf iso cap , so if you sing into a mic you gonna feel something like steel wool on yo lips,
UL listed or approved?
note lack of grid resistor on first stage, and a .01 cap across the OPT,
Attachments
schematic redrawn and voltages included.
tired a bunch of different caps and found a 0.0022 sounds better than a .005 from 12SL& to the 50L6 grid,
there was a 0.01 uf tone cap across the first cathode which was removed as it rolled off too much high end,
sounds great with the new tube,
tired a bunch of different caps and found a 0.0022 sounds better than a .005 from 12SL& to the 50L6 grid,
there was a 0.01 uf tone cap across the first cathode which was removed as it rolled off too much high end,
sounds great with the new tube,
Attachments
You found the Wikipedia article.
It has *multiple* "issues". Clearly some of these editors have never dropped the radio in the bathtub.
But yes. The radio had safety issues and the guitar amplifier of the same technology has REAL safety issues. You can use a radio and never touch metal. There's no simple way to use a guitar amplifier without sweaty hands *directly on* (or barely isolated) circuit common.
Which why I will say again: ISOLATION POWER TRANSFORMER.
For lurkers: Allied Transformer Isolation, Chassis Mt, 115V, 35VA, Out: 115V, Part#: 6K112HF
For CJ only: DIY it. 30+VA core. Probably already has a 120V winding. Strip the secondary, insulate good, calc the volts/turn, wind a new secondary. Aim for 1:1 ratio; you'll have some sag but also wall-voltages have gone up since 1960, so 1:1 off a modern wall, sagged, will be near a 1960 wall.
It has *multiple* "issues". Clearly some of these editors have never dropped the radio in the bathtub.
But yes. The radio had safety issues and the guitar amplifier of the same technology has REAL safety issues. You can use a radio and never touch metal. There's no simple way to use a guitar amplifier without sweaty hands *directly on* (or barely isolated) circuit common.
Which why I will say again: ISOLATION POWER TRANSFORMER.
For lurkers: Allied Transformer Isolation, Chassis Mt, 115V, 35VA, Out: 115V, Part#: 6K112HF
For CJ only: DIY it. 30+VA core. Probably already has a 120V winding. Strip the secondary, insulate good, calc the volts/turn, wind a new secondary. Aim for 1:1 ratio; you'll have some sag but also wall-voltages have gone up since 1960, so 1:1 off a modern wall, sagged, will be near a 1960 wall.
> lack of grid resistor on first stage, and a .01 cap across the OPT
There's always a DC path through the guitar, right? (This is no longer true, guitarists "need" pedals, and not all pedals have DC bleed.)
A naked pentode into a common (low buck) speaker will have a rising high because of voice coil inductance. To a point, this is "good". But in AM radio, anything over 3KC-5KC is just hiss. Radios used a 0.05u cap to form a 2-pole roll-off. This rig lets a few more highs through, because it truly is a low-hiss circuit.
There's always a DC path through the guitar, right? (This is no longer true, guitarists "need" pedals, and not all pedals have DC bleed.)
A naked pentode into a common (low buck) speaker will have a rising high because of voice coil inductance. To a point, this is "good". But in AM radio, anything over 3KC-5KC is just hiss. Radios used a 0.05u cap to form a 2-pole roll-off. This rig lets a few more highs through, because it truly is a low-hiss circuit.
