[Ron Ramirez]

The 37-690 lives!

Yesterday, I finished up the amp/power supply chassis. One of the wires coming out of one of the power transformers - which just so happened to be one side of the AC line primary - had a broken spot just outside the transformer shell. That received a piece of heat shrink tubing, which now extends into the shell for safety and the "just in case" factor.

Tonight, I pulled the tweeters out of the cabinet. Everything else was already out. Spread everything out across my workbench, plugged the 690 into my Variac, and slowly brought it up. By 90 VAC I was picking up music from my home transmitter. I took it up to 115 and let it play for an hour or so, and it did well.

Next move: to give the chassis an alignment.

[morzh]

Yeah, it's a huge relief when a beast like this one plays right out. I felt it when my 37-116 played right after recap. Otherwise the imagination starts scaring you into thinking how awful and complex the troubleshooting will be.
Congrats, the rest should be farly smooth from bere.

[Ron Ramirez]

Happy Independence Day!

This morning I gave the 37-690 a complete alignment. For those of you who have never done this on a 37-690 or 37-116, let me tell you that this is no easy task!

Here is the 37-690 spread out across my workbench:

[Image: http://www.philcoradio.com/images/phorum...e00002.jpg]

The dial lamp has been replaced with an LED light from Pinball Life. It is warm white in color, and looks just like an incandescent:

[Image: http://www.philcoradio.com/images/phorum...e00003.jpg]

Observations:

1. The radio does not slowly come up - it comes on, hesitates for a few seconds, and then the audio "pops" in.

2. Band 2 is dead - typical on almost every 37 and 38-690.

3. The two 50 watt heatsink resistors I used to replace the defective Candohm get HOT HOT HOT after two hours of operation. But these are more substantial than the two-section Candohm they replaced.

4. The speaker's outer cone is separating from the inner, stiffer cone. This can be easily repaired with some fabric glue, however.

5. Whoever worked on this set last really cut some corners. Take a look at one of the four Magnetic Tuning lamps:

[Image: http://www.philcoradio.com/images/phorum...e00004.jpg]

Naah, don't bother rebuilding the lamp socket - just solder the 6.3 volt "hot" lead to the end of the lamp!

If I can, I am going to rebuild these four lamp sockets the right way, and then stop at that point. It plays very well and I think my buddy will be happy with it overall. While it isn't 100% perfect, it is in better shape than it was when I acquired it.

Thoughts? Comments?

[morzh]

Ron

Is band 2 the same as in 37-116? I got no dead bands, all have at leat some stations on them.

Candohms hot...if it dissipate so much (for a 50W resistor to get so hot you have to dissipate at least 15W or so) - and I do not have a good qualty sch - do you think it makes sense to try to play with the values to maybe reduce currents while keeping voltages?

[Ron Ramirez]

All four of the Magnetic Tuning lamp sockets have now been rebuilt, and the incorrect #55 lamps replaced with #44.

Mike (Morzh), and anyone else interested - Take a look at the partial schematic linked here:

[Image: http://www.philcoradio.com/images/phorum...c0157a.jpg]

The resistor in question is part (177). Philco made a production change in which the 325 and 80 ohm sections became separate resistors and the values changed to 300 and 75 ohms, respectively. This 37-690 had that production change.

I have not measured voltages at the taps, and Philco does not give that information in the Service Bulletin.

Candohms hot...if it dissipate so much (for a 50W resistor to get so hot you have to dissipate at least 15W or so) - and I do not have a good qualty sch - do you think it makes sense to try to play with the values to maybe reduce currents while keeping voltages?

I am open to all suggestions.

Oh, and regarding Band 2: It receives the alignment signals of 4.5 and 1.7 mc, but very weak. The oscillator adjustment screws on that band have no effect.

And, yes, the RF unit in the 37-690 is identical to that used in the 37-116 (and 37-675).

[morzh]

I think I was close in my power estimate, if the voltage across that is aboute 300V the power will be close to 16W. 250V will give you 11W.

From this part of the sch the to resistors bias the cathode of ABAC, so it is not straightforward to change values. Maybe some concessions were made to minimize number parts (who counted kWh back then.....), sacrificing power dissip. Maybe not. Has to be studied.

Well, If the chassis can disipate that (I think it can) then good.

[Ron Ramirez]

Okay!

After stopping for a planned three day excursion into Ohio last weekend, during which time I attended my first ever Midwest Audiofest at Parts Express in Springboro, my thoughts are turning back to buttoning up this 37-690 and getting it ready to go home to its new owner.

I am just not comfortable with the amount of heat generated by the 3K and 2.2K chassis mounting resistors. They make that portion of the chassis get really hot...way too hot to keep my hand on...and a lot of components are in that area including output tube sockets, the special sockets for the cables going up to the upper chassis, and the speaker cabling. Plus, both power transformers are close by.

I have made an executive decision to replace the chassis mount resistors I installed (3K & 2.2K) with ceramic resistors that mount on the chassis but do not use the chassis as a heat sink.

I will also mount them away from the mass of components where the original Candohm was, and where the chassis mount replacements currently are.

I'll get the parts ordered this weekend.

[Mondial]

Ron, are you still going to mount the new resistors under the chassis? If so the same amount of Watts will be dissipated and they will heat the inside of the chassis more than the chassis mounted ones.

The heat has to go somehere, and heat sinking to the chassis still results in lower under chassis temperature, as the heat is not trapped underneath.

[Ron Ramirez]

What else can I do? I can't mount these things above the chassis...shock hazard...

Okay, guys: Can the circuit be redesigned to use Zeners instead of resistors? And if so, will there still be a major heat factor, or not?

I am definitely not comfortable with the amount of heat these things are generating, and then that chassis sitting on the bottom of the wooden cabinet (even though the cabinet has slots for a modicum of ventilation).

[Mondial]

Why not separate the 3K and 2.2K resistors so they do not heat the same concentrated location on the chassis? You might leave the 2.2K where it is and relocate the 3K to the right side of the chassis where there is no resistor now. The 3K dissipates the most power and generates the most heat.

In the end a certain amount of heat has to be dissipated, but if it can be distributed more evenly over the surface of the chassis, a lower peak temperature will result.

These resistors form part of a voltage divider for the power supply, providing B+ and bias for the circuits, while at the same time serving as a bleeder. They also provide a load to produce the required field current for one of the tweeters. There is no easy way to redesign it.

In any case, keep in mind that the heat generated now is no more than it was when the set was new, and was then considered acceptable by the designers.

[Ron Ramirez]

Acked.

That would be an easy way out...to just move the 3K...actually, someone had been in this set before and had placed a 25 watt 3K sand resistor off to the side, replacing the open 3K section of the original Candohm.

I'll do that this weekend. Thanks.

(actually, I would suspect the heat would be slightly higher now, since the AC line voltage has crept up from 110-115 volts in 1937 to 125 volts now.)

[Ron Ramirez]

As a side note: I am now curious about just how much heat is generated by the original Candohm? I owned a different 37-690 several years ago which still had the original Candohm in good shape, but I never thought to check and see if that area of the chassis became hot. I suspect it must have, due to the amount of current flowing through those resistors and the resultant heat being generated.

[Mondial]

That's a good point, Ron. Today's higher line voltage will increase the power dissipated in the resistors by about 15% to 20%, but I would assume the engineers involved designed the radio to withstand worst case high line conditions.

All the more reason to use a bucking transformer when operating these old sets.

As you mentioned, I would expect the chassis to get as hot with the old Candohm as it did with the new replacements. As long as the resistance and the mounting location is the same, the power dissipated and the resulting temperature would be identical.

[morzh]

Heatpipes.

[Ron Ramirez]

Here's something about the chassis mount resistors: They are encased within their own heat sinks, but are designed to be mounted to metal for maximum heat dissipation. As a result, they are smaller than their ceramic brethren.

A 50 watt ceramic resistor is 4 inches long. The same 50 watt heatsinked chassis mount resistor is less than 3 inches long.

If the ceramic resistor gets hot due to the amount of current flowing through it and the watts dissipated by it, the surrounding air will become hot. Anything very close to the resistor, or touching it, will also get hot.

If the heatsinked resistor gets hot, the area of the chassis where it is mounted gets just as hot.

This is why I was considering replacing the heatsinked resistors with ceramic.

Your thoughts before I proceed to merely move the 3K heatsinked resistor are welcome.