Restoring a Supreme Model 562 Audolyzer
The model 562 Audolyzer was advertised by Supreme as a dynamic RF and AF signal tracer and radio analyzer. It contains a 5-band tuned radio frequency (TRF) receiver with a built-in speaker. The receiver is also integrated with a built-in vacuum tube voltmeter (VTVM) and an ohmmeter. It looks like the model 562 was manufactured from 1939 to 1947 before being replaced by a newer model 688.
I did a little research online where I read comments that the earliest units had a gray face plate in an oak case, and the earliest circuit had a #80 rectifier tube, did not have the ohmmeter, and used a bias battery in the first audio amplifier. There is a 1939 schematic available online for this circuit.
Later units were said to have a black metal cabinet and face plate, and circuit-wise the audio circuit was modified to eliminate the bias battery, the rectifier was updated to a 5Y3 and, of course, it had the ohmmeter.
But from what I've seen in period catalogs and radio magazines it isn't as simple as that. One of the earliest ads I found for an Audolyzer is in the September 1939 Radio Craft magazine and it shows a black front panel with what appears to be the metal cabinet (handle on top). The 1940 Supreme catalog also shows a black front and metal cabinet. And there is an ad in the October 1940 Radio Service Dealer magazine showing a 562 with a black front panel in what appears to be the oak case (handles on the side). But, if the 562 came out in 1939 and was already advertised with a black front panel and a metal case, then when were all the gray units in oak cases made? Having a gray panel or black panel is not an accurate guage of the instruments age.
One of my favorite ads is from 1941 and shows an Audolyzer, the Vedolyzer and a model 561 signal generator all with black panels and matching oak cases. And interestingly, the Supreme 1946 catalog shows a black 562 in a guess what: a clearly stated quarter-sawn oak case. The 1945 catalog also shows the oak case. So the oak case is also not an indication of age.
My particular Audolyzer has a gray face in an oak case, and the circuit has the #80 rectifier like the original schematic, but the ohmmeter and the rest of the circuitry matches the schematic for later versions. A schematic I found that's dated 1946 says it's for units S/N 2100 and up and shows a 5Y3 rectifier, but my Audolyzer is S/N 2443 and still has a #80 rectifier. Serial number-wise this unit isn't one of the earliest, so at least some gray face units must have been made well after 1939 and with #80 rectifiers.
Supreme apparently did make a lot of incremental changes over the years, but you can't just look at the color or the case and say one unit is earlier or later than another. The only way to tell for sure is to refer to the serial number.
Later units were said to have a black metal cabinet and face plate, and circuit-wise the audio circuit was modified to eliminate the bias battery, the rectifier was updated to a 5Y3 and, of course, it had the ohmmeter.
But from what I've seen in period catalogs and radio magazines it isn't as simple as that. One of the earliest ads I found for an Audolyzer is in the September 1939 Radio Craft magazine and it shows a black front panel with what appears to be the metal cabinet (handle on top). The 1940 Supreme catalog also shows a black front and metal cabinet. And there is an ad in the October 1940 Radio Service Dealer magazine showing a 562 with a black front panel in what appears to be the oak case (handles on the side). But, if the 562 came out in 1939 and was already advertised with a black front panel and a metal case, then when were all the gray units in oak cases made? Having a gray panel or black panel is not an accurate guage of the instruments age.
One of my favorite ads is from 1941 and shows an Audolyzer, the Vedolyzer and a model 561 signal generator all with black panels and matching oak cases. And interestingly, the Supreme 1946 catalog shows a black 562 in a guess what: a clearly stated quarter-sawn oak case. The 1945 catalog also shows the oak case. So the oak case is also not an indication of age.
My particular Audolyzer has a gray face in an oak case, and the circuit has the #80 rectifier like the original schematic, but the ohmmeter and the rest of the circuitry matches the schematic for later versions. A schematic I found that's dated 1946 says it's for units S/N 2100 and up and shows a 5Y3 rectifier, but my Audolyzer is S/N 2443 and still has a #80 rectifier. Serial number-wise this unit isn't one of the earliest, so at least some gray face units must have been made well after 1939 and with #80 rectifiers.
Supreme apparently did make a lot of incremental changes over the years, but you can't just look at the color or the case and say one unit is earlier or later than another. The only way to tell for sure is to refer to the serial number.
The chassis removed from the wood cabinet. The cabinet is falling apart at the joints and needs reglued and refinished. I like the oak cabinet better than the metal cabinet.
|
This Audolyzer is not in great shape, but it is complete inside, even though it's obviously been worked on, looks like at least a couple of times. The older repair work was mostly well done, but later someone else was trying to troubleshoot it, snipping leads and tack soldering components back. Just wiggling some of the resistors and capacitors was enough to break the leads free. Real poor work, but it made it a little easier to remove the old components.
The paint is literally falling off the front panel and there are rusty areas on the chassis. The chassis can probably be touched up but the front panel will have to be recreated from scratch. |
|
The chassis is dirty but not too bad except for the rusty places. Not the worst I've ever worked on. At least the paint isn't falling off the chassis like it is on the front panel. There was a battery for the ohmmeter on the left side where it's rusted. The metal strap that held the battery is also very rusty.
The chassis is actually painted with the same gray wrinkle finish as the front panel, something I don't often see. A little better quality than just a bare metal chassis. |
The tube complement, starting with the glass tubes, from front to back, are a 6J5 for the VTVM, a 6V6 audio output amp and the #80 full-wave rectifier. On the metal tube side the first two are 6SG7 RF amplifiers and then a 6SQ7 detector and 1st audio amp. The schematic shows two 6SK7 tubes but the 6SG7 is a valid substitute.
|
The front panel removed from the unit. Every time I touch it more paint falls off. I want to scan it before I loose too much paint so I can recreate the dials and lettering on my computer. I hope to be able to print decals to relabel it.
The top left corner is bent, apparently from having been dropped. The oak cabinet corner isn't damaged so it must have been out of the case when it happened. It will get straightened when the panel is repainted because if I do it now more paint and lettering will fall off. |
|
Under the chassis is clean but there's a lot of paper and electrolytic capacitors to replace. There are 22 paper caps in all plus 5 electrolytics. The shielded box contains the band coils and trimmer caps for the receiver plus several paper caps and resistors.
All the resistors need checked and since they are the older uncoated "roundy" style it's a good bet some will need replaced. The input and output jacks are oddball Motorola-style antenna jacks as used on automotive radios. |
I noticed in a picture, from an Audolyzer manual of a metal-cased unit, that the two jacks on the right were changed to 1/4" phone jacks. Only the probe jack was still an antenna-style jack on that unit. I thought about replacing the other two jacks with either 1/4" phone jacks or BNCs, but neither fit the existing holes all that well so I just left the originals and will make an adapter cable.
|
The top of the chassis cleaned up nicely and I found an off-the-shelf spray paint that was a close color match to the original gray finish. The rusty spots were treated with naval jelly, then paint was sprayed into a cup and brushed on. The touched up spots were dry brushed into the existing finish and it came out much better than expected. The touched up areas looked so good I ended up dry brushing the whole chassis top.
|
|
Inside the shielded compartment are the band coils and trimmer capacitors for the TRF receiver. There's also three paper capacitors to change and several resistors. Only two of the resistors were out of spec, both are labeled 1500-ohms on the schematic but were actually 1600-ohm resistors in this unit. They are visible in the center of each half of the box.
The TRF receiver has two RF stages, a detector and then two AF stages. It tunes from 95Kc to 14.5Mc which covers the common IF frequencies. It also has an antenna input coax in the rear to use it as a radio. The antenna is connected to the receiver RF input when the PROBE SELECTOR is in the AF OUT position. The radio's audio output from the first audio amp is then routed to the test probe and also on to the 6V6 audio amplifier driving the speaker and AF OUT jack. The output of the probe can be used as an input source to test audio circuits. Or you can just listen to the radio. |
|
|
I removed the variable capacitor and cleaned it with a product called Super Clean. It came out almost as shiny and clean as new. It's a degreaser so a coat of WD-40 was applied after cleaning to protect the metal.
The RF Multiplier switch has two paper caps to replace. The hard part was just getting the cover off. In the early schematic there are three capacitors on the switch. The 0.15uF for the 1000 position was removed and that switch position grounded in the later schematic. Not sure if that was just a cost savings move, so I think I will add the capacitor back to see what, if any, difference it makes.
(Update: I added the 0.15uF cap back and compared it to just a GND and didn't see any difference in signal levels for the 1000 position.)
On the subject of cost savings, I notice quite a few resistors that don't match the schematic, such as substituting a 360K resistor where the schematic calls for a 350K, or a 910K for a 1Meg. I wonder if Supreme might have had a large stock of certain values and tried to use them wherever possible.
The RF Multiplier switch has two paper caps to replace. The hard part was just getting the cover off. In the early schematic there are three capacitors on the switch. The 0.15uF for the 1000 position was removed and that switch position grounded in the later schematic. Not sure if that was just a cost savings move, so I think I will add the capacitor back to see what, if any, difference it makes.
(Update: I added the 0.15uF cap back and compared it to just a GND and didn't see any difference in signal levels for the 1000 position.)
On the subject of cost savings, I notice quite a few resistors that don't match the schematic, such as substituting a 360K resistor where the schematic calls for a 350K, or a 910K for a 1Meg. I wonder if Supreme might have had a large stock of certain values and tried to use them wherever possible.
|
The speaker was also rusty and needed painted. It's an R&A brand and says it's made in England, and if the schematic is correct it is not original. The schematic shows a speaker with a field coil, not a permanent magnet. And the output transformer was hanging from it by just a single screw, and looking under the chassis revealed the wires from the 6V6 to the transformer had been cut and spliced. So it's not original either. It must have been changed a long time ago though, because the spaghetti tubing on the splices was dry rotted. The field coil is labeled as 1000-ohms on the schematic and has been replaced in the power supply by two 2000-ohm power resistors in parallel.
|
(Update: I found this R&A speaker online and it dates from the 1960's, so this repair/replacement is about 50 years old.)
|
The spliced-in audio transformer is very small and not a good match to the 6V6, if you want any kind of fidelity, so I ordered a replacement designed specifically for it. It's more than twice as hefty as the old transformer and mounts to the rear panel under the chassis where there's already a couple of mounting holes and a vacant space, probably where the original transformer was mounted to begin with. This one drops right in and matches the pre-punched holes.
The new transformer is rated at 5 watts and has a primary winding impedance of 5000 ohms to match the 6V6, and a DC resistance of 500 ohms, exactly what's on the schematic. |
|
|
The oak cabinet joints were reglued and then it was sanded, stained and coated with clear enamel. The original handles were missing and replaced with "screen door" handles from a hardware store.
The metal back panel was also missing and a new one was cut from a decorative metal panel. Grommet holes were also added for the power cord and antenna coax cable and I made a reproduction power specs label. I cut the power cord hole large enough to get the plug through and I still need to buy a big grommet for it. It's not an exact match for an original back but close enough. It buttons up the rear nicely and still allows for good ventilation.
The metal back panel was also missing and a new one was cut from a decorative metal panel. Grommet holes were also added for the power cord and antenna coax cable and I made a reproduction power specs label. I cut the power cord hole large enough to get the plug through and I still need to buy a big grommet for it. It's not an exact match for an original back but close enough. It buttons up the rear nicely and still allows for good ventilation.
|
I converted an old probe found in a junk box for use with the Audolyzer. I replaced the original internal resistor with 100KΩ, and changed the connector to a Motorola-type. The probe end is a two-piece affair, it must have originally had interchangeable tips but the others are long lost. It uses the same antenna-style connectors to plug the two halves together.
One of the advertised advantages of the Audolyzer was that a single probe was used for troubleshooting a radio from antenna coil to voice coil. You don't have different probes for RF, AF or the VTVM like on a Rider Chanalyst. But using the probe with the VTVM is optional, a set of regular test leads connected to the meter jacks can also be used, while using the main probe for a different function at the same time (with meter in EXT. VOLTS position). Ohms measurements are also made using the test leads connected to the meter jacks. |
|
|
The front panel was stripped and repainted in a black wrinkle finish. I like the black face but decided to recolor the panel gray to match the chassis. The wrinkle paint is VHT brand and available at auto stores. VHT also makes gray wrinkle paint but it was not available locally.
I've found a couple of fonts that are pretty close to the Supreme lettering so now I can recreate the dials and print them on water slide decal paper. I also found a couple of products called Microsol and Microset to "melt" the decals onto the finish of the panel.
I've found a couple of fonts that are pretty close to the Supreme lettering so now I can recreate the dials and print them on water slide decal paper. I also found a couple of products called Microsol and Microset to "melt" the decals onto the finish of the panel.
All the new decals are on. That was a lot of work. It's not perfect but I'm still very pleased with how it came out. After the labels dried the panel was coated with multiple coats of clear matte enamel spray to protect them.
Matching colors of the decals was a pretty impossible task. They are one shade of gray on screen, and another shade when printed and applied. Plus I printed several sheets and each sheet is a little different. The matte spray also seemed to tone down the labels some. In the end they are close enough for me. Also my camera exaggerates the difference between panel and labels.
Matching colors of the decals was a pretty impossible task. They are one shade of gray on screen, and another shade when printed and applied. Plus I printed several sheets and each sheet is a little different. The matte spray also seemed to tone down the labels some. In the end they are close enough for me. Also my camera exaggerates the difference between panel and labels.
Some assembly required, but it's looking like an Audolyzer again. This thing was taken down to the point where putting it back together was almost like building a kit. During reassembly all rusty hardware was replaced with new. The black oxide fillester head screws on the front panel are hard to find so they were cleaned and reused. Most of the nuts and washers for the front panel controls and switches were also replaced. It looks so nice I can't wait to get it powered up.
The chassis is all back together and instead of reusing the metal tubes I installed a full set of glass tubes, just for looks. Also, on the left side plate I added a "D" cell battery holder for the ohmmeter. Hardly looks like the same unit now.
The three lowest ohmmeter ranges only use the 1.5-volt battery and can be used even when the Audolyzer is turned off. So I was able to test them already and all three work great. Readings are well within the tolerance of the divider resistors.
The three lowest ohmmeter ranges only use the 1.5-volt battery and can be used even when the Audolyzer is turned off. So I was able to test them already and all three work great. Readings are well within the tolerance of the divider resistors.
|
As soon as the face plate was back on the chassis I finished replacing the last of the paper capacitors, a few more resistors, and reconnected some loose wiring. All together I replaced 30 resistors, 22 paper capacitors and 5 electrolytic caps, plus one of the original metal tubes was bad, and I installed a new audio transformer.
|
A look under the completed chassis. All the old waxed paper caps and old electrolytics are now gone. The next steps are powering up the set and checking the power supply voltages, checking the voltmeter and higher ohms ranges, and then aligning the TRF receiver.
|
Since I decided not to replace the Motorola connectors, I made a shielded adapter cable for connecting a scope or patching audio in or out of the Audolyzer. The other probe is the main probe I made earlier, which is used for testing RF, IF and AF circuits, and for voltage measurements. That's it for cables and probes except for a simple set of meter leads for use with the ohmmeter and also with the voltmeter in the Ext. Volts setting.
|
|
The 1946 schematic has a 5Y3 rectifier and says the B+ should be 250VDC. However, the older 1939/1940 schematic has a #80 rectifier and shows the B+ as 270VDC. In my unit with a #80 rectifier it is actually around 284VDC, so as a test I used a Variac to set the AC line to 117.5VAC (as specified on the rear panel) and this dropped the B+ to 269VDC, right where it should be.
|
The reason my B+ is reading high is because my line voltage here is 122VAC. For safety I added a fuse to the AC line and used a pigtail fuse holder so I wouldn't have to drill a hole in the chassis. And while I was at it I inserted a 10-ohm resistor to lower the line voltage to the transformer primary back down to 117VAC, just like in the good 'ol days. This is much healthier for the transformer and the set as a whole.
Originally the instrument had a 2-wire power cord, but I found it needs a good ground to keep hum out of the RF stages, so I changed to a 3-wire power cord. I figured this was simpler than keeping the coax shield always connected to an external ground.
|
To align the voltmeter there are three controls. Adjusting three potentiometers at once sounds like a recipe for insanity, so I set the Zero Adjust (R43) to 50 and left it there. Then I connected a 1 volt source to the meter (on 1-Volt scale) and adjusted the 1K control (R42) in series with the meter to read +1 volt. Next I reversed the 1 volt source and checked the negative reading, tweaking the chassis mounted control (R44) to set it. Changing one setting affects the other so I repeated these steps until the meter read 1 volt in both directions and was centered on zero.
(Update: The meter zero kept drifting and I had to replace R42. Looks like the metal bands crimping the ends of the rheostat's nichrome wire are rusty and not making good contact.) |
|
There are no alignment instructions in the manual for the TRF receiver, but normally trimmer capacitors are adjusted at the high end of the band where they have the most effect. For Bands A, B, and C there are both parallel and series trimmers, but bands D and E have only a single trimmer, which according to the schematic is in parallel with the coils. They are really series trimmers, the schematic is in error.
I adjusted the parallel trimmers on the coil forms of band A, B and C to align the dial frequency and adjusted the series trimmers for maximum signal. After adjusting each series trimmer I scanned the band for oscillations using an oscilloscope. If oscillations were occurring I detuned the series trimmer of the second RF amplifier for that band just enough to completely eliminate it. It didn't take a lot of adjustment and I did not find it necessary to readjust any other trimmers to stop oscillations. I only encountered oscillations on bands A and B. |
Other alignment issues included:
- on the bands with both parallel and series trimmers (Bands A, B and C), the two trimmers interact. If a trimmer won't peak (all open or tightened down) then back off both trimmers a turn or two and try again starting with the parallel trimmer.
- Band E was completely dead, couldn't hear a peep from the signal generator. It turned out to be an open secondary coil to the 2nd RF amp. The winding was damaged and the wire broken in one spot on the coil, but with the help of a microscope it was repaired. I also needed to add a small NPO capacitor across the trimmer for this band to get it to peak.
- I started the alignment on Band E. The tuning capacitor's range is wider than the dial scales and turns past both ends of the scales, which raises the question of where it should be initially aligned with respect to the dial. To start I set it by splitting the difference so it traveled the same distance past each scale end. Then I aligned the top end of band E. Next I checked the alignment at the bottom end and loosened the dial set screws and moved it to correct the error on the low end. I kept repeating these steps until I was able to get band E to track over the entire scale. I wanted band E to track correctly because it covers WWV at 5 and 10MHz and CHU Canada at 7.85MHz.
- on band D fully tightening the trimmer did not allow me to peak the signal. I added a 12pF NPO capacitor across the trimmer and then was able to peak it. Probably the inductance of the coil has changed over the last 75+ years.
- on the bands with both parallel and series trimmers (Bands A, B and C), the two trimmers interact. If a trimmer won't peak (all open or tightened down) then back off both trimmers a turn or two and try again starting with the parallel trimmer.
- Band E was completely dead, couldn't hear a peep from the signal generator. It turned out to be an open secondary coil to the 2nd RF amp. The winding was damaged and the wire broken in one spot on the coil, but with the help of a microscope it was repaired. I also needed to add a small NPO capacitor across the trimmer for this band to get it to peak.
- I started the alignment on Band E. The tuning capacitor's range is wider than the dial scales and turns past both ends of the scales, which raises the question of where it should be initially aligned with respect to the dial. To start I set it by splitting the difference so it traveled the same distance past each scale end. Then I aligned the top end of band E. Next I checked the alignment at the bottom end and loosened the dial set screws and moved it to correct the error on the low end. I kept repeating these steps until I was able to get band E to track over the entire scale. I wanted band E to track correctly because it covers WWV at 5 and 10MHz and CHU Canada at 7.85MHz.
- on band D fully tightening the trimmer did not allow me to peak the signal. I added a 12pF NPO capacitor across the trimmer and then was able to peak it. Probably the inductance of the coil has changed over the last 75+ years.
|
One of the biggest problems I found was the coax for the antenna. I don't understand how this ever worked. Remove the coax ground and the receiver has good sensitivity, reattach the ground and the receiver goes deaf to all but strong signals. And it's not a bad piece of coax. A new piece does the same thing. Supreme attached the coax center conductor through a capacitor directly to the grid of the first RF amplifier. Attaching an antenna directly to the grid is an outdated but common circuit for a TRF receiver of the 1920's, except minus the coax. It looks like an impedance mismatch but how could Supreme not have seen this?
To fix it I added an untuned tapped coil to the grid circuit (grid to gnd) and removed the existing 100K resistor. The coil came from a parts drawer and looks like it's an old AM radio oscillator coil. I connected the capacitor from the coax center conductor to the coil's tap and It works very well. The receiver is quite sensitive now and grounding the coax shield causes no problems. But I'm still scratching my head trying to understand what Supreme was thinking. |
|
Another issue Supreme seems to have missed the class on is coil shielding. On an antique radio forum I read a post about an Audolyzer that said the coils were too close together, which allows the band coils of the selected band to couple with the coils of the next lower band, causing portions of the band to go dead because the lower band coils are absorbing signal from the active coils. Their fix was to short out the lower band coils, but you need to add a switch wafer to do this. I tested each band and found this problem affecting band E.
I fixed it by adding a piece of brass sheet, insulated in heatshrink, between the coils for bands D and E, and soldered it to the ground bus wire of the coils. Sliding it closer to the coil for band E provided the best shielding. Also had to be careful to steer clear of the bare bus wire on the left which is B+. |
|
The final issue with the receiver was very low, choppy audio out of the detector for all but the strongest signals. This is another problem where I'm not sure what Supreme's reasoning was, but by adjusting the value of the cathode resistor of the 6SQ7, lowering it to 680 ohms, the audio is much improved and the receiver detects much weaker signals now.
Supreme also attached the volume control wiper directly to the grid of the 1st AF amp in the 6SQ7. I changed this by adding a 4700pF capacitor between grid and wiper, and a 4.7M resistor from grid to ground. |
After the repairs and modifications the TRF receiver does quite well. The selectivity is a bit wide, to be expected from a TRF, but the sensitivity is good. I'm using about 30 feet of wire from a loft window as an antenna and can develop 30 volts or better of AVC voltage on a strong shortwave station. And the audio fidelity is much better than an AA5. The audio is clean and hum free.
Now that the TRF receiver is aligned and has some sensitivity, it's time to see how the Audolyzer works as a test instrument. I'm restoring a 1936 Sears Tombstone radio and it gives me the perfect opportunity to test the Audolyzer's capabilities. As I began testing the Sears radio I quicky discovered a possible answer to my questions about what Supreme was doing with the TRF sensitivity. The TRF receiver is so sensitive after my mods that just tuning in the test radio's IF on the Audolyzer, and then bringing the probe near the radio's IF amplifier, sounds like you just tuned in the radio's station on the TRF's dial. Actually touching the probe to the IF amplifier circuit might damage the speaker if the volume isn't turned down.
Supreme's procedure says to feed a signal generator (such as the Supreme model 561) to the radio, and since the signal level of the generator can be adjusted as needed, the lower gain of the original 562 circuit probably wouldn't be a problem. Maybe Supreme was intentionally throttling the gain of the TRF receiver. Not sure if that's what they were up to, but I'm happy with my Audolyzer as it is.
Now that the TRF receiver is aligned and has some sensitivity, it's time to see how the Audolyzer works as a test instrument. I'm restoring a 1936 Sears Tombstone radio and it gives me the perfect opportunity to test the Audolyzer's capabilities. As I began testing the Sears radio I quicky discovered a possible answer to my questions about what Supreme was doing with the TRF sensitivity. The TRF receiver is so sensitive after my mods that just tuning in the test radio's IF on the Audolyzer, and then bringing the probe near the radio's IF amplifier, sounds like you just tuned in the radio's station on the TRF's dial. Actually touching the probe to the IF amplifier circuit might damage the speaker if the volume isn't turned down.
Supreme's procedure says to feed a signal generator (such as the Supreme model 561) to the radio, and since the signal level of the generator can be adjusted as needed, the lower gain of the original 562 circuit probably wouldn't be a problem. Maybe Supreme was intentionally throttling the gain of the TRF receiver. Not sure if that's what they were up to, but I'm happy with my Audolyzer as it is.
Page created 12/7/2019
Last updated 6/17/2020
Last updated 6/17/2020
