Old PA - Lorenz SMV4/8W

[Richardgr]

Firstly, if anyone has a schematic for this, I'd be very grateful for a copy! I have just about reverse engineered it, but some component values are missing, and the original schema would be very valuable.

It is a nice single ended amplifier from the late 30's. I have decided to update it as sensitively as possible (i.e. no chassis work) and provide it with a telepropp input so it can be used as a guitar amp or microphone amplifier (which is close to its original use).

The tubes are a pair of EF12s, one being dedicated to the gramophone input, and an EL11 for output.

Power supply is quite refined! CLC with AZ11, and decoupled supplies for each EF12.

All tubes have the TUS-Y8 'German steel' sockets.


My first question is whether I should try and save the resistors. They are the red ceramic style, and are used to create solder pads on the paxolin board. I doubt if I will be able to salvage any passive components, but interested what others think!

[wd40addict]

Surely you can measure the resistors and only change if they're way out of tolerance, don't change them for the sake of it!

Problems are much more likely to lie with leaky capacitors, but again you should not automatically assume they're faulty without testing.

BTW the label says Lorenz, but the amp appears to have been built by Siemens?

[Richardgr]

Thanks for the reply. Yep, it has a Siemens plate on the rear, but a Lorenz logo on the front.


Yes, I should measure a few resistors and see if they have drifted much, before I decide on their fate. I plan to reuse the paxolin board, and use the same construction style with the component leads doubling up as tags.

I'm thinking I will retain the tube sockets - I'm sure contact cleaner will fix those.

I'd like to keep the old toggle switches - but could this be a safety issue?

The AZ11 and EL11 are a little loose in their bases - any tips for fixing them in place? I have seen acetone mentioned as a solution, but dies that really work? Otherwise I was thinking a little loctite juditiously applied around the rim of the base.

There are 3 or 4 5Meg resistors, and some odd paralleling and serial compononents that I do not full understand. Hopefully I will have a schematic drawn soon.

[wd40addict]

https://www.vintage-radio.net/forum/...5&postcount=21

[Richardgr]

Thanks for the tip on fixing the tube bases. I'll give it a whirl.

The paxolin appears to have had some arcing on one point, so I don't think it is reusable.

The resistors were surprisingly accurate - like just 2% off. The capacitors were way off, and a couple are missing their text, so I will have to try and work out what they were.

I think it is a bit later than I originally thought - probably just after the war. The metal work is quite flimsy, so smacks of post war austerity, and the tubes tally with some models being built in '48.

[Techman]

I can't see any evidence of arcing. perhaps staining by flux or damp.

I wouldn't go mad with this amp, just work out what the critical capacitors are and replace them, then power it up and see what happens. It might be an idea to try a bit of reforming of the smoothing capacitors before applying power, but if you use a lamp limiter and do the power up in stages and feel for them getting warm and monitor the HT while you're doing it, then there shouldn't be a problem. Remember that this is just an old PA amp, so was built for quantity, not quality audio wise!

PS - do some static checks first to check for shorts or open circuits etc.

[trobbins]

Has any insulation on wires gone rigid or crumbly?

This type of restoration is helped if you have an insulation resistance megohm meter and can test circuitry at up to 1kVdc to confirm no poor insulation in transformers and around the circuitry, and across parts and to ground on the board, and across any switch to case. Certainly if any caps are confirmed to be way off in capacitance then change, and otherwise I'd suggest using a variac to bring up voltage (without valves) to confirm all coupling caps aren't leaking and filter caps are ok up to their rated voltage. Then if you add in one valve at a time, and bring up voltages with a variac, then check levels, then that is an easy way to provide confidence that all is as it should be. But many don't have a variac so the light bulb limiter is the typical path taken, although for older amps that is becoming more and more a leap of faith type test. A middle ground is to energise the power transformer primary with say a 24VAC feed.

Those resistors certainly look at least late 1940's and appear to have a nice tropical coating, and are large, so its good that most show no change in resistance.

[Richardgr]

Hi trobbins, the insulation was generally OK - it was an insulated sleeving. However there had been previous surgery, so a bit of PVC in their too. From the short length of mains lead with blue and black wires, pre-1970.

I have eviscerated it, so now I need to decide where I will go with this thing.

The tube bases look OK, along with the choke and OPT. The power transformer looks like it is not original, since only two bolt holes on the base matched the holes in the top plate. It is impregnated with tar, so it is ugly but serviceable. I'll test it soon.

From a previous post, the paxolin could be re-usable. Since the resistors tested quite well, I could start with that for testing, however one of the end caps of a resistor popped off, so possibly this is a false economy.

I have tried to reverse engineer the schematic, but I am confused around the tone control. Don't understand those two 5M resistors.

Also there is a 1000pF capacitor from one of the filament wires to earth - is that some sort of hum defeating measure?

And the 1800pF capacitor from output tube anode to earth is confusing - presumably to stabilise the circuit in some way?

I don't have any voltages yet - I'll test the power transformer with the variac to see what it is producing.

Appreciate any comments regarding obvious nonsense from the schematic!

[trobbins]

The 1800pF cap was likely for tapering off treble, and constraining voltage transients across the winding. C1 at 50nF doesn't sound right, as it appears to be local feedback (ie. it could be 5pF or 50pF). There is no schematic indication of output transformer speaker connections and if there was feedback. The source switch would cause some popping, as the coupling cap to the input triode mode stage is not leak grounded.

Did you confirm there was no heater winding direct connection to ground (ie. a CT)? It is possible that the heater winding didn't have a direct connection to ground, and the the heaters float to some DC voltage between the limits of the cathode voltages of the three valves (due to heater/cathode leakage), and the 1000pF capacitor provides a DC filter for that voltage - for hum suppression - that was not a common technique but has been used by some manufacturers.

[Richardgr]

Thanks for the pointers! There was no feedback from the OPT secondary.

Now it is gutted, I’m wondering what my best course of action would be. I do have a nice pair of UCL11, and some Y8 sockets, so considering a bijou SE amp. The case has a nice patina and the power supply should be salvageable. I’m considering triode connected EF12 per channel, and the UCL11, then a couple of small ECL86 SE OPTs. Rectifier would be silicon, to keep the tube count to 4.

If I replicate the circuit, then it is probably a bit of a waste of nice tubes for a non-descript guitar amp.

[trobbins]

One of the advantages of restoring the amp to working condition, that is now not available to you, is that you could have checked the performance of the power transformer, and the valves in a real amp (not via simpler testing such as in a valve tester, which likely wouldn't confirm operation at nominal working voltages).

So the valves and PT, imho, remain a risk for your plan. Sometimes a lot of effort can go in to preparing a new build only to find that a particular part may have seemed ok, but ends up having a problem. I've been caught with PT's and OT's that can measure ok for DC resistances, and even turns ratio testing, but have shorts between turns or insulation breakdown that only shows up when fully energised and passing power or signal, or doing additional pre-testing such as with an Insulation Resistance tester.

[Richardgr]

Originally Posted by trobbins

One of the advantages of restoring the amp to working condition ...

The construction did not lend itself to maintenance, and there were some very crummy looking 80 year old electrolytics that must have been soldered in as the first components, before many layers of connectivity on top. The OPT is cheap and cheerful - 6cm wide, and the laminations aren't even properly clamped.

The components have just been separated into modules - the main 'PCB' with its connections marked - so I am still able to fix the critical components that have aged too much, and see if I get any sound out of it.

I had already acquired some NOS ECL11, UCL11 and AZ11, so I had already had a plan to make something with them, and this case is so small (30cm x 14cm) that it would sit on a bookshelf.

The original intention was to turn it into something useful, and sell it on; I am floundering in what my other half calls 'junk'.

Decisions, decisions!

[Richardgr]

Originally Posted by wd40addict

https://www.vintage-radio.net/forum/...5&postcount=21

Thanks for the meths tip for fixing loose valve bases - I was a little skeptical, but it worked a treat!

[Richardgr]

I have had to decide where to go with this project. After separating the main components, the chassis is just a rusty tin box, not particularly strong (1 mm plate), and is looking pretty sad.

Also, there is no intrinsic purpose with the device. At some point it mixed a mic signal and amplified it to 8 watts. I have tried reverse engineering the circuit, but I am not able to make sense of the way the EF12 drives the EL11 with the two tone elements - I am wondering if the notes I made before dismantling are correct.

I have a plan. I have a simple circuit with EF12 and EL11 from a Russian site, so I will build that in the chassis. That exercises the PSU and OPT. If that has good sound, then I need one more EL11 to build a simple SE amp on that chassis. I would need to use solid state rectification, and buy new OPTs, but it will have a small footprint and can be used with headphone too (the OPTs have 32 ohm taps).

After that there is an alternative schema from the Russian site with Schade feedback, but more complicated PSU. Alternatively to that, I am very interested in the Multiplied Transconductance Amps on the Linear Audio site, and since I have some ECL11’s, am tempted to go that route later.

So now I just need to get the single channel up and running as a proof of concept.

[Richardgr]

Here is the schema I will implement. The second one with feedback is included (it is not anode to anode as I thought).

[Richardgr]

Originally Posted by trobbins

The 1800pF cap was likely for tapering off treble, and constraining voltage transients across the winding.

I have tried to capture that role better on the schematic.

Originally Posted by trobbins

C1 at 50nF doesn't sound right, as it appears to be local feedback (ie. it could be 5pF or 50pF). There is no schematic indication of output transformer speaker connections and if there was feedback. The source switch would cause some popping, as the coupling cap to the input triode mode stage is not leak grounded.

This is the big problem I have with the schema I have drawn - I would have assumed that the 50nF cap was the coupling capacitor from the EF12, but I have checked the wiring left on the sockets and the view I created before dismantling, and I can't see how it worked. I am trying to find the other pictures I took before dissassembly ;-/

Originally Posted by trobbins

Did you confirm there was no heater winding direct connection to ground (ie. a CT)? It is possible that the heater winding didn't have a direct connection to ground, and the the heaters float to some DC voltage between the limits of the cathode voltages of the three valves (due to heater/cathode leakage), and the 1000pF capacitor provides a DC filter for that voltage - for hum suppression - that was not a common technique but has been used by some manufacturers.

Yes, that makes sense now.

Thanks for all of that!

Cheers, Richard

[Richardgr]

Originally Posted by trobbins

One of the advantages of restoring the amp to working condition, that is now not available to you, is that you could have checked the performance of the power transformer, and the valves in a real amp (not via simpler testing such as in a valve tester, which likely wouldn't confirm operation at nominal working voltages).
...

Yes, agree totally. The proposed schema above is very minimalist, and will exercise all the main components (transformers, choke and tubes).

I have a valve tester, an old AVO in fact, but I was surprised to see it did not have the footless base, so I will need to cobble something together to test the tubes.

[Herald1360]

Anyone else find circuits like these much easier to follow with HT+ disentangled from the rest and feedback likewise?

[Richardgr]

Originally Posted by Herald1360

Anyone else find circuits like these much easier to follow with HT+ disentangled from the rest and feedback likewise?

Yep, that looks cleaner.

Another interesting difference in schemas is whether to show the PSU with all voltage drop resistors and decoupling caps and have multiple B+ points on the main schema, or to put them all on the main schema, with just a single B+ supply.

Is there there a preferred approach?

[Richardgr]

Some progress, but not really sure where I am going with this thing.

I have re-wired the power supply as it was originally, so I can test the rectifier, choke and PT. I had tested the PT in isolation, but I wanted to see what sort of B+ there would be with the rectifier and choke.

The original can caps were 16uF + 16uF, 500VDC. I had a go at stuffing the original caps, and it went quite well, so that was a good exercise for the future. I did not want to exceed the capacitance for the AZ11 rectifier, but I only had 16uF 350 VDC caps, so I used serial caps for the first cap after the rectifier (= 8uF 700VDC), hoping B+ to the next cap was not going to be over 350Vdc.

As it turned out, B+ was 350v with the variac at 200VAC. The heater voltages were going to be more than 10% over, so I need to see if there is another tap on the transformer for 245VDC (I think the one below that that I am using is 220VDC, and my line voltage is 235VDC).

Since the supply was unloaded, I tried to load it with 10 x 470R 5W resistors in series. I did not have the confidence to keep the variac going beyond 80V since the voltage did not seem to be rising, and I did not want to damage the choke or the PT. How do other people test a PSU under load?