Collins 51J-4 Deluxe General Coverage Receiver.|
February 16, 2014, Collins 51j series PTO rebuild.
When I was a young man, I worked as a commercial radio operator at Alert N.W.T. located on the northern tip of Canada's most northerly Island called Ellesmere. In the radio room there were three Collins 51j-4 receivers which were used on a daily basis. I learned to appreciate these fine quality tube radios and I always wanted one for my home shack.
Here I am at the operating position at Alert taken while on shift in 1964. Note the 51j-4's and Hammarlund SP-210LX. There was also a fully equipped Collins-S line with the big Collins linear at Alert. We used the call sign, VE8ML and I was very active chasing DX while there. Until 1999 Alert was located in the North West Territories with the prefix VE8. When the boarders between Canada's northern territories were re-aligned a new territory was formed called Nunavut. It has it's own distinct prefix now, VY0.
In October 2012 I acquired a 51j-4 that was in good shape that had previously been used in Collins Laboratory.
I really enjoy turning the big Collins 51j-4 switch to the ON position and waiting in anticipation for the tubes to warm up. Reminiscent of days long ago leading up to when I received my ham radio licence a friend and I would listen to his father's table top short wave receiver to voices from far away lands. Radio New Zealand, Cuba, China, Japan and others are often heard on the Short Wave bands. I often tune to W1AW at 5 PM local time to listen to the ARRL bulletins then check WWB and WWBH or CHU for the time signals.
Much to my delight the 51j-4 is now situated next to a restored Viking Ranger transmitter and I often use them together to make both CW and AM contacts.
When the receiver arrived it didn't take me long to begin the process of cleaning it up.
I found on the WEB, "The Hollow State Newsletter #26" authored by Dallas Lankford that provides helpful information about changing the existing 51j-4 BFO circuit to a combined BFO/product detector as well as improvements to the AGC circuit.
The Dallas Lankford papers #26-37. BFO and Product Detector.
These modifications entail removing the front panel, rewiring and running cables to the switches on the front panel
This is a photo of the BFO circuit after I modified it to a combined BFO/Product Detector.
When I purchased the receiver it came without mechanical filters as it had originally been used in one of Collins factories as a lab instrument (hence the light grey front panel colour). The original 500 KHz Collins mechanical filters are sometimes available on the WEB but tend to be very expensive. I found a surplus electronic dealer on eBay located in Slovenia that sold Russian manufactured 500 KHz mechanical filters. He sent me the specifications for the filters so I ordered the 6.0 3.0 and 0.5 KHz BW filters at very reasonable prices. The item between the pliers is an orginal Collins coupling circuit that is inserted in one of the filter slots when the receiver does not have any mechanical filters installed.
The physical dimensions of the mechanical filters are not the same as the Collins filters so I fabricated a mounting plate from PC board for the new filters.
Here you can clearly see where the filters are to be installed.
Mission accomplished. The two wider filters have a small amount of ripple. I decided not to match them to the existing circuit as I found when listening to SSB or AM signals that I am unable to notice any ripple.
Update February 16, 2014, Collins 51j series PTO, Model 70E-15, rebuild.
After owning the receiver for 9 months I noticed that the receiver frequency began wandering up and down, sometimes as much as 100 to 150 Hz and then usually it would drift back to the original frequency I was listening to. I decided to find out why. Checking with a frequency counter I measured both the BFO and Xtal hetrodyne xtals and found them stable. When I connected a counter to the PTO (permantaly tuned oscillator) I noticed that it was not stable which was obviously the problem that caused the receiver frequency to wander.
The Collins model 70E-15 PTO is common to the 51j series and R-388A military receivers. Removing the PTO from the 51j-4 is a major undertaking. The PTO removal procedure is outlined in Section 5.4 in the 51j-4 manual. Here are a few points that are not mentioned in the manual:
1. Remove the PTO tubes and shields.
2. After removing the front Panel, see Sec.5.6.2, and the KHz Dial, remove the 3 spacers that are located between the PTO and the gear mounting plate. This is not easy but once they are loosened they can be unscrewed from the PTO with your fingers.
3. You can turn the PTO while tilting it towards the back to remove it.
4. Note it is not necessary to remove the PTO couplers.
5. I tuned the receiver to 20.0 Mhz before removing the PTO and then measured the PTO output frequency at 2.5Mhz. When re-installing the PTO remember to tune it to 2.5MHZ otherwise your dial can be way off frequency and there is no way to correct it without removing the PTO again.
The following photo is the PTO after removal.
Following suggestions I found on the WEB, I replaced the two bypass capacitors C005 and C005 as well as C008, the capacitor between the main inductor tap and V002 screen (specified as a 0.01 UF or 1000 pF). After applying filament and DC voltage to the PTO and letting it warm up the PTO continued to wander.
Next I decided to replace C004, C007 and all the resistors located under the tube sockets. I made a hand drawn picture of the location of each part and where the capacitor and resistor leads were soldered in place. The components are tightly packed so work slowly and carefully as you remove and replace the components. Here is a photo with the components removed.
After replacing all the components under the PTO tube sockets, I fired it up and it did not oscillate. I checked all the tube voltages and they were correct. Then I removed the capacitors one by one and checked the values with a capacitor checker and they were all OK. Not knowing what to do next I decided to try another type of capacitor for C008. For the first replacement I used a 0.01 Polystyrene capacitor. I found a new 820 pF Silver Mica capacitor and soldered it in place. To my surprise the PTO oscillated? I checked the capacitor I removed and it measured OK so maybe the polystyrene capacitor had too high an internal resistance?
After letting the PTO warm up for an hour I went back to make a chart of the PTO drift. NO oscillation. OH NO. Further checking with a VOM I found that there was no DC supply voltage on the V001 plate. There is a 27K dropping resistor from DC supply to the plate of V001 and the VOM indicated that the 27 K resistor was open. I could not believe it. It is buried underneath other components and very difficult to get to. I was able to snip the one lead off and insert another 27K another resistor on top and leave enough room for the cover.
Here is a photo of the PTO after insertion of all new parts with the 27K located on top.
I was glad to see that it oscillated but after only a few minutes I noticed that the PTO would make an abrupt jump in frequency. It turned out the main HV rectifier a 5V4, must have been in the process of breaking down as the HV would drop dramatically and then rise to the proper voltage. I didn't have any 5V4's on hand but several 5U4's so I used one of them and that fixed the problem. Success?
After letting the PTO warm up and stabilize I made a graph of the frequency stability. I noted the frequency every 15 seconds over a period of 10 minutes and it only drifts plus or minus 3.5 Hz. After installing the PTO back into the receiver and letting it warm up over night I can set the receiver to WWV at 20 MHz with a beat frequency of 650 Hz and returning to check the frequency several times in a day I can not notice that the frequency varies from a audio generator tone of 650 Hz. Amazing technology for a receiver that is nearly 70 years old.
Here is a photo of the 51j-4 after a thorough cleaning, re-alignment and the product detector and mechanical filters are installed. What a beautiful receiver.