Thanks. I would still advise repairing the IF transformer if possible, these replacements do not match the selectivity of the original transformers, but do allow the radio to become usable again.
Excellent work 👍👏👏. I enjoyed and learned from this video. Thank you for taking your valuable time to make this video. Have a great day. Regards, JZ from Florida USA
Not my original idea, I will go into it's background in the next video. I did some extensive testing on the design idea, it's not as good as an original IF transformer but it works as a replacement for one of the stages. Replacing both stages as I did here was an extreme test; it works but I suspect that the selectivity suffers. Sorry for not responding sooner, I am not getting notifications for these comments, need to figure out how to get that working so I have to go to each video and look. Cheers,
Excellent idea no matter who thought of it. The inductor spacing idea is very elegant. Why not make the base out of stacked PCB's? For those who don't have a 3d printer. Maybe make the traces for the trimmer caps a little longer so that they can be soldered in easier. I am not knocking anything. I am very impressed.
42:50 I keep a roll of masking take handy to cover up "hot spots" on a board if I think I can't avoid them well. That..... is a great design for a replacement IF coil, and most could also resin cast a base and just drill the needed lead and adjustment holes if they don't have a 3D printer. I don't know if you have gerber files available for that board layout, but I'm betting you could sell them with a Patreon subscription. Great work and good demo, thank you for sharing!
I think easy way - all oneband ( AM MW ) radio's repair to TRF circuit. :) But idea is good - make modern IF transformers from fixed inductor coils and trim caps.
Really nice video. Question: wouldn't you get way higher Q factor if you use a multi-stranded wire like they did very old but high performance radios? Tnx.
So you couple two inductors at a fixed distance and tune them with a variable capacitor to create the IF transformers rather than having fixed capacitors and tuning the transformer coils by adjusting the ferrite core? Is there anywhere you can point me to or any books you can suggest for descriptions or formulae to calculate all of this? or is it just something you test and characterize by trial and error? Regardless, color me impressed! Thanks for the video! Edit: I only ask because I'm currently trying to make a radio receiver from the various bits parts and trash I have around and most of my previous knowledge is tube related but I'm trying to use transistors, and those tiny IF cans are hard to come by these days without ripping apart old walkmans. I'm loathe to use more modern IC designs because it starts to become an incomprehensible black box which isn't so great for actually learning things.
It was mostly empirical testing, see the video on how I tested different gap widths for effect on Q-factor. I may have mentioned something on transistor based IF cans, they use ferrite cores and windings are close ( bad for Q-factor) but they tap the primary (B+ point) to "multiply" the Q-factor so it is useable again. Google for "rfcandy if-cans" for a web page description of this effect.
Howdy. Quite ambitious to build one's own IF cans. In my super het. experimens I have settled for an easier way. I use two inductors and two varcap.s. in a dual series resonant lineup. To interconnect the resont circuits I use a cercap.. of suitable values to get desired tightness of coupling. If tight coupling is desired one may use a toroid core with two windings. Regards.
I am hoping that the repair is only for one IF stage and the other stage is intact, this video shows the worst case repair where both IF cans are trash. Having a 6 - 9 Khz bandwidth (+/- 3 to +/- 4.5 KHz) where the edge of the band is the "half power point" or the -3 dB point is not ideal for good selectivity between stations but... when the alternative is junking the radio it becomes a more viable choice.
It may be in Part 2. I think the inductors were 1300 - 1500 uH range. Using a LC resonant frequency calculator, eg. www.omnicalculator.com/physics/resonant-frequency-lc a 455 KHz IF filter and 1500 uH inductor would require 81.5 pF and this would be a combination of a fixed value and a small trimmer cap. In reality, you would have to test this and probably change to a different cap as there are cumulative capacitance in the circuit that throws the calculated values off.
Hi Scott, I tested that orientation (co-axial) and found that the coupling was better when they were in parallel. I think it works better when "in the can" as well. Additionally the parts placement works better when they are in parallel and I can get the air gap into the PCB or breadboard. Please feel free to experiment on your own! Science is all about proving it to be true and not just taking someone's word for it. :) Thanks for the comment and season greetings. Cheers,
@@EngineeringVignettes I think this is great! I have some Halicrafters that may need this to work properly again They are also close to 455KHz Happy New Year!
I have a question. I've spent hours on the internet and searched my books on valve circuits regarding the Q of an i.f. transformer and written to a radio magazine letters page and nobody responded. Most tell you what Q is. I know about moving the coils to alter the Q and a damping resistor from my books to lower the Q but how to increase the Q when the coils are glued in position? I have a 19 Set tx/rx where one capacitor in the i.f. can has been replaced by a smaller value, 120 pF down to 47pF. From what I have seen so far this seems to indicate it is raising the Q for communications purposes for Morse Code. They say reducing a resistor in an LCR circuit raises the Q but nothing about a smaller value capacitor. I've never opened an i.f. transformer and seen different values of capacitor used.
I've rewound cores, and have refastened tubes to substrates & brackets. It's not hard, you just have to act like a surgeon. I also hate Litz wire, ordinary enamel magnet wire is excellent. The screws in my Hammarlund IF chain are all mangled, but still serviceable. The slots are exactly a hacksaw blade wide, and if you're really careful, like, plan all day for this one slot careful, they were made into screw slots once again. Why not just put it on a single toroid? I have never been able to repair a squished can. This technique is great for that. New installations, too. But watch that voltage.
Yeah, this technique is for replacing a missing IF can or one so mangled that it cannot be fixed. I try to save the transformer can if I can. As I stated in the videos, this is a "cheesy" way to make a transformer, but it works surprisingly well. If an OEM transformer is available I would use that though. There seems to have been a design migration, back when these 1/2" square transformers were made, from the genuine silk litz wire to a cheaper form using multiple twisted strands of normal enamled wire. The latter is a bit easier to repair. Traditional Litz is pretty thick in diameter and the windings can get quite big, especially if using a "universal" winding. The cheaper newer method allowed the "pie" to be a smaller diameter. The Radiotron Designer's Handbook 4th Ed. Chapter 11 details some of the reasoning for using litz... mainly to try and get the Q factor over 100. More modern transistor IF cans use a centre tap to get Q multiplication while using normal single strand enamled wire but the transistors can drive the transformer with more current than a tube... perhaps there could be a way to try a centre tapped transformer on a tube circuit but it would negatively affect the gain... Thanks for watching.
@@EngineeringVignettes The Hammarlund HQ-129-x I have has the IF outputs on autotransformer with the output to grid low on the windings, close to ground. Voltage is reduced, but current goes up. It's not all that innovative a circuit, nor even all that great. It was supposed to be cheap, and the set is showing its age. I thought for sure I let a transformer short out with all 300 volts, but it wasn't. Time to replace another tube socket...
Sorry, this comment was in my "for review" bin and I did not see it until now. Yes the project looks similar but I do not speak the language of the narrator. Cheers,
I used to use one for DXing. I was able to find a station at almost every 10 kilohertz using a small external antenna. No Rf amp, 1 IF stage. I am on Canada. Some radios are better than others.
Wow, that's pretty cool! The one job we all hate is repairing IF cans! Can't wait for Part 2!
Thanks. I would still advise repairing the IF transformer if possible, these replacements do not match the selectivity of the original transformers, but do allow the radio to become usable again.
I have been thinking about this for some time. You have proven my idea. Thanks to proving me right and thanks for a great video.
Thanks for sharing your knowledge. This gives new life to my dead radios.
Excellent that you went to the full effort needed to replace bad IF’s 🤓
Very nicely done! I like your straightforward style of combining tons of relevant information while actually doing it. I learned a lot, thank you.
nice design much more reliable
Excellent work 👍👏👏. I enjoyed and learned from this video. Thank you for taking your valuable time to make this video. Have a great day. Regards, JZ from Florida USA
This is absolutely fantastic. Thank You! I have become a subscriber.
You Eddy_D are a super genius!
Not my original idea, I will go into it's background in the next video. I did some extensive testing on the design idea, it's not as good as an original IF transformer but it works as a replacement for one of the stages. Replacing both stages as I did here was an extreme test; it works but I suspect that the selectivity suffers.
Sorry for not responding sooner, I am not getting notifications for these comments, need to figure out how to get that working so I have to go to each video and look.
Cheers,
Thanks
Excellent idea no matter who thought of it. The inductor spacing idea is very elegant.
Why not make the base out of stacked PCB's?
For those who don't have a 3d printer.
Maybe make the traces for the trimmer caps a little longer so that they can be soldered in easier.
I am not knocking anything. I am very impressed.
42:50 I keep a roll of masking take handy to cover up "hot spots" on a board if I think I can't avoid them well. That..... is a great design for a replacement IF coil, and most could also resin cast a base and just drill the needed lead and adjustment holes if they don't have a 3D printer. I don't know if you have gerber files available for that board layout, but I'm betting you could sell them with a Patreon subscription. Great work and good demo, thank you for sharing!
I think easy way - all oneband ( AM MW ) radio's repair to TRF circuit. :)
But idea is good - make modern IF transformers from fixed inductor coils and trim caps.
Thank you for sharing.
very nice, especially pico mem
Excellent engineering!!👍
Thanks :)
Great information. Well done!
Thanks. Hopefully some similar videos on local oscillator coils (design, replace, rewind) in the near future... on my todo list.
Cheers,
Very interesting. Thanks.
Really nice video. Question: wouldn't you get way higher Q factor if you use a multi-stranded wire like they did very old but high performance radios? Tnx.
So you couple two inductors at a fixed distance and tune them with a variable capacitor to create the IF transformers rather than having fixed capacitors and tuning the transformer coils by adjusting the ferrite core? Is there anywhere you can point me to or any books you can suggest for descriptions or formulae to calculate all of this? or is it just something you test and characterize by trial and error? Regardless, color me impressed! Thanks for the video!
Edit: I only ask because I'm currently trying to make a radio receiver from the various bits parts and trash I have around and most of my previous knowledge is tube related but I'm trying to use transistors, and those tiny IF cans are hard to come by these days without ripping apart old walkmans. I'm loathe to use more modern IC designs because it starts to become an incomprehensible black box which isn't so great for actually learning things.
It was mostly empirical testing, see the video on how I tested different gap widths for effect on Q-factor.
I may have mentioned something on transistor based IF cans, they use ferrite cores and windings are close ( bad for Q-factor) but they tap the primary (B+ point) to "multiply" the Q-factor so it is useable again. Google for "rfcandy if-cans" for a web page description of this effect.
@@EngineeringVignettes Thank you so much! I'll definitely have to drag my test equipment upstairs and play around with this.
Howdy.
Quite ambitious to build one's own IF cans.
In my super het. experimens I have settled for an easier way. I use two inductors and two varcap.s. in a dual series resonant lineup. To interconnect the resont circuits I use a cercap.. of suitable values to get desired tightness of coupling.
If tight coupling is desired one may use a toroid core with two windings.
Regards.
I'd be interested in a kit with the pcbs, capacitors, inductors and the 3D printed base as I restore a lot of these tube radios.
If you want the audio to be between 6KHz and 9KHz the IF amplifier is going to have to have twice the the bandwidth as the audio -- 12KHz to 18KHz.
I am hoping that the repair is only for one IF stage and the other stage is intact, this video shows the worst case repair where both IF cans are trash. Having a 6 - 9 Khz bandwidth (+/- 3 to +/- 4.5 KHz) where the edge of the band is the "half power point" or the -3 dB point is not ideal for good selectivity between stations but... when the alternative is junking the radio it becomes a more viable choice.
I don't remember you mentioning the component values anywhere in the video. Any chance you could post them?
It may be in Part 2.
I think the inductors were 1300 - 1500 uH range.
Using a LC resonant frequency calculator, eg. www.omnicalculator.com/physics/resonant-frequency-lc
a 455 KHz IF filter and 1500 uH inductor would require 81.5 pF and this would be a combination of a fixed value and a small trimmer cap.
In reality, you would have to test this and probably change to a different cap as there are cumulative capacitance in the circuit that throws the calculated values off.
Would it work better if the inductors were on the same axis?
Hi Scott,
I tested that orientation (co-axial) and found that the coupling was better when they were in parallel. I think it works better when "in the can" as well.
Additionally the parts placement works better when they are in parallel and I can get the air gap into the PCB or breadboard.
Please feel free to experiment on your own! Science is all about proving it to be true and not just taking someone's word for it. :)
Thanks for the comment and season greetings.
Cheers,
@@EngineeringVignettes I think this is great! I have some Halicrafters that may need this to work properly again They are also close to 455KHz Happy New Year!
I have a question. I've spent hours on the internet and searched my books on valve circuits regarding the Q of an i.f. transformer and written to a radio magazine letters page and nobody responded. Most tell you what Q is.
I know about moving the coils to alter the Q and a damping resistor from my books to lower the Q but how to increase the Q when the coils are glued in position?
I have a 19 Set tx/rx where one capacitor in the i.f. can has been replaced by a smaller value, 120 pF down to 47pF. From what I have seen so far this seems to indicate it is raising the Q for communications purposes for Morse Code. They say reducing a resistor in an LCR circuit raises the Q but nothing about a smaller value capacitor.
I've never opened an i.f. transformer and seen different values of capacitor used.
I need some of those boards any chance you could help me out
Redoing the IF cans for a SMD problem is too long. Just change both caps for 100pF.
I've rewound cores, and have refastened tubes to substrates & brackets. It's not hard, you just have to act like a surgeon. I also hate Litz wire, ordinary enamel magnet wire is excellent. The screws in my Hammarlund IF chain are all mangled, but still serviceable. The slots are exactly a hacksaw blade wide, and if you're really careful, like, plan all day for this one slot careful, they were made into screw slots once again.
Why not just put it on a single toroid?
I have never been able to repair a squished can. This technique is great for that. New installations, too. But watch that voltage.
Yeah, this technique is for replacing a missing IF can or one so mangled that it cannot be fixed. I try to save the transformer can if I can. As I stated in the videos, this is a "cheesy" way to make a transformer, but it works surprisingly well. If an OEM transformer is available I would use that though.
There seems to have been a design migration, back when these 1/2" square transformers were made, from the genuine silk litz wire to a cheaper form using multiple twisted strands of normal enamled wire. The latter is a bit easier to repair. Traditional Litz is pretty thick in diameter and the windings can get quite big, especially if using a "universal" winding. The cheaper newer method allowed the "pie" to be a smaller diameter.
The Radiotron Designer's Handbook 4th Ed. Chapter 11 details some of the reasoning for using litz... mainly to try and get the Q factor over 100. More modern transistor IF cans use a centre tap to get Q multiplication while using normal single strand enamled wire but the transistors can drive the transformer with more current than a tube... perhaps there could be a way to try a centre tapped transformer on a tube circuit but it would negatively affect the gain...
Thanks for watching.
@@EngineeringVignettes The Hammarlund HQ-129-x I have has the IF outputs on autotransformer with the output to grid low on the windings, close to ground. Voltage is reduced, but current goes up. It's not all that innovative a circuit, nor even all that great. It was supposed to be cheap, and the set is showing its age. I thought for sure I let a transformer short out with all 300 volts, but it wasn't. Time to replace another tube socket...
Similar to the video of my friend Carles Botia and me.
th-cam.com/video/UYcY25AvJAo/w-d-xo.html
Sorry, this comment was in my "for review" bin and I did not see it until now. Yes the project looks similar but I do not speak the language of the narrator.
Cheers,
None of the AA5 :radios were worth a dam.
I used to use one for DXing. I was able to find a station at almost every 10 kilohertz using a small external antenna. No Rf amp, 1 IF stage. I am on Canada. Some radios are better than others.
Very boring Eddy