radiomuseum.org

 

Looking for Gorler 312-2427 FM front end or SF8636 n channel

Moderators:
Ernst Erb Jacob Roschy Günther Stabe Bernhard Nagel 
 
Please click the blue info button to read more about this page.
Forum » Valves / tubes - Semiconductors » Semiconductors » Looking for Gorler 312-2427 FM front end or SF8636 n channel
           
Gast Gast
 
 
D  Articles: 0
Schem.:
Pict.: 0
22.Mar.16 21:02

Count of Thanks: 5
Reply  |  You aren't logged in. (Guest)   1

The Gorler front end in my SAE MK VIB fm tuner is defective. I'm trying to find a replacement front end or rf amplifier jfets as listed in the title. I've not been able to find a sustitute listed for the SF8636.

 

Thanks,

John

Andreas Steinmetz
 
Editor
D  Articles: 638
Schem.: 81
Pict.: 39
25.Mar.16 20:43

Count of Thanks: 5
Reply  |  You aren't logged in. (Guest)   2

Hi John,

I think I can help:

Though I don´t know a Görler 312-2427 FM front end, I suppose that it is similar to the well known 312-2433. I am just looking onto a 312-2433, manufactured 8-1974, and there I can see two transistors BF245B, not any SF8636. I also remember that my varicap-model, the 312-0522, which I bought one or two years later, has no SF8636 but BF245 transistors, too. I am really sure about this, but I don´t remember if they were BF245B or BF245C types. So it seems that Görler changed the transistor types from SF8636 to BF245B or C in the later production, without change of any resistance-value. Additionally we know that Görler preselected the FETs for minimum noise.

John, please control the resistors and capacitors around the defective device and then replace the defective transistor by a BF245B or C of good quality. (The C-type will produce a bigger drain-current than the B-type and perhaps a bigger signal gain.) Be careful because of the pinout, it could be different! I´m sure you´ll be successful repairing the very good Görler front end.

Heribert Jung
 
Editor
D  Articles: 910
Schem.: 12337
Pict.: 5295
25.Mar.16 20:40

Count of Thanks: 6
Reply  |  You aren't logged in. (Guest)   3

Hi John,
during a program change you could write as a guest in the Forum. This can usually only members, because not every post is factual as yours. We do not have "staff" to check whether the text is sensible and humane in order before we could give it free. Guests enter via "Contact" on the home page.

I am not able to answer your question, because I don't know which model is SAE MK VIB. Did you find it in Radiomuseum.org?

I can find the SF8636  2 replacement parts are listed: BF244 ; BF245.

The best way to get contact to RM-members is, to become a member. As member you can create a new model page, if your model is not in RM. There you can upload pictures from the tuner as well as from the part you need.

Regards,
Heribert

This article was edited 25.Mar.16 21:02 by Heribert Jung .

Bernhard Nagel
Bernhard Nagel
Officer
D  Articles: 1436
Schem.: 2014
Pict.: 5675
26.Mar.16 02:11

Count of Thanks: 5
Reply  |  You aren't logged in. (Guest)   4

Hello John

the model Mark Six (Mk VI) without "B" we have listed in RMorg. I think, the differences are minor, it also shows a Görler (Gorler) front-end. The hints of Andreas and Heribert given above for replacement FETs like BF245 are perfect, the chance to re-activate your Gorler FM-front end is very good.

BTW, unfortunately these answers will not reach you direct, since you are not a member in RMorg. So I hope you have a look into the forum occasionally.

Regards
Bernhard

This article was edited 26.Mar.16 02:12 by Bernhard Nagel .

John Vogt
 
 
USA  Articles: 17
Schem.:
Pict.: 5
12.Apr.16 03:16

Count of Thanks: 4
Reply  |  You aren't logged in. (Guest)   5

Hi Andreas, Heribert & Bernhard,

 

Sorry it took so long to thank all of you for your help & advice as I received my user name & password only today. 

The information that the BF245 replaced the SF8636 in later production units was the information I needed.

The BF245 is scarce here in the US, I had to order replacements from a dealer in Britain, hopefully the replacement will work properly.

My other issue with the Gorler front end are 3 cracked ferrite slugs that were damaged by a previous repair attempt by another technician. I would appreciate any suggestions in finding replacements.

Regards,

John

 

This article was edited 12.Apr.16 03:16 by John Vogt .

Andreas Steinmetz
 
Editor
D  Articles: 638
Schem.: 81
Pict.: 39
13.Apr.16 23:02

Count of Thanks: 1
Reply  |  You aren't logged in. (Guest)   6

Welcome John,

the 3 cracked ferrite slugs could really be a big problem. I don´t know any source. Somewhere in the www I read that the original slugs can´t be replaced by other types. But never say never.

The best solution were if you could remove all the parts of every defective slug out of the coil bodies, collect them and then conglutinate (is that the correct english expression?) them with a Cyanacrylate-glue for example. If there is not too much material missing, you should be able to do a perfect re-alignment of the RF coils (be careful to avoid a pseudo-maximum). The IF coils could produce more problems because missing material will influence coupling and bandwidth of the IF filter. Probably it could become necessary to increase the capacitors of the IF filter. But these things should better be done by a specialist.

If all the above hints don´t help, you could try to buy a used FM frontend via ebay for example, or try to get the original slugs. The last way is to look for alternative slugs. Mind that the RF coils are working at about 100 Mc and the IF coils at about 10 Mc! The slugs must be made of a suitable material, and the Al value should match!

Good luck!

This article was edited 13.Apr.16 23:12 by Andreas Steinmetz .

John Vogt
 
 
USA  Articles: 17
Schem.:
Pict.: 5
14.Apr.16 05:14

Count of Thanks: 4
Reply  |  You aren't logged in. (Guest)   7

Hi Andreas,

Thanks for the welcome.

I'm also under the impression that the slugs need to be an exact replacement. I've also read that gluing the slugs back together with cyanacrylate will change their magnetic properties preventing them from working properly although I don't understand why that would be.

Two of the slugs are stuck so removing them without damaging them further is a real problem. I don't know what I could use to free them so they can be safely removed. The third can still be adjusted so I'll leave that one alone.

I have a second SAE MKVIB tunerthat works well and the ferrite slugs in the front end of that unit match the position of the slugs in the defective front end so the broken slugs are probably not far off from their optimal position. I'll know better once I install a replacement FET.

I have ordered both BF245B & BF245C Jfets. Which version would you install, the "B" or "C" version?

Thanks again,

John

 

 

Andreas Steinmetz
 
Editor
D  Articles: 638
Schem.: 81
Pict.: 39
16.Apr.16 23:10

Count of Thanks: 4
Reply  |  You aren't logged in. (Guest)   8

Hi John,

step 1, the FET: I would install the B version, because it was originally implemented in my late 312-2433 FM front end. If you don´t have the schematics of your 312-2427, you should look at the 312-2433 schematics. Probably the differences will be insignificant. Unfortunately the FET´s connections are not labled in the original schematics and I am not sure if the BF245 is symmetrically constructed. So please mind that the source of an n-channel JFET is usually connected to a more negative potential than the drain.

After installing the BF245B you can calculate the resulting drain current separately for every stage based on measurements (without antenna signal) of the voltage drops across the resistors, which are mounted outside of the metal case. You have one big advantance: the second fully working SAE MK VIB fm tuner! So you just need to compare the voltage drops between the two tuners to decide if the tolerances are o.k. or not. I have no values but I would say: If the differences are lower than 10 %, then it´s o.k.. If the voltage drops are much more than 10 % lower, then you could try to install a BF245C type (if the drops are much larger, then you could try a BF245A). Alternatively you could slightly modify the responsible resistor(s).

Btw: Which FET was defective? The one in the RF stage or the one in the mixer?

After the FET is changed successfully, let´s go on with step 2, the slugs. Until then it could be a good idea not to change anything with the slugs. Perhaps the tuner then already works resonably and just small changes are necessary.

This article was edited 16.Apr.16 23:44 by Andreas Steinmetz .

John Vogt
 
 
USA  Articles: 17
Schem.:
Pict.: 5
17.Apr.16 17:43

Count of Thanks: 4
Reply  |  You aren't logged in. (Guest)   9

HI Andreas,

Thanks for the help & suggestions.

I unfortunately don't have the schematic for the 312-2427.

The defective jfet appears to be the first rf amplifier as the gate is connected to the antenna coil. What I assume to be the source was reading about 2V lower than the properly funtioning tuner. I removed that jfet tested it with my transistor checker which indicated it was bad. The second rf amplifer is another SF3836 and in checking that I was able to determine the gate and that it checked good.

What I assume to be the drain of the first SF8636 is connected directly to the gate of the second SF8636.

I should receive the BF245B's in the next day or two and I'll let you know what the outcome is.

I've attached a pix. of the top of the tuner. Note this model has 2 BC237B that are not on the 312-2433. One lead of the first SF8636(source?) is connected to the collector of a BC237B through a 330 ohm resistor.

Thanks again!

John

Attachments:

John Vogt
 
 
USA  Articles: 17
Schem.:
Pict.: 5
19.Apr.16 19:55

Count of Thanks: 3
Reply  |  You aren't logged in. (Guest)   10

Hi Andreas,

I received the BF245's and installed a BF245B as suggested. I had to make an educated guess as to the source & drain connections but the spec sheet for the BF245 indicated the source & drain were interchangeable.

In any event, the sensitivity improved from about 70 uV to 3 uV for -30db quieting.  A big improvement but not as good as the my other SAE MKVIB & MKVI tuners, their sensitivity meets spec at 1.6 uV.

Wanting to try and improve the sensitivity further I next tried to free the two cracked slugs that were stuck with isopropyl alcohol and later freon. I had no success with either of those methods. I might try freeing them with silicon lubricant and also heating the slug with my iron.

I ordered a few rf coils from Coilcraft here in the States. Unfortunately the slugs used in these coils are fractional and not metric with the closest being #6 (3.5mm) instead of about 4mm. Also I don't know how the magnetic properties of the replacements compare to that of the originas.

What further suggestions might you have in repairing the coil slugs and how the jfet has been connected? The drain is connected to the gate of the second SF8636 and the source is connected through a 330 ohm resistor to the collector of a BC237 mounted on top of the tuner box( see jpeg above).

Thanks again!

John

This article was edited 19.Apr.16 19:56 by John Vogt .

Andreas Steinmetz
 
Editor
D  Articles: 638
Schem.: 81
Pict.: 39
19.Apr.16 22:35

Count of Thanks: 1
Reply  |  You aren't logged in. (Guest)   11

Hi John,

your two last postings are very interesting. Unfortunately I don´t have much time at the moment. Could it be that the drain of the first FET is connected to the source (instead of the gate as you wrote) of the second SF8636? Please have a look at the schematics of the Görler varicap model 312-0522: Could it be that your 312-2427 has a similar input stage with a FET cascode like the 312-0522 does? The circuit with the two external npn transistors (delayed AGC) seems to be the same, too (except of the additional 330 Ohm resistor). It would be very helpful if you could draw a little schematic diagram of your input stage and post it here until I´m back.

Andreas Steinmetz
 
Editor
D  Articles: 638
Schem.: 81
Pict.: 39
20.Apr.16 11:23

Count of Thanks: 4
Reply  |  You aren't logged in. (Guest)   12

Hi John,

though I really understand that you wish to see positive results now, please don´t modify anything with the ferrite slugs at the moment. We have to finish step 1 before we go on with step 2. Step 1 is finished when we are sure that the FETs are working properly. That especially means that the working points (DC voltages and drain currents) are o.k.. For that a little schematic diagram (it doesn´t matter if it is self made and of non-professional quality) of the input stage would be very helpful.

Btw: Please tell us which of the slugs are cracked or can not be adjusted.

John Vogt
 
 
USA  Articles: 17
Schem.:
Pict.: 5
20.Apr.16 14:54

Count of Thanks: 6
Reply  |  You aren't logged in. (Guest)   13

Hi Andreas,

Not to worry, I won't make any changes to the slugs as you suggested. The stuck slugs are in the coil for the input stage and what I think is the mixer. They are marked "Ant" and "Rf2" on the SAE schematic.The SAE schematic does not give the schematic of the Gorle howeverr, it is only shown as a box marked "2 fet 4 gang tuner".

I think your right about the two outboard transistors being an AGC stage as the layout and connections of the 312-2427 are strikingly similar to the 312-0522.

I'll work on drawing the schematic after removing the tuner from the chassis.

Regarding the pinout of the SF8636, I reconnected my Peak Atlas Semiconductor Analyzer to the second SF8636 and did verify the gate.

I think I was able to determine the pinout of the SF8636. If you look at the picture of the external connections for the 312-0522 there is an SF8636 used for AFC connected to the wiring strip. Comparing the picture to the schematic, the source is connected to a 15K resistor, the gate to a 10n capacitor, drain to a variable resistor(?).

So we have with the flat side facing up and moving from left to right: S   G   D                                  

Assuming the pinout above is correct, it appears the replacement BF245 has been connected properly.

Thanks again,

John

 

 

John Vogt
 
 
USA  Articles: 17
Schem.:
Pict.: 5
26.Apr.16 21:26

Count of Thanks: 3
Reply  |  You aren't logged in. (Guest)   14

Hi Andreas,

 

I finally had some time to draw up a schematic that I've attached. Hopefully I drew it correctly, let me know what you think.  I also measured the voltages for the six transistors on both the defective tuner and a properly functioning tuner as follows:

                        Defective Tuner        Functioning Tuner

 (1) SF8636                                                                  

         Gate          -12.93V                        -12.91V                        

         Drain         +01.47V                       +01.17V

      Source         -11.71V                         -11.97V

 (2) SF8636

        Gate         +01.47V                         +01.17V

       Drain         +11.16V                          +11.32V

    Source         +00.08V                          -00.32V

BF245

       Gate         +00.00V                          +00.00V

       Drain         +11.02V                          +10.90V

    Source         +03.61V                          +04.11V
 

(1) BC237

             C         -12.98V                           -12.94V

             B         -12.40V                           -12.38V

             E         -13.08V                           -13.04V

(2) BC237

            C           -12.40V                          -12.38V

            B           -12.55V                          -12.53V

            E           -13.08V                          -13.04V

One interesting thing I noticed on the functioning tuner was that the connections of the of gate & source on the first jfet were reversed. I think a different jfet was used in place of an SF8636 (maybe a BF245?) but I couldn't see any marking on the case. Also, there was a 100 ohm resistor added in series with the small choke and 1n capacitor connected to the gate and drain.

The voltages on the second SF8636 are off by more than 10% so maybe that jfet also needs to be changed?

Your thoughts?

 

Thanks!

John

Attachments:

This article was edited 27.Apr.16 21:39 by John Vogt .

Andreas Steinmetz
 
Editor
D  Articles: 638
Schem.: 81
Pict.: 39
26.Apr.16 23:11

Count of Thanks: 4
Reply  |  You aren't logged in. (Guest)   15

Well done John! I know that was much work, but we can see much more clearly now. Your schematic is not free of mistakes, but that is quite normal. Here some short hints:

  1. Your 312-2427 definitively has the same FET cascode incl. the AGC circuit like the 312-0522. The only difference is that your circuit is fed by a +12V/-13V power supply whereas the 312-0522 circuit is fed by a single +24V power supply and that there is a 330 Ohm resistor in the source connection of your circuit. (The cascode version should have about +6 dB more gain than the normal input stage of a 312-2433.)
  2. The connections of the second SF8638 are incorrect: The gate must be connected to both the 330K resistors outside of the box! Please refer to the schematics of the 312-0522.
  3. The connections of the oscillator transistor are incorrect, too. Please refer to the schematics of the 312-2433.
  4. There must be a capacitor between the source of the mixer FET and the oscillator coil. Otherwise the 1.2K resistor (isn´t it a 8.2K resistor?) at the source of the mixer FET would have a DC short circuit. Please refer to the schematics of the 312-2433.

The absolute values of the supply voltages are not so very important. But the resulting drain currents should fit. So please measure the voltage drops along the following resistors (i.e. the voltage differences between the two connections of the following resistors):

  1. The 330 Ohm resistor outside of the box (cascode).
  2. The 120 Ohm resistor outside of the box (cascode).
  3. The 120 Ohm resistor outside of the box (mixer and oscillator).
  4. Only if possible: The 1.2K (or 8.2K?) resistor inside of the box.

More details later on.

Andreas

 

This article was edited 26.Apr.16 23:15 by Andreas Steinmetz .

John Vogt
 
 
USA  Articles: 17
Schem.:
Pict.: 5
28.Apr.16 07:04

Count of Thanks: 3
Reply  |  You aren't logged in. (Guest)   16

Hello Andreas,

Thanks much for your encouragement! Your critiques of the the schematic were spot on!

Turns out I reversed the pinouts on two devices, misread a resistor value & left out a capacitor. I redrew the schematic with corrections and replaced the original schematic with the new one. Please take look and see if it is now correct.

Here are the voltage drops across the resistors you requested:

                                                      Defective Tuner           Operating Tuner

          1. 330 Ohm outside box-Cas  -1.27V                             -0.98V

          2. 120 Ohm outside box-Cas  +0.50V                            +0.39V

          3. 120 Ohm- Mixer/Osc.          +0.61V                            +0.60V 

          4. 8.2K Ohm inside box           +4.07V                            +4.59V

Assuming the schematic is correct, I am unclear as to how the amplified RF signal enters the mixer. Could you shed some light on that for me? I see on the other two Gorler schematics a double coil mounted on the same form with 150pf & 500pf capacitors wired across one of the coils, similar to the dual coil in this unit. Is the signal coupled from one coil to the other?

John

 

Andreas Steinmetz
 
Editor
D  Articles: 638
Schem.: 81
Pict.: 39
28.Apr.16 22:16

Count of Thanks: 4
Reply  |  You aren't logged in. (Guest)   17

Hi John,

I think the schematic is o.k. now. Just one last question: Is there really a resistor connected in parallel to the oscillator coil? If yes, then it would be a damping resistor. Not unusual, but nevertheless I expected a small capacitor at that place. (O.k., that are peanuts...)

Due to the corrected connections of the second FET, the table of the measurements have to be corrected as follows:

 (2) SF8636

     Gate Source         +01.47V                         +01.17V

               Drain         +11.16V                          +11.32V

     Source Gate        +00.08V                          -00.32V

To save time please let me give you just a short answer at this moment: The resulting drain currents seem to be o.k.. Detailed calculations and more explanations will appear soon here.

John, could you please take some photos of your FM unit, from the top of the box, from the side with the Görler label and from the inside, as long as it is still opened? We´ll need them later when we (or better you) install the model into our RMorg database. Thank you!

Andreas

This article was edited 28.Apr.16 22:21 by Andreas Steinmetz .

John Vogt
 
 
USA  Articles: 17
Schem.:
Pict.: 5
29.Apr.16 03:13

Count of Thanks: 5
Reply  |  You aren't logged in. (Guest)   18

Hi Andreas,

I would be happy to take pictures for the database, I'll add pictures of this model SAE tuner also.

Regarding the "resistor" across the oscillator coil. I'm not really sure if it is a resistor. It looks to me like a themistor. It is a small grey disc with a lead on either side. It does have the number "89" on it. It very well could be a type of capacitor I'm just not familiar with. I've attached a picture.

I'll await your recommendations as what to do next to get the tuner operating properly.

 

Thanks!

John

Attachments:

This article was edited 29.Apr.16 03:15 by John Vogt .

Andreas Steinmetz
 
Editor
D  Articles: 638
Schem.: 81
Pict.: 39
30.Apr.16 23:37

Count of Thanks: 3
Reply  |  You aren't logged in. (Guest)   19

Hi John,

first of all: The oscillator component is a capacitor, not a resistor. Read the printing upside down: 6.8. That means that the capacity is 6.8 pF (perhaps the dot or comma is missing on the print, but 68pF would be much too much capacity for an FM oscillator).

The calculations:

1. Cascode:

If you calculate the no load voltages at the points where the gates of the first and the second FET are connected to, then you see that the voltages are the same as measured (with connected gates). That means that the gate currents are zero. This and the other voltages make me sure that the FETs are o.k..

The resulting currents are (values of the defective tuner; values of the working tuner in brackets):

ID = ΔU(120Ω) / 120Ω = 0,5V / 120Ω = 4,17mA (= 0,39V / 120Ω = 3,25 mA)

IS = ΔU(330Ω) / 330Ω = 1,27V / 330Ω = 3,85mA (= 0,98V / 330Ω = 2,97 mA)

Due to the zero gate currents, ID and IS should be identical. The reason for the diffferences can only be tolerances of the 120Ω and 330Ω resistors. Assuming that these resistors are +-5% types the tolerances of <10% between ID and IS are o.k..

We see that the cascode current of the defective tuner is approx. 4mA and the current of the working tuner is approx. 3mA. The power dissipation of the FETs is still far away from a critical value. So I would say: don´t worry about that difference. (It was a good idea to use a BF245B because with a BF245C the current would be much higher.) (If you really wanted to reduce the current you could increase the value of the outside 330Ω resistor, but I wouldn´t do that.)

2. Mixer and oscillator:

With similar calculations you can find that the current of the mixer is approx. 0,5mA and the current of the oscillator is approx. 5mA. The values of the defective tuner are nearly the same like the values of the working tuner.

John, step 1 is successfully finished now.

Please give me some time to prepare step 2. Some questions: Did you ever make alignments on an FM frontend? Do you have equipment like a service signal generator up to more than 100Mc? Do you perhaps know a specialist who could assist you?

Andreas

This article was edited 30.Apr.16 23:45 by Andreas Steinmetz .

John Vogt
 
 
USA  Articles: 17
Schem.:
Pict.: 5
01.May.16 00:37

Count of Thanks: 3
Reply  |  You aren't logged in. (Guest)   20

HI Andreas,

Ha! Ha! Well isn't that embarrassing, so much for my electronics expertise............ I should stick to my day job.;))

In any event I'll correct the schematic and update the attached file.

I have performed an alignment on the front end of these tuners. I have a Sound Technology 1000A FM stereo generator as well as a Tek 100Mhz scope and Sound Tech 1700B distortion measurement system.

I've adjusted the coil slugs and trimmer caps for max signal level with the tuner receiving a low level signal (10-100uV) from the Sound Tech. The tuners built in oscilloscope makes it easy to monitor the signal level as the adjustments are made. The oscillator coil and trimmer I've not adjusted as the Nixie display show the correct station frequency across the band.

Thanks again for the help, its been most enlightening,

John

Andreas Steinmetz
 
Editor
D  Articles: 638
Schem.: 81
Pict.: 39
01.May.16 21:19

Count of Thanks: 3
Reply  |  You aren't logged in. (Guest)   21

Hi John.

just in order to guard against misunderstandings: You wrote that you have adjusted one or more tuners. Is one of them already the defective tuner with the replaced FET or not? Do you still need assistance for step 2?

Here are some more "peanuts" for the schematic:
1. I suppose that there is a capacitor missing in parallel to the first tuned circuit (3.5 pF).
2. Is the antenna input grounded at the bottom or at a tapping in the middle of the coil or/and at the balun?
3. Is the oscillator transistor an AF125 or an AF124?
3. What are the values of the following capacitors?:
- capacitor between C and E of the oscillator transistor (10pF?)
- capacitors of the secondary IF coil (160pF and 500pF?)
- coupling capacitor between the oscillator coil and the connection for the FET buffer (counter)

Step 2:

Let´s begín with some answers to your questions:

1. The inductor and 1nF capacitor between gate and drain of the first FET is a neutralisation network which improves the stability of this stage. Depending on the used FET type or example, sometimes an additional resistor (for example 100Ω) is necessary.
2. The bodies of the output coil of the RF cascode amplifier and of the input coil of the mixer stage are mounted not far away from each other. So the magnetic flux lines can meet each other: the coils are magnetically (or inductively) coupled. Additionally there is a mixed (inductive and capacitive) coupling by parasitic inductivities and capacities via the connection to the variable capacitor and by the crosstalk inside the variable capacitor itself. Sometimes a little shielding plate between the coils is necessary to avoid a too tight coupling.
3. The coils of the IF transformer are relative tightly coupled; they are on the same coil body in a specific distance.

Now an important thing: Your tuner doesn´t have a conventional dial glass. So aligning of the oscillator differs from the normal procedure! It doesn´t matter how the oscillator is aligned: The stations are always found at the correct frequency shown by the counter (as long as the IF frequency is o.k.)! That is nice but that gives us a problem during the alignment procedure. Question: Which frequencies (lowest & highest) can be reached when turning the dial knob to the left and right maximum? Are there differences between the properly working tuners and the defective one?

Andreas

This article was edited 01.May.16 23:32 by Andreas Steinmetz .

John Vogt
 
 
USA  Articles: 17
Schem.:
Pict.: 5
02.May.16 01:19

Count of Thanks: 3
Reply  |  You aren't logged in. (Guest)   22

Hi Andreas,

I haven't aligned the defective tuner, that was done by a repair technician attemping to repair the tuner. He attempted an alignment but quickly realized there was an internal problem with the front end. Unfortunately he cracked the slugs during the attempted alignment. I didn't know this until I started to delve into the repair. Also, other than the front end only the discriminator is adjustable. The IF utilizes 10 pole & 5 pole toriodal filters by Filtec that are non adjustable.

I would appreciate further assistance with Step 2. 

Schematic questions:

1.There is no missing capacitor in parallel to the first tuned circuit.

2.There is a tapping in the middle of the coil, I missed that and will make the correction. The balun and coaxial shield are grounded at the balun end. At the other end of the coax the shield is soldered to the chassis of the box and not connected to the ceramic pass through that one end of the coil is connected to.

3. As best I can tell the oscillator is an AF125. Because of the paint/glue covering the transistor cases I was only able to make out an "F" & "5" on one of the three tuners I examined. I tried chipping off the paint, that didn't work as doing so appeared to have removed the markings.

4. capacitor at C & E of oscillator--10n

   capacitors at secondary IF coil- 160pf, 500pf

   coupling capacitor to FET buffer--3.5pf

 

Thanks for the explanations, one of the other tuners does contain a 100Ω resistor in the series network of the first FET. The FET is not an SF8636 (BF245?) as the gate and source connections are reversed.

The working tuners reach 87.7 to 108.7. The defective tuner will only go from 87.7 to 108.1. Part of this may be due to operating the tuner without the cover though.

 

John

 

 

Andreas Steinmetz
 
Editor
D  Articles: 638
Schem.: 81
Pict.: 39
02.May.16 22:18

Count of Thanks: 1
Reply  |  You aren't logged in. (Guest)   23

Hi John,

correction: The capacitor at C & E of oscillator is not 10nF but 10pF!

Because we don´t have any official alignment hints and we have some cracked ferrite slugs, we must extemporise a little bit.

  1. Let´s choose the alignment frequencies as follows:
    Low frequency (lf): 89.0 MHz
    High frequency (hf): 106.0 MHz
  2. Refer to the schematics of the 312-2433 or 312-0522. In one of the pictures the aligning points are labeled with letters (H,J,K,L,M,N,O,P,Q). Let´s use the same letters.
  3. First please try to memorize the positions of all the aligning elements on a piece of paper and/or take a photo. Then you have the chance to undo every alignment modification if necessary.
  4. Please close the box by the original cover, switch on the receiver and let it heat up to normal working temperature. Then choose the maximum frequency with the dial knob and increase it from 108.1 to 108.5 MHz by slightly turning trimmer P (counter-clockwise?). This procedure will not influence the lowest frequency significantly, but gives us more space at the upper end.
  5. Connect the signal generator to the antenna input, set the frequency to 106.0 MHz (hf). Use a signal amplitude as low as possible but strong enough to check the signal strength with your built-in oscilloscope during the aligning procedure.
  6. I would do the aligning procedure in the inverse order of the aligning point letters. First we try to increase the sensivity by aligning the trim capacitors due to the ferrite slungs can´t be turned.
  7. Trim the IF filter (R,Q) to maximum. Repeat this step of the alignment procedure until no further improvement is possible. Be careful, just a slight correction should be necessary.
  8. Trim the intermediate bandfilter (N,K) to maximum. Repeat this step of the alignment procedure until no further improvement is possible.
  9. Trim the antenna filter (J) to maximum.
  10. Is the tuner sensivity o.k. now at the frequency of 106.0 MHz? Please also check the sensivity at 89.0 MHz and tell us the results.

Andreas

John Vogt
 
 
USA  Articles: 17
Schem.:
Pict.: 5
03.May.16 05:37

Count of Thanks: 3
Reply  |  You aren't logged in. (Guest)   24

Hi Andreas,

I followed your suggested procedure unfortunately there was little improvement. The sensitivity is abour 3uV at 106 Mhz and 6uV at 90Mhz.

I'm tempted to try replacing the broken slugs with those I removed from a couple of tunable coils that were ordered from Coilcraft. The slugs are about 1mm to large in diameter but I was able to reduce them to the proper size by rotating them while wrapped in emery paper.

John

 

Andreas Steinmetz
 
Editor
D  Articles: 638
Schem.: 81
Pict.: 39
03.May.16 22:39

Count of Thanks: 5
Reply  |  You aren't logged in. (Guest)   25

Hi John,

I suppose that the reduced sensitivity at 89.0 or 90.0 MHz compared to 106.0 MHz is caused by a too low inductivity of the first three coils and that a small amount of ferrite material brought into every coil body could solve the problem. But where should we get the material from? Replacement of a ferrite slug by your coilcraft slug could work fine but I am sceptical, because it is unsure if that material is suitable for frequencies around 100 MHz.

I have a better idea: Isn´t the oscillator ferrite slug (O) completly o.k. and could be removed without damaging anything? If yes, then please replace the original oscillator slug with your coilcraft slug and adjust the oscillator minimum and maximum frequencies as they were before. Then check if the sensivity at 106.0 MHz is still the same. That would make us sure that the oscillator amplitude is still high enough. If the oscillator doesn´t work properly we are sure that the coilcraft material is not suitable for us. With this procedure we would have one spare original ferrite slug which we could use instead of a broken slug...

Andreas

This article was edited 03.May.16 22:42 by Andreas Steinmetz .

John Vogt
 
 
USA  Articles: 17
Schem.:
Pict.: 5
04.May.16 03:46

Count of Thanks: 2
Reply  |  You aren't logged in. (Guest)   26

Hi Andreas,

That was a great idea, replaced the oscillator slug with the coilcraft after reducing it slightly further to prevent binding and lo and behold it worked! Frequency extremes and sensitivity are the same.

Now, should slug M or slug H be replaced first?

One thing that continues to bother me is the signal level displayed on the oscilloscope is about one third lower in vertical height than that displayed on the properly funtioning tuners. Also, with no antenna connected the noise displayed horizontally on the baseline is less also. This makes me question whether there is an issue with one or both of the remaining original FETS.

Anyhow, I'l replace the broken slugs pending your instructions. BTW, only slug M & H need replacing, slug L is adjustable and works properly.

John

Andreas Steinmetz
 
Editor
D  Articles: 638
Schem.: 81
Pict.: 39
05.May.16 16:32

Count of Thanks: 5
Reply  |  You aren't logged in. (Guest)   27

Hi John,

you should remove all the slug parts from coil body M but save them to eventually add them into coil H if the inductivity is too low there. Then screw the ex oscillator slug into coil M and start the aligning procedure: 89 MHz slugs and 106 MHz the capacitor trimmers. Repeat and finish with the trimmers.

If the first FET had a short then it could be that the second FET got damaged. Also please check the mixer, especially the 8.2K resistor.

Sorry John, I can not assist you for the next days. Good luck!

This article was edited 05.May.16 16:32 by Andreas Steinmetz .

John Vogt
 
 
USA  Articles: 17
Schem.:
Pict.: 5
05.May.16 17:47

Count of Thanks: 2
Reply  |  You aren't logged in. (Guest)   28

Hi Andreas,

Thanks for the response. I understand your busy so whenever you're able to respond will be fine.

I went ahead and replaced the slug from coil M with the oscillator slug, another alignment resulted in no improvement so it appears the stuck slug was in the correct position.

Before replacing the slug in coil H I inserted a hex wrench into the slug to see if increasing the inductance would detune the circuit. That made little if any difference so I decided to leave that slug in place for now.

I'll replace the second FET, hopefully we'll get lucky. If the mixer needs replacing that's going to be a real challenge as there's very little room to maneuver..

John

John Vogt
 
 
USA  Articles: 17
Schem.:
Pict.: 5
10.May.16 02:25

Count of Thanks: 6
Reply  |  You aren't logged in. (Guest)   29

Hi Andreas,

I wanted to give you an update as to my progress on this repair.

Before doing so I wanted to tell you that I can't thank you enough for all of the help, time and effort

you spent walking me through this problem. You a great mentor and it is most appreciated!

Now, onto the update..... The front end is fixed! I replaced the second FET and improved the sensitivity

across the band back to spec. The problem with low signal level appearing on scope remained however.

In doing some further checking I noted the voltage at the IF output was about 1V low compared to a properly functioning tuner. The output feeds directly into a CA3053  linear IC, used, I assume, as the first IF amplifier . I have spares of this device so I installed a replacement---- problem solved. The tuner now finally works the way it should. This after 1 1/2 years and 3 technicians.

I have redrawn the schematic again and would appreciate if you could review it for correctness before I add it to the database along with  pictures of the unit.

John

This article was edited 10.May.16 03:07 by John Vogt .

Andreas Steinmetz
 
Editor
D  Articles: 638
Schem.: 81
Pict.: 39
12.May.16 23:02

Count of Thanks: 1
Reply  |  You aren't logged in. (Guest)   30

Hi John,

thank you for your compliments. And congratulations for your success after all the years!

You asked me to review your redrawn schematic for correctness. Of course I would do that but are you really sure that the last uploaded schematic (see post #14) is your newest version? The attached version still contains nearly all the errors we discussed in posts #18 and #21 ... #23.

Andreas

John Vogt
 
 
USA  Articles: 17
Schem.:
Pict.: 5
13.May.16 04:05

Count of Thanks: 2
Reply  |  You aren't logged in. (Guest)   31

Hi Andreas,

I have redrawn the schematic incorporating the corrections we have discussed but have not uploaded it as yet as I ran into a problem trying to update the original schematic in post #14 so the original drawing is still there along with a double posting of the post itself. I sent a message to the moderators to have it corrected but have not heard from them as yet. If you don't mind I'll send you the latest schematic directly in the interim.

Thanks,

John

Andreas Steinmetz
 
Editor
D  Articles: 638
Schem.: 81
Pict.: 39
13.May.16 20:53

Count of Thanks: 3
Reply  |  You aren't logged in. (Guest)   32

Hi John,

yes, please send me the schematic directly via e-mail.

Andreas

John Vogt
 
 
USA  Articles: 17
Schem.:
Pict.: 5
24.May.16 20:16

Count of Thanks: 2
Reply  |  You aren't logged in. (Guest)   33

I wanted to give an update to those that were following this thread of the final outcome.

Andreas and I worked on the final version of the schematic via email.

The final schematic, along with pictures of the tuner, is now posted in the database for reference.

It can be found in the database under Gorler 312-4247

Thanks again to all those who helped, especially Andreas.

 

John

Andreas Steinmetz
 
Editor
D  Articles: 638
Schem.: 81
Pict.: 39
26.May.16 21:16

Count of Thanks: 3
Reply  |  You aren't logged in. (Guest)   34
. 

Hi John,

it was a pleasure for me to assist you. Thank you for adding the model to our database incl. schematic and pictures. Here is the link to the model Görler 312-2427 (not 312-4247).

Andreas

John Vogt
 
 
USA  Articles: 17
Schem.:
Pict.: 5
24.Mar.20 17:33

Count of Thanks: 1
Reply  |  You aren't logged in. (Guest)   35
. 

Hi All,

 

I wanted resurrect this thread as I am experiencing drifting with two of my SAE FM tuners that use the Gorler front end.

The two tuners wil drift off the assigned staion frequency for the first 30-45 minutes after being powered up. The drift is about -75kHz, lower in frequency, over that tine period.

I thought the 6.8pf oscillator capacitor would be the most likely candidate, however, in corresponding with Editor Andreas Steinmetz, he thought it could also be the gemanium oscillator transistor or two other ceramic capacitors in that stage. He suggested I post in my original thread to see what suggestions other experienced members might have.

The Gorler 312-2427 is listed in the database and can be accessed in the reply just above this post as I am having a problem with pasting the link in my message.

Also, attached is a picture of the oscillator capacitor, you'll note that the capacitor has an orange mark on the capacitor body and Andreas thought that the capacitor could have an N150 temperature coefficient.

Lastly, does anyone know of a source for N150,N220,N330 ceramic disc capacitors as these parts appear to be scarce?

 

Thanks for your help and suggestions!

John

Attachments:

This article was edited 24.Mar.20 18:57 by John Vogt .

  
rmXorg