1968 Ford AM-FM Stereo overhaul and sound system upgrade

Radio Diagnostics Continued



At point 3.) the signal is attenuated greatly and below the conduction threshold of the detector diode, hence no output on the radio. I checked the diode and it checks fine (diode check on a multimeter) and I checked that C65 is not shorted and it wasn't. So next I checked the resistance on the IF transformer.



Checking the primary coil of the last IF transformer.



8.3 ohms. Yeah I have newer multimeters but I do like the vintage stuff when appropriate to use. These old nixie tube Flukes are from 1968 and are spot on calibration wise and never had a problem with either of them.

Anyway, the schematic calls out 7.5 ohms, but I'm using a 2 wire resistance measurement so the resistance of the leads is adding to the real value some. The primary is fine.



Measuring secondary.



It's open. Well there's the reason why AM is dead.



Now I know when I take the radio apart I'll remove this IF transformer and see if I can't repair the broken winding. Now if I listened to the internet <big sigh> I would have taken the radio apart, replaced all the electrolytic's then put it all back together, only to have to take it apart again to fix this. Once you better see what it takes to take apart and put back together you'll really understand why it's important to diagnose first what you can, before you attempt any repairs.

Onto the FM stereo portion and giving it a once over to see why the FM radio is alive but not the stereo lamp.

This operation does require some specialized kit.



In this case I am using a Sencore SG-165. Sencore did make a more modern version of this it's an SG 80 I believe. There are also some other brands. I have a Leader stereo generator as well, but it doesn't have all the functionality of this Sencore. The big problem with older test gear is that often it's either out of calibration so much it's not useable or it's failed and out of calibration to boot. Then it costs even more to repair and get it to the point of using. I've already went through and rebuilt this Sencore and calibrated it.

How this works; with this function selected, it will generate an FM station frequency based on the setting of the big round dial (88-108MHz) and then you can manually turn on the stereo pilot and also the left and/or right speaker test tone. The output of this is fed into the antenna of the car radio and you tune the car radio dial to whatever this is set at station wise, or vise versa, doesn't matter. Then you can turn on the stereo pilot frequency of which should turn on the stereo lamp but otherwise the speakers should be silent and then you can turn on the left speaker or the right speaker test tone or both to check stereo separation.



So here's the setup, since both amplifiers in this radio are so neutered (way less than 1 watt) I am using the Fluke multimeters to measure the AC signal across the speakers. I have the car radio tuned to the stereo generator.



I have the stereo pilot on and the right side speaker tone on.



I have right side speaker with a decent amount of separation. There's a little on the left.

continued in next post.

Radio Diagnostics Continued



Selecting the stereo left side speaker tone.



Decent stereo separation, so the decoder is working, but not the stereo lamp.



Next obvious step is to measure voltage at the lamp as maybe the bulb has simply failed.



Well I only get 2.3 volts for a 12 volt bulb. If I put my peeper right up to the amber jewel on the radio I can see a hint of glow in a dark room. I doubt the stereo decoder is so far out of alignment as to cause the dim lamp because it has pretty good stereo separation still. I bet it's the transistor or R70 or both even.

So now I know to pull and check that transistor and resistor.

Ok that hits all the key areas of the standard AM-FM stereo. I realize I'm skipping a lot and if there are questions on how FM stereo works or even how AM works (AM is real simple) I'll be more than happy to delve into it. Oh there was one more thing I failed to mention, the radio will not shut off. It's a physical problem with the On/Off Volume control. For whatever reason, it's stuck on and will not click off.

At this point, we can take apart the radio with a rational solid repair plan.

You really want to take copious amounts of pictures taking intricate items like this apart. This is true if there are many wires that must be unsoldered.

Here's some of the more interesting pictures.



This is one of the output amplifiers. It's a TO3 standard package, but geeze these look like the TO3 prototypes.



It starts getting unruly pretty quick.



There's the other channels amplifier output transistor. I've worked on a lot of vintage electronics and I've never seen this rendition of a TO3 till now. Learn something new every day.



Almost there.



That little board flipped up is the entire AM radio portion. The remainder large 'L' shaped bottom board is the FM and stereo decoder. AM radios are very simple and very open if you take off the top and very easy to work on. This is antithesis of that.

Ford offers integrated 8 track but with AM radio only for 1968. Now you know why. There's no bloody room for an 8 track player with the FM stereo portion.



This may look scary at this point, but if you take your time and loads of pictures it's not. There's no need to take this apart further. So we'll stop here and start the long road to recovery on the next round.

More to come.

Cheers

Radio Repair Part I

I'll start with the easy peasy mechanical stuff first. I'll address the slipping manual tuning.



The manual tuning has a little clutch assembly. There's basically a flywheel, if you will, friction disc and a little pressure plate.



Here's the parts stack up.



The friction disc is, or was, a piece of serrated rubber that turned hard due to age. This needs to be replaced.



First, will be to estimate its original hardness. This is a Shore A gauge and so we can assume it originally was under 70 shore A.



I have two sheets of EPDM rubber to choose from; 50 A or 60 A rubber. You can buy these at McMaster. I choose the 60 A rubber sheet that came in under the rating a bit.



The original disc measures 0.049"



The replacement is ~0.059" There is an adjustment on the fork for the little pressure plate so a slightly thicker piece of friction material will work.



Since the replacement sheet was smooth, I used a very course sandpaper to scratch radial lines on one side of the friction and roughed up the pressure plate with finer sandpaper. I also glued the friction disc to the flywheel so there's only one side to slip on.



Cleaned and greased (white lithium) the manual tuning knob and reinstalled.



Adjust the tension on the pressure plate and it works like new, no more slipping manual tuning knob and the preset buttons disengage the clutch fully when pressed. Cross that off the list.

Continued in next post.

Radio Repair Part I - Continued

Next is the light insulation from the stereo lamp to the front jewel. It just disintegrated into dust. Plus this gives me a chance to clean and lube the pointer and band selection mechanical portions.



I repainted the dial needle the correct orange.



The stereo lamp looked low hours, but since this is pain to replace, I put a brand new old stock bulb in.





I had to get creative with the missing light pipe foam. This is a roll of generic medium density EPDM foam.



I had to use two layers, but I round punched out this and stacked them.



I carefully cleaned the dial face and directly lit the stereo lamp to check if the foam seal worked. It did very well, no ambient stray light into the jewel except the stereo lamp.

The old turd of a radio is starting to clean up nicely.



As you can see you need to have a lot of work space for something like this.

OK next up is the volume control -On/Off switch. It won't mechanically switch off.



I removed the back switch off the volume/tone control assembly and found the contact slightly bent. I simply straightened it and put it back on. That fixed that. I also sprayed control cleaner into the assembly since I was there.

That takes care of the mechanical problems. Now onto the more interesting electrical ones.



This is the AM radio board. Not much too it. In the diagnostic section I found the secondary winding of the 2nd IF transformer open. The IF transformers are in these tall metal cans for shielding against radiant electrical noise. I will unsolder the entire can, then remove the insides for inspection.

Continued in next post.

Radio Repair Part I - Continued

This is the actual IF transformer assembly. Here you can see the primary and secondary windings separated by an air gap. They are not wound atop one another like you would probably see in a power transformer or audio transformer. In the plastic base are the small mica tubular capacitors as shown on the schematic. Together with the inductance from the windings and the capacitance this forms a tuned circuit which peaks at 262.5 kHz when adjusted. What you can't also see are two ferrite tuning slugs, one under each coil. They screw up and down the centre tube changing the resonant frequency. These are the slugs you adjust when electrically aligning the radio.

Back to the problem at hand. The terminals look extremely corroded. Let's have a closer look.



Well, there's the problem. However this begs the question why did the solder and copper wire oxidize so badly and only at the terminal.

Fortunately there is enough slack in the copper windings to reattach. However all that corrosion must be carefully and meticulously removed first. I can't use chemicals without damaging the rest of the assembly. It was a painfully tedious slow scraping to remove as much corrosion as I could.

What happened here is actually pretty simple. Someone on the assembly line making these used a solder with an acid core base and not rosin core. The acid will slowly oxidize the joint over time but look just fine right after soldering initially.



And here it is repaired.





You get the idea. I redid all the connections in this transformer and removed as much of the corrosion as I could. I use 3% rosin core solder and sometimes use no clean rosin flux liquid on stubborn areas. All these are safe long term on connections.

There's a high probability that even my other parts radio has the same problem so trying to find a usable part may prove to be very difficult if not impossible. So sometimes you have to make difficult repairs.

Now even though the 1st IF transformer was working I was worried it might have the same problem only to rear its ugly head later.

Sooooooooooo



I pulled that off the AM board and pulled it out of the can.



This is what I've found. Factory fresh 50+ year old solder that looks like it was just soldered. This person on this assembly line used the correct rosin core flux solder. No worries here.

This will fix the AM band.

Much more to come.

Cheers

Radio Repair Part II

I need to include a few things I've forgotten to comment about. There's a condition for a certain kind of silver mica capacitor in the IF transformers that causes either a dead AM/FM band, or poor sensitivity or even poor selectivity. Sensitivity is receivers ability to receive a weak station. Selectivity is the receivers ability to focus in on one station. Too little selectivity can result in a side or adjacent station bleeding over into the station your tuned into. Too much selectivity can reduce signal bandwidth and cripple fidelity.

Back to the point. If the IF transformer has the capacitor where it's a sheet of mica plated with silver on both sides to form the capacitor and it's squeezed by contacts in the plastic base of the IF transformer this can be a problem and those need to go and be replaced with a regular silver mica encapsulated capacitor.

What happens in these IF transformers, is with time/age and use the plating of silver is so thin on the mica sheet with contaminants such as humidity and salts from the air, the current and voltage across the capacitor will start to lift the ions of the silver around the contact and it will plate the contact. When enough of them have migrated over the connection to the rest of the mica capacitor can be lost or lost intermittently. If you hear crackling on the speakers either intermittently or all the time, there's a high probability this is the cause. On vacuum tube/valve radios because of the higher voltages, you can see this type of capacitor arcing if you can look into the IF transformer either through the bottom or top with the set running. It makes one of hell of a racket on the speaker(s) when doing so.

This radio doesn't use those kinds of capacitors in the IF transformers, they are round mica ones with leads so it's not a problem.

OK that's a good segue onto electrolytic capacitor replacements.

Unfortunately, just like automotive parts, the electronic parts industry has, over the years, been plagued especially with cheap crappy parts or even counterfeit parts that look like brand name parts. These parts are junk and to be avoided.

You can give yourself the best chance to avoid counterfeit parts by ONLY buying from places like Digikey, Newark, Mouser, Allied, etc. These places do their very best to vet the parts they buy and sell. When you buy parts from radio supply houses, parts clearance houses or E-Bay or Amazon there's no guarantee and most likely those are either plain junk or counterfeit parts, which are also junk.

The primary brands of capacitors to stick to are:

1.) Panasonic
2.) Nichicon
3.) Sprague which is now owned by Vishay
4.) Illinois Capacitor

There are other brands like Kemet that are ok, but try to stick to these when possible for your application.

If you see a no name capacitor or a funny name one, just say no.

That covers the quality and supplier topic of capacitors, now lets cover the actual electrical characteristics of capacitors.

In a perfect world a capacitor has no series resistance, no parallel resistance and no series inductance. But they do. Electrolytic capacitors especially. There are all kinds of different classes of capacitors; electrolytic's, film, disc, mica, variable plate air, tantalum, etc. Here I am just going to discuss the problems with electrolytic's.

Electrolytic's are simply two rolls of conductive foil separated by an insulator soaked in a juicy and often toxic electrolyte. How's that for a sandwich

As a result of its construction, electrolytics have a service life in hours (most often from 1000 hours to 10,000 or more hours). The life depends on the temperature of which it operates and the quality of the materials used.

Here's some of the failure modes of electrolytic capacitors.

1.) The electrolytic paste-goo dries up and the capacitance goes way down, sometimes to near zero.
2.) The dielectric insulating film between the foil sheets can break down causing shorts between the two sheets of foil.
3.) The dielectric paste-goo can eat away at the foil causing the equivalent resistance of the capacitor to go way up or the dielectric breaks down causing the delayed dielectric effect to increase resistance.
4.) The dielectric paste-goo can turn caustic and eat away at the aluminum can structure then leak out of the capacitor, not only rendering the capacitor inoperative but then eat away at the traces on the PCB board around the capacitor. This happens in equipment, consumer goods and now even happening in ECU's in cars from the 80's and 90's. Fox body guys are starting to feel this with the Ford ECU's and so are the imports of that era. I bet some thought they never would have to recap their ECU's.......

I know that was a lot for a beginner, so I'll stop there and let that soak in.

The main characteristics of interest for a capacitor for uses like this are series resistance, capacitance and parallel resistance. Parallel resistance is also called leakage resistance of a capacitor and isn't so much a factor in a bi-polar transistor radio such as this, mostly because bi-polar transistors are a current device and even a slight leakage in a coupling capacitor probably will not greatly affect the Q (quiescent) point of the bias of the transistor.

However if it's a FET radio or a vacuum tube/valve radio even a slightly leaky coupling capacitor can cause high currents in FET's and even red plate tubes causing them to fail.

To further compound the complexity of capacitors, certain test equipment can test falsely to really screw with you. I'll demonstrate this and why it's just best to replace nearly all old electrolytic capacitors when in doubt.

Let's jump in.

I've removed all the electrolytic's from the radio. I'll start testing the two off the AM board.



These two are circled in blue and green on the schematic.



This is the 50 uF capacitor circled in blue. This is the AVC/AGC (automatic volume control/automatic gain control) capacitor in an AM radio. If this were to open (dry out) your AM radio would squeal over most or all the band.

OK, onto testing. I have a little handheld capacitance meter here first. The rating on the capacitor is only 50 uF. If this is all you had, you might think this is an over achiever and is still good. At this point I wish I could insert Dr. Cox's wrong song from Scrubs.



Here's the same cap on a lab grade LCR bridge. It's a little high, but not terribly. So what's going on here and why is the capacitor still bad.

Well it's about the ESR or equivalent series resistance. Imagine there is a resistor in series inside the capacitor. Ideally the series resistance should be 0 ohms, but in reality for a capacitor like this it should be 1/2-1 ohm, 2 ohms max.

Now from the D or Q reading we can use some simple trig to calculate ESR.



Or just press the ESR button and it does it for you. So 14 ohms is way too high. This capacitor is still alive, but it's dying and very much on its way out.

So why does the handheld capacitance meter read nearly twice the rated capacitance?

I should note it isn't because the handheld is some cheap thing, I've seen way overpriced Fluke handheld multimeters with a capacitance function give the same B.S. result.

The answer is simply how it measures capacitance. The handheld meters are applying a charging voltage to the capacitor and integrating that over time to calculate capacitance. What happens when the series resistance goes up..... it takes longer to charge the capacitor and fools the meter into thinking it's a larger capacitor than what's actually connected.

The big LCR bridge is using frequency applied across the capacitor to determine is true capacitance and that method is unaffected by any resistance.

There's the reason why. Hopefully that makes sense.



Lets test the other capacitor on the AM board. This is the filter capacitor in the AM B+ supply and is circled in green.





I'm going to keep using both meters to show how easily it is to be fooled.

So this is saying it's 26.1 uF and it's rated at 20 uF. So that seems good, or does it....



Oh my it's real value is 0.246 uF. It's dead. Nobody home. Why did the handheld read nearly ok then.



Because the ESR is really high at 1.3k ohms. By sheer dumb luck the tiny residual capacitance this had left and the high ESR it had fooled the handheld into thinking it's close to the rated value.

Having this capacitor open and running in a car with all the electrical load noise would have probably made it next to impossible to listen to AM. This also could cause unwanted oscillation or spurious squealing. Since I initially tested this on a clean lab supply there is no electrical noise and the AM radio portion worked with the spectrum analyzer.

Continued in next post.

Radio Repair Part II - Continued

Next I'll examine main power supply filter capacitor. It's the big can that has three capacitors in it. They are circled in red.





Here's one cap in the can. Now this is the first time the handheld has really indicated trouble. All three capacitors in this can are rated at 500 uF.

Here's the other two sections:



not good.



and also not good.

Since we know that an old failing electrolytic will read much higher on a handheld, this thing must be nearly expired.



Yuppers, that's pretty darn low at only 8.8uF out of 500 uF. That's only about 1.8% of capacity.



This is showing an ESR of 143 ohms. For this kind of application I would want to see a capacitor with a 1/2 ohm or less ESR.

I won't bore you testing the other two sections as they are all about the same condition. I'm trying to impart a feel of what to expect.

With this input power supply filter being nearly dead, you would hear every electrical noise in your car over the speakers on AM or FM. You'd hear the ignition, turn signals, blower motor, etc as a symptom and perhaps even the alternator whine.

There's only 4 more electrolytic's to check.



The ones circled in pink must have some life left in them as they couple the audio on the L & R amplifiers, if these were dead the radio would produce no sound, maybe some hiss from the output transistors cranked all the way up, but that's about it.

Here's how they checked out.



The handheld is measuring 3uF and the capacitor is rated at 2uF. On the surface that seems pretty good, albeit high.



The bridge isn't that far off either.

Sooooooo are these good or...........



No. Not even close. 172 ohms is bad, when these should be around 1-2 ohms ESR for this size and voltage rating of an electrolytic capacitor.

But now let me throw you a further curve ball. Up to this point I've been checking ESR and capacitance. I've mentioned that the LCR bridge uses frequency to measure capacitance, but I never mentioned what that frequency is.

Till now.

Continued in next post.

Radio Repair Part II - Continued

The industry standard is 120Hz. Why 120Hz... because the North American power grid operates at 60Hz and capacitors used after a full wave bridge rectifier will see 120 Hz pulses of current, which is a common application. My HP LCR bridge can measure at 120Hz, 1kHz and 10kHz. Now healthy capacitors will operate near their rated capacitance (usually around 1% variance) at higher frequencies, but what about old failing capacitors....

Let's take a look. This is the audio coupling capacitor again.



On 120Hz, it's measuring close enough to 2.6 uF.



At 1kHz it's measuring 1.7uF



And at 10kHz it's only 0.9uF.

Where these are used in the circuit, if you change the capacitance or resistance, you effectively are changing the allowable pass band of certain frequencies, just like what an equalizer does. Now imagine an equalizer that changes automatically on its own with frequency of the audio coming through it. It would sound terrible. Well that's what's happening here with these old capacitors. Whilst they work, they sonic quality is severely degraded.

The other audio coupling capacitor is in similar condition so no need to show that.

The last group of electrolytic's to check are the ones circled in orange.



These in this circuit function as emitter bypass capacitors. In other words they shunt the AC audio signal past the biasing network of resistance allow the proper quiescent DC biasing voltages to keep the output transistors operating near their ideal current conditions.

If these capacitors start to go open then the gain falls off severely in the audio amp. Which is exactly what this radio was displaying when I powered it up as you had to nearly max the volume control out to hear normal listening level. With that I already expected these to be dead or nearly dead.



I'll be honest I wasn't expecting that. I mean they are just a pile of bones dead.



The LCR bridge says 0.0076 uF and they are rated at 1000 uF. I guarantee that tiny capacitance is from the test leads and not the capacitor.

This is the one time the handheld and the bridge were in complete agreement.



Holy cow it's no longer a capacitor but a 550k Ohm resistor.

The other one was no better and certainly no worse. I mean 0 is 0.

That concludes the learning curve on electrolytic capacitors. I know I covered a great deal (and still left a great amount out), but hopefully some will find this useful.

Last item on the bucket list was the driver transistor for the stereo lamp. The resistor in series with the emitter checked fine so I did a quick test with a pocket checker.



Now this can also be like the handheld capacitance meter. You need to be aware of the equipments limitations and how exactly it's testing the part. On the surface this says it's just fine.

I have a curve tracer, but that's a bit much to get into, if you thought simple electrolytics was a complex subject I could drown you with bi-polar solid state theory and applications. I will say this, these early epoxy domed transistors are notorious for failure with age. And this transistor is indeed bad, regardless of what the tester says.

I think that's enough for now. On the next round, I'll cover testing the new parts, assembling and testing the radio to see how well I did to correct the first round of problems and see if any new problems are now detectable.

More to come.

Cheers

kbuhagiar said:

Hello DXL, hope you are well!

Thanks again for he detailed - and entertaining - restoration write-ups. I'm finding this one especially interesting since I spent my first semester of college taking electronics technology courses, before moving into telecommunications and eventually IT. BTW that is an impressive array of test equipment you have at your disposal - do you do this for a living?

Since you seem to be so good at so many different pursuits, I'm curious if you have advanced education or training in any specific field? Or are you just naturally talented?

Click to expand...

Hello kbuhagiar,

How are you doing? I am doing well, thank you for asking. I am happy you are finding this thread interesting. I am trying to balance a fine line of giving enough information for someone wanting to try something similar, but at the same time not overwhelm someone with too much information.

I really do not do vintage repair for a living. There simply is no money in it as no one is going to spend for the labour to have something old fixed properly. I have about 40 solid hours into that radio and that's not including letting it play for a few hours a day and checking in on it. Very few people would pay lab rate to have the radio fixed properly.

I have a bachelors degree but none further. I did have a unique job for several years as a metrologist in a NIST comparable environment and that imparted a really good background in accuracy and testing. Then I worked as an automotive testing and design engineer for awhile. In those jobs it was sink or swim, there's no mentoring. You have to find answers on your own and think out of the box most days. In the automotive world it's sink or swim, no life vest.

Now I do have an insatiable desire for knowledge. My better half and I have collected over 400 hardcover books on general engineering subjects. Many of them are older books; from the 60's all the way to modern day. I will say this there is much more information in the older books than what is imparted in newer texts. I'd go so far as it's a dumbing down of American education. You couldn't pay me to go back to college.

I do like to read and experiment for myself. Now I wouldn't mind teaching it, that's for sure.

Anyway I hope you are having fun

Cheers

Radio Repair Part III

This section will deal with the replacement parts.

On the topic of the stereo lamp driver transistor.



The old epoxy domed one was indeed faulty and I found a suitable one from my parts selection. All these transistors in this radio are silicon and there are no germanium ones. Now the Sams schematic is in error and shows a couple of them as germanium but they are silicon, so you can of have to use instinct to find the errors.

Germanium transistors will have a Base to Emitter forward voltage drop around 0.2 volts. Silicon will have around 0.6 volts.



Just to rehash, this is the old lamp driver and according to this it's ok. But it's not testing under a heavy load and that's where the problem lies. In this case it's only 6.1 mA, that lamp even with the 10 ohm resistor will draw well over 100 mA. A curve tracer will confirm the collector current falls off under a heavier load.



This one tests nearly the same. This kind of tester is fine for catching dead devices, it doesn't do a comprehensive test. Again it's all about knowing exactly how your test equipment operates.



On the topic of other replacement parts, in this case capacitors, with the current supply shortage, finding the perfect part replacement was a bit of a challenge as the major parts houses I listed earlier were really skint on inventory and I didn't want to wait 16-30 weeks, in this case I settled for electrically critical criteria but had to settle in some cases for case size or just pricing. In this case these capacitors replace the audio coupling caps from earlier. These were over 7 dollars each.

I will not bore you with more capacitor testing pictures. The handheld meter does agree with the LCR bridge on the new capacitors. My previous pictures were to demonstrate how easily it is to be fooled by readings with failing parts.



These are to replace the main power supply filter capacitor can. Here's where you have to think out of the box a bit. There are the equivalent of three capacitors in that can. Finding a replacement can that fits in the radio will be next to impossible. The only multisection can capacitors that are still being made are mostly reproductions for old tube gear and those are physically much larger and higher voltage.

About replacement capacitor and voltage ratings. Whilst you never want to put a lower rated capacitor in place of a higher rated voltage one, on the flip-side you do not want to put a really high rated voltage capacitor in place of a lower rated one. The reason is electrolytic capacitors occasionally need to 'reform'. When an electrolytic capacitor is dormant for long periods the dielectric insulator starts to break down in some areas. When you apply voltage to a dormant electrolytic capacitor it can heal itself. But you need to be ~1/2-3/4 of the rated voltage to do so.

So you wouldn't want a 100 volt capacitor for the 16 volt filter capacitor sort of thing. Just something to keep in mind.



Here's how I mounted them. This worked really well. I used low acid RTV to effectively glue the three together. This makes them very stout and resistant to vibration.

You want to use low acid RTV, typically the Ultra series RTV (in this case Ultra Black) are low acid. Some other RTV's cure by releasing acetic acid and the acid will eat through some materials and you never want to use that in electronic equipment. Basically if the RTV has a strong odor, don't use it as glue.



OK, so I've replaced the electrolytic's on these boards and I reinstalled the AM board.



The balance control is added. This was cleaned with Deoxit.



Cleaned and installed the front support plate.



This rheostat is the stereo separation adjustment. It's adjustable from the exterior of the radio.

Continued in next post.

Radio Repair Part III - Continued

That's all put back together and it's starting to look less hopeless and more like a radio.



All that's left is the volume/tone control assembly and audio amplifier sections.



But first, the old wire is, well, old, and needs to go. It just cracks if you bend it on a tight radius.



With that I added new GXL wire to the radio. Now I did add a remote wire as well for the other part of the sound system enhancements.



And for some reason the general illumination lamp wire was cut on this radio. Why on earth would someone cut that. I mean the wire is old and needs to be replaced, but I don't get why hack it.



Anyway they do sell the lamp terminal crimps like the factory for making your own new wire assemblies.



Much better. Now I do eventually use a soft white bright LED. So this incandescent doesn't hang around for too long.



These were the last of the electrolytics to replace.



At this point you'll have to forgive me as I was so engrossed on reassembly I forgot about the camera. I do pick up again after realizing I'm SUPPOSED to take pictures.

Here's the events that transpired without pictures. I powered up the radio driving just 8 ohm monitor speakers and what an improvement. At just 1/4 volume it's really loud in the lab. AM is alive albeit not really sensitive as it should be, I'm counting on it just needing an alignment after I was poking around mending that IF transformer.

FM works well, the stereo lamp is bright on stations, so that's done. I hit all my initial problems and they are remedied.

However, 3 new problems were now evident.

1.) The right side channel is noticeably louder than the left.
2.) The sound is really flat, tone control works, but being just a tone control it doesn't address the noticeably flat spectrum.
3.) The left side speaker sometimes doesn't work and if I press on the output transistor is works. That's an easy one and we'll tackle that last as I'll have to replace the output transistors.

Continued in next post.

Radio Repair Part III - Continued

To address the first problem, I set up the stereo generator with both test tones on so the input signal was equal on both speakers. Then with a steady signal I took some measurements to figure out why the right side was louder. I could be partly the stereo decoder, it could be the volume/tone controls, and or it could be the amplifiers as the gain may be higher in one than the other.

Didn't take long to find the problem.



These are ac voltages from the test tones entering the volume/tone control circuitry. To make sure the amplifiers weren't dragging the signals down I removed the disc capacitors with the red X through them. From this you can see at the balance control I only have a R & L difference of ~ 1%. At the input to the amplifier section (out of the volume/tone controls) I have roughly 45% difference between R & L. Here's the problem area.

I removed all the film capacitors from this area to check for leakages, although film capacitors rarely fail. I checked the resistors and they all checked ok. Although some of the film capacitors were crumbling. So they will get replaced because they are physically falling apart.

The problem was the actual volume controls themselves. Das ist nicht sehr gut. That's not good because they are very specialized controls for this particular radio design and not something that can be just purchased.

Looking at the schematic you can see the end to end resistance is supposed to be 225 k Ohms with a tap at 75 k Ohms.

Here's my results.

Right side volume control potentiometer:

Full range Ohms: 319 k (supposed to be 225 k)
Tap to audio source end Ohms: 242 k (supposed to be 150 k)
Tap to ground end Ohms: 77 k (supposed to be 75 k)

Left side volume control potentiometer:

Full range Ohms: 211 k (supposed to be 225 k)
Tap to audio source end Ohms: 131 k (supposed to be 150 k)
Tap to ground end Ohms: 81 k (supposed to be 75 k)

So you can see if you add the tap to one end and the tap to the other end resistance that should total close to the full range ohms of the volume control.

Lets have a better look at the volume control and where the tap is at.



You can see the four stacks of controls amalgamated into one unit. The front two sections are the tone for the R & L side, the rear two sections are the volume controls for the R & L side. You can see the tap connection atop and to the one side. The wiper and end contacts are on the bottom on all 4 sections.

I do have the spare radio so it might be a donor candidate. With that I took it apart to measure its controls.



Well that took bit of time, but here's the parts radio results.

Right side volume control potentiometer:

Full range Ohms: 192 k (supposed to be 225 k)
Tap to audio source end Ohms: 136 k (supposed to be 150 k)
Tap to ground end Ohms: 57 k (supposed to be 75 k)

Left side volume control potentiometer:

Full range Ohms: 260 k (supposed to be 225 k)
Tap to audio source end Ohms: 1621 k (supposed to be 150 k)
Tap to ground end Ohms: 99 k (supposed to be 75 k)

I think the parts radio is actually worse off.

Well..... !

Here's where you have to think out of the box again. I have accepted I'll probably never find a decent donor part so I have to work with what I have. What I can do is try to normalize both R & L potentiometers to the least resistance portions of each other.

Let me explain that, since each volume control is divided (unequally) into two sections by the tap, I can focus on the resistance from the tap to the audio source end connection on both volume controls and whomever has the higher resistance will be reduced to match the lower resistance of the other volume control. And you do the same with the other side of the tap.

A diagram may be better:


Here I've drawn both volume control potentiometers atop the page. The blue ink represents the measured resistances of the two sections of each volume control.

Let's look at the lower section from the tap to the bottom terminal. On the right it's currently 77 k Ohms and on the left it's currently 81 k Ohms. The idea is to reduce the 81 k Ohms to 77 k Ohms by adding a parallel resistance. Then you do the same for the top portion. In the end both volume controls will measure the same across their terminals. Now this isn't exact as wiper in the potentiometer will not track exactly, but it will be close enough to improve this situation.

The equations at the bottom are just simple parallel resistance calculations to determine the extra resistors I need to add to each volume control.

After adding the resistors this is the new measurements based on % of volume control.



Pardon my handwriting. Aft and Fore refer to the volume controls as positioned in the radio from the front to back. You can see the L & R now track each other pretty decently till about 63% of travel on the volume knob. And after running the radio, anything past 50% is just distortion anyway and bloody loud as the amplifiers are clipping.

So this will work just fine.

Next I'll tackle the second problem of why is the audio so flat. Looking at the broader picture as in not just the radio, but the whole system as installed in the car, I am not taking into account the original type speakers used. Chances are they would have been paper cone high efficiency speakers (greater than 100 db per 1 watt at 1 meter). New speakers today aren't that efficient, mainly because there's no need to make them efficient as audio power is aplenty these days. But in the vacuum tube/valve era and early transistorized radios power output was expensive and heavy, so it made sense to make efficient speakers to make the most out of the little the radios were able to produce in terms of power output.

It is very possible the original speakers greatly accentuated the mids and highs and what seems like a flat response to me using a high quality monitor type speaker isn't what you would have heard in the car when the entire system is considered.

That begs the question, is this flat response built in or a fault?

Only one way to find out and that is model the volume/tone controls in SPICE and run an AC analysis and look at the frequency response.



I have the original circuit here. Just need to draw this in SPICE.



For those wondering what this is about, SPICE is circuit emulator. It allows you to test your circuits under different conditions to see how well they work before you even build a prototype. It's handy.

SPICE stands for Simulation Program with Integrated Circuit Emphasis, although you don't have to use integrated circuits, you can model discrete components as well like this simple volume and tone control.

The program I am using is freeware and is it's call Quc's. I have a couple computers loaded with Ubuntu on them and Quc's is in the Ubuntu software store for free. Just search download in the Linux free app store. Last time I checked Quc's did not run natively on Mac OS. You'll have to use Wine to run Quc's on a Mac. It does run natively in Windows as well. Although the last time I used Windows the operating system was XP 2nd edition. I probably wouldn't recognize the latest edition of Windows if I tripped over it and believe me I'm grateful.

Ok, back to the simulation. Here's how this is setup. On the input I am applying a 1V AC signal on the (Audio In) that varies from 0 to 15 kHz and it'll plot the output at (Audio Out).

In this circuit rendition, you'll notice I have the tone control set at full base (full anti-clockwise) and volume at 1/3.



This is the audio spectrum sent to the amplifiers in the radio. From about 3 - 15 kHz it's pretty darn flat. Let's focus on the lower end and enhance that.



So in full bass setting that's exactly what it's doing, allowing more bass signals to get through and attenuating the rest of the spectrum. Still pretty flat otherwise.



Here I've set the tone control in the middle. Let's see the result.



The bass stays more or less the same but the rest of the spectrum is elevated. So this would boost the mids and highs together. Still pretty flat.

Continued in next post.

Radio Repair Part III - Continued



Now I've set the tone control to full treble (max clockwise). Let's see what this does.



Well all it really did was attenuate the bass. Looking a little closer.



It just killed the low bass chiefly.

And this is exactly what I'm hearing in the lab using the monitor speakers. This simulation proves a couple things.

1.) This is how the radio was designed and is not a fault.
2.) The cars original factory speakers were such that they probably further enhanced mid and highs for the tone/bass control to really only affect the low frequency range.
3.) Since even if you had the factory speakers, they would be so dried out and decrepit by now you'd have to use an aftermarket speaker which more than likely wouldn't have the sonic characteristic as the originals, so you would need to use an external equalizer and amplifier.

Spoiler:
The upgraded total sound system is in fact using equalizers, amplifiers, fader control, antenna injection of ancillary audio input and very probably reverberation along with this 1968 head unit. But the head unit must work extremely well first otherwise all the rest is meaningless if you put a garbage signal into it.



Back to the head unit. I replaced the disintegrating film capacitors on the volume control board with Nichicon film capacitors (yellow ones).



And here we are back together..... again, for another test. This time we are going to use one of the equalizers and amplifier that's going in the car. Now in order to do that. I need to convert the low output impedance of the radio back to a higher impedance to feed into the equalizer. So I have two line matching transformers, one for each channel.

I temporary put the front plastic cover on this radio, I will use the plastic cover and knobs from the radio in the XL as those are in really nice shape. I put the cover on, because the LED back light is otherwise really bright.



It's messy but it's working really well at this point and it still needs an electrical alignment and the power output transistors changed.

Both left and right speakers track in volume up to stupid loud levels, so that's fixed and of course the EQ corrects the flat response for these speakers.

Since the proof is always in the pudding. I offer the pudding.

Here's a video of it playing at this point. http://galaxieworks.com/sites/forum_videos/1968 Radio Play.mp4

Mind you I'm using my still camera and it's limited video feature. Your playback may vary in quality, but in person it sounds pretty darn good for once beat up and left for dead radio.

More to come.

Cheers

Speakers - Rear

I thought I would take a break from the radio portion, I know it's boring to many, and focus on the speaker situation. My particular XL just has the rear speakers and none in the door. The plan is to add the door speakers and I think I've got a decent solution to that. The goal is to look as either factory or 60's era as possible. But for now, I'll focus on the rear speakers.

Here's what's in the car now.





They look very old and very paper. They sound terrible, but the current 1968 AM-FM stereo radio is to largely blame.

Plus lest not forgot the exquisite installation someone did on these in the past.



Ohhhhhhhhhhhhhhhh brother.



Good grief, what a joke. Well I had enough of a laugh over these time for these to go in the trash, where they belong.



The paper just disintegrated when you touch it.



This is the back side of them. They look like some 1990's Circuit City bargain line.

Since the amplifier I am going to be using is a 4 x 50 watt rms, I decided to pick a 6x9 that was similarly rated.



I bought these.



At least on paper these seem to be a good match for the amplifier.





Just to keep in mind on this sound system I am putting together, I am going for quality not over the top quantity.

Continued in next post

Speakers - Rear - Continued

Now for grills I wanted some more 60's looking than a garish looking modern grill. So I found these for a 60's Mustang.



These look the part.



Just a compare and contrast. Yuppers the right one looks like it belongs in the car.



Now the Mustang reproduction grills are all metal, no cheap plastic, so I'm really digging them. Plus it has studs which makes installing the speakers even easier as I don't need 20 foot arms to reach around and hold a screwdriver from the boot in the car where I'm securing the nut. I used Locktight on the studs.



However Rockford Fosgate has quality problems already in that their 6x9 holes are slightly off. I thought maybe it was the reproduction Mustang grills but the old speaker mounting holes are right on target. So I had to file all the mounting holes on the new speakers a bit. More wonderful aftermarket mantra "nothing works as intended".



Easy peasy to mount with these grills.





I am happy with the looks, although the previous owner who installed black carpeting over the package shelf hardboard should have gave the carpet a haircut first.

I can't really judge how these sound yet, because, well, they sound exactly the same as the decrepit old speakers. You know your radio is in dire need of repair when that's the case.

More to come.

Cheers

Radio Continued

It's been a while since I've updated this. I have had the chance to finish the radio and acquire some more of the parts for the complete sound system planned for this '68 XL.

Without further ado.



The output transistors needed to be replaced because the left channel would still quit working and it was the transistor, if you pressed on it, it would start working again.

You can see the left channel mounted above the volume/tone control. It's in the shape of a TO-3 package, but it's not. I have two TO-3 package transistors next to it as an example.



This is a first, even for me, this is some weird TO-3 prototype that made it into production or is the prototype that was installed. It wouldn't be the first time a prototype escaped and I ended up with it. The actual transistor crystal is the blur disc and it's bonded to the plate, but the bonding is cracked, hence the intermittent contact.



Now the astute will notice a spot where rationally the right side output transistor would go. However, for whatever reason it's not there and installed in the back panel of the radio. After doing the testing on this radio I have a good idea why. This is a direct drive class A amplifier and they get warm. I think it was deemed too much heat to have in the front of the radio, so they put the other channel on the back to better dissipate the total heat of the output amplifier in the radio case.



This is one of my spare '68 AM-FM stereo radios and it has a TO-220 transistor in a TO-3 mounting plate.



New output transistors through Mouser, so there's a high probability these are high quality genuine parts and not cheap knock-offs.





Removing the old TO-220 off the TO-3 mount in the spare radio and installing the new NPN transistor for this radio I am working on.



This is the right side channel transistor mounted in the back plate. This of course was changed to a new TO-220.



Just to summarize, these are all the parts that were replaced in this radio to make it work for the long future.



Even the radio seemingly works, I want to do a full electrical alignment on the AM, FM and FM stereo decoder to see how sensitive and selective the radio is. So this involves injecting signals and looking at signals at different points in the radio.

I had some pretty interesting results.

Continued in next post.

Radio Continued-1



This is the FM IF section. This is the bandpass of the entire IF chain just before the FM detector. The FM IF bandwidth should be 150 KHz wide. I fiddled with all the adjustments (ferrite screws in each of the FM IF transformer cores) until I was rewarded with (or so I thought) a good looking spectrum. You can see the middle 10.7 MHz marker and the +/- 75 KHz side markers on the outer edge of the slopes. It's textbook.



This is the detector waveform, again close to textbook, just a tad off but it's as close I could get it and centre the 10.7 MHz marker.

However whilst all this was textbook adjustments the real world had other ideas. This killed the sensitivity of the FM. It would only pick up nearby stations. Using the signal injection, it was only picking up stations with 50 or more uV and those had static and the FM stereo decoder wouldn't work until closer to 100 uV signals were injected. For reference you should be able to hear FM stations, even if not perfectly clear with around 10uV on the antenna and it's OK if the stereo decoder doesn't engage.

With that I went with Plan B and adjusted the FM IF and detector for maximum sensitivity and good separation, so all by ear and not pictorial scope displays. I was able to get decent FM (Monaural) all the way down to 7 uV and stereo reception around 30 uV. That's a world of improvement.

OK, so why bother with textbook examples if adjusting for the peak works. In this very particular case, peaking the FM IF still allowed enough bandwidth through to allow stereo reception.



Here's a breakdown of the typical FM station bandwidth (all 75 KHz). Here's a little explanation of it. If you had an old monaural pocket FM radio it would utilize the first 15 KHz of the total information on a single station spectrum because the left + right audio information is contained there. This also applies to stereo receivers with too low a signal to activate the stereo decoder, all they use at that point is first 15 KHz. The 19 KHz pilot is what the stereo decoder looks for to activate. If the signal is too weak, the decoder turns off and you can still hear the FM station, it just will be monaural. Also there are still a couple FM stations, mostly talk radio that do not transmit in stereo, so there's no 19 KHz tone present and the stereo decoder stays off.

When a strong enough stereo signal is received, the stereo decoder needs the L-R information (from 23 - 53 KHz) along with the L+R information (normal monaural information) to mathematically produce the separate Left and Right channels. I'll get into that more in a bit.

The SCA portion of the FM radio station is the piggy backed station that use to be very popular well before even cassette tapes, never mind the internet as this was an original specification of the FCC to define FM stereo broadcast and the adherence to it. SCA like I mentioned at the beginning of this thread is monaural mid fidelity station that use to play commercial free musac for places like department stores, dental offices, etc.

I do believe there are still a couple of stations that have the SCA active with another broadcast. Way back when, you were supposed to rent equipment to listen to it, but it's easy to add a PLL to a standard FM radio and listen to it with little to no cost.

The RDS section was added a couple decades ago, there was juuuuuuuuust enough spectral space to add a carrier for digital information. This is what carries the information on car radios that tell you what song is playing. Obviously the station is sending this information along with the audio.

Now back to this little Ford AM-FM stereo radio. For the FM stereo to work the bandwidth of the IF doesn't need to be textbook 150 KHz/2 wide. It only needs to be 53 KHz wide for the AM-FM stereo portion to work. So if peaking the FM IF results in more sensitivity but narrower, it will work better so long as the IF is at least 53 KHz x2 wide.

At this point, the FM radio front end was done, but I needed to adjust the stereo decoder. But first here's a quick review of how stereo actually works. This is from Sencore SG-165 manual. It's simple to understand so why not have a look.








It's a simplistic explanation, but that's all one needs to do the alignment. Designing your own would be beyond the scope of this thread.




At this point, I used 2 channels on the scope; one for left channel output and one for right channel output, then just injected the stereo signal on the antenna (basically acting as a radio station but I can control the left and right audio being transmitted).




I only have two pieces of equipment that can do this. This is the Leader. It is currently set to send a 1KHz audio tone on both L and R channels in stereo (19 KHz pilot active by blue lamp). The Leader only transmits on one equivalent FM radio station which is 100 MHz.



The radio is tuned to 100 MHz and the stereo lamp is on, which it should be with the 19 KHz pilot signal on the Leader.



OK, you can see both channels (L & R) of the radio on the scope. And you can see the vague hint of stereo choppiness in the sine waves. That's the 38 KHz chopper working in the stereo decoder. Humans can't hear it, but I bet it drives dogs and cats insane if the tone controls, audio amplifier and speaker are allowed to reproduce that high.

Continued in next post.

Radio Continued-2



So here's a good test of the stereo separation the radio has. The Leader is set for L channel only transmission.



OK, so the radio is doing a pretty good job, there is a little on the right side, but it's mostly attenuated.



Checking for stereo balance by just sending the R channel now.



Again pretty good for a 1968 car radio. All this is fine tuned by adjusting the ferrite cores in the stereo decoder and the separation rheostat.



Now to transmit the L & R channel but with the pilot off, meaning no stereo, just monaural.



Stereo lamp off.



Now you can see the same sine wave but it's a smooth sine wave in monaural.



Since I have the Sencore I can use it to double check the Leader.



Continued in next post.