UK Vintage Radio Repair and Restoration Powered By Google Custom Search Vintage Radio and TV Service Data

Go Back   UK Vintage Radio Repair and Restoration Discussion Forum > General Vintage Technology > Components and Circuits

Notices

Components and Circuits For discussions about component types, alternatives and availability, circuit configurations and modifications etc. Discussions here should be of a general nature and not about specific sets.

Closed Thread
 
Thread Tools
Old 26th Oct 2017, 1:39 pm   #21
G0HZU_JMR
Dekatron
 
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 3,077
Default Re: SMD Inductors

I'm not sure if the T37-2 toroids I used were genuine Micrometals parts. Using their design tool below I was expecting to be using about 7 turns for both inductors but I ended up using 6 turns of 0.45mm wire. I think this is partly due to the fact that the coil has some self capacitance and the 5.6pF cap probably needs to be smaller to account for this. But my coils were not wound very neatly and I ended up with a fairly tightly squidged 6 turns for the 270nH inductor and 6 turns with about 200 degrees of coverage for the 220nH inductor. Not very elegant I'm afraid but it worked well like this.
Attached Thumbnails
Click image for larger version

Name:	T372_270n.gif
Views:	50
Size:	26.0 KB
ID:	151438  
__________________
Regards, Jeremy G0HZU
G0HZU_JMR is offline  
Old 26th Oct 2017, 11:14 pm   #22
Wendymott
Octode
 
Wendymott's Avatar
 
Join Date: Jun 2013
Location: Huddersfield, West Yorkshire, UK.
Posts: 1,782
Default Re: SMD Inductors

Hi Jeremy... I have compared the T37-2 cores and the SMD inductors I got today.. Please see attached photo's... in my case the SMD devices seemed better.
I will run with the SMD parts.
Thanks for your help and guidance.
Attached Thumbnails
Click image for larger version

Name:	Filter 1.jpg
Views:	50
Size:	90.8 KB
ID:	151466   Click image for larger version

Name:	Filter 2.jpg
Views:	46
Size:	90.1 KB
ID:	151467  
__________________
Should get out more.

Regards
Wendy G8BZY
Wendymott is offline  
Old 27th Oct 2017, 1:01 am   #23
G0HZU_JMR
Dekatron
 
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 3,077
Default Re: SMD Inductors

Looks good and there's not much difference between them so I guess the SMD parts will be neater to fit.

Quote:
Re the Fet Oscillator...... it works ok up or down to 800 nH then stops...I will try a NPN version in the next couple of days, with a device with better Ft. But the principle works ok. As I dont need a stable oscillator, just a "ballpark" value.
I must say I quite liked your osc approach for measuring smallish inductors and I think your JFET version could be tweaked to work better although an NPN BJT version can also work well.

It's possible to analyse your JFET version on something like RFSIM99 and look at how well it generates a negative resistance looking into the gate. Obviously, this negative resistance needs to overcome the real loss resistance or ESR of the inductor under test or the circuit can't oscillate. If it doesn't oscillate then I'd expect the JFET to get quite hot because it might have an Idss of 60mA meaning 300mW Pdiss at 5V. But this depends on the Idss of your particular J310 and this can vary a lot from device to device as we all know. But this could be overcome with a minor tweak to the circuit I think.

Even the old and freebie RFSIM99 simulator program can be set up to show how marginal things look with your current JFET circuit. It's then quite easy to tweak the circuit to make it have more negative resistance up at (say) 40MHz to help it oscillate even with low Q inductors that have high ESR.

I can show you how to do this and post up an RFSIM99 project file that has this all set up for you. It will show where oscillation is expected and how much margin can be expected in terms of negative resistance offsetting the ESR of the inductor under test. Would this be of any help?
__________________
Regards, Jeremy G0HZU
G0HZU_JMR is offline  
Old 27th Oct 2017, 6:47 am   #24
Radio Wrangler
Moderator
 
Radio Wrangler's Avatar
 
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,800
Default Re: SMD Inductors

The T37-2 inductors have more Q, but this only shows up in a little bit more notch depth in your plots.

Another method of measuring small inductors without needing a very wide range oscillator is to put them in series with the inductor of a lower frequency oscillator and look at the percentage frequency shift.

David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done
Radio Wrangler is offline  
Old 27th Oct 2017, 12:57 pm   #25
Wendymott
Octode
 
Wendymott's Avatar
 
Join Date: Jun 2013
Location: Huddersfield, West Yorkshire, UK.
Posts: 1,782
Default Re: SMD Inductors

Hi Jeremy and David.
Jeremy.... Many thanks for the offer of a RFSIM99 file set up..... If I can get my head around it.. I could be a good starting point. I Must admit that when I down loaded it.. it all seemed "gobbledygook" mind you as I have said before I am not really good at analytical stuff.... so yes please.
David. of course I should have just put the "unknown" in series with a known component..that should also be ok on the bridge, as long as the ratio is not too big...I will make another jig for the TF2700 with a known inductor in series on one leg.
The oscillator using a NPN device was interesting...it went "off" straight away using a 1uH coil and I set the frequency as though it had a 100pf in parallel = 15.91 Mhz.... unfortunately when I placed a 2u2 or a 3u3 in circuit.. the frequency was WAY off... not as it should be..... I assumed the capacitance ratio's were wrong...
As David suggests.. even putting an unknown inductor in series would allow the circuit still to operate and a calculation made.
Lots to think about.... thanks guys.
__________________
Should get out more.

Regards
Wendy G8BZY
Wendymott is offline  
Old 28th Oct 2017, 12:14 pm   #26
G0HZU_JMR
Dekatron
 
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 3,077
Default Re: SMD Inductors

I'd still worry about the impact of the ESR of the SMD inductor when you add it to the bigger reference inductor. The reference inductor might have an ESR of 0.3 ohm but the SMD inductor might have an ESR or 0.5 ohm due to its ultra skinny wire. So although you might only be adding 10% to the reactance with the SMD inductor, the resistance could be doubling. Will this affect the setup of the 2700 bridge? I can't remember because it's 25 years since I last used an Mi 2700. We eventually threw ours in the skip at work because it was a waste of time for RF stuff.

You also have the issue of mutual inductance effects but this should be tiny I guess. It just seems a mad way to do it when there are so many other ways that will be better.

Argus suggested a scope method that looks for a dip although I think the resistor values are way too big for this range/type of inductor with its low Q. This assumes I've understood his circuit description correctly of course.

Probably the simplest method at your disposal would be to put a decent/accurate 120pF SMD cap in parallel across the SMD inductor on a test PCB and then look through it with your Rigol analyser and its tracking generator. There should be a suckout dip in the response at the resonant frequency because it will act as a trap. Then just work out the inductance from this trap frequency and the resonating cap value. This way you are measuring the inductance of the 220nH inductors up around 30MHz which is close to your filter cutoff frequency.

But by far the best method is to simply adjust the inductor in the actual elliptic filter such that the notch is in the right place for the filter design. This assumes the caps have been fitted correctly to the filter circuit. You could probably get your analyser to display the notches better but the Rigol might be a bit too deaf to see the 130MHz notch. Also, if your PCB layout is poor with poor components you might not be able to see the notch at 130MHz anyway. The notch could get lost in stray effects in the components and PCB layout. But you could try the tracking gen method with the inductor in parallel with a test capacitor?

If you make this on a test PCB with an input and output connector then the SMD choke will mounted correctly in its natural environment on a PCB and its inductance will be affected by the jig PCB in the same way it will be when it is on the PCB of your DDS LO circuit.

The oscillator method should work fairly well as long as you build the circuit with a tight layout and decent quality SMD parts. I'm surprised that it doesn't seem to work for you even with 1uH or 2uH chokes and the 470pF feedback caps. To get it to oscillate up past 40MHz and up towards 200MHz the feedback caps need to be smaller, eg 100pF each. This will extend the frequency range it can generate adequate negative resistance and so it can measure smaller inductors like this. But for this to work well you would have to make a very tight PCB layout. Get it right and it would do a reasonable job of measuring regular RF inductors over the range of maybe 56nH through 3uH.

You would only need a frequency counter and a spreadsheet or conversion chart to get the inductance with the oscillator method. But because you have the Rigol analyser and the tracking gen you can try the 'trap' method instead? This is going to be quite reliable I think.
__________________
Regards, Jeremy G0HZU

Last edited by G0HZU_JMR; 28th Oct 2017 at 12:44 pm.
G0HZU_JMR is offline  
Old 28th Oct 2017, 1:06 pm   #27
G0HZU_JMR
Dekatron
 
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 3,077
Default Re: SMD Inductors

Of course, you could try removing the shunt caps (or not fitting them initially to the filter) and then look for the notch frequencies of 80MHz and 130MHz with your Rigol analyser. This should make it easier to see the notches. Of course, there will be some self capacitance in the inductor itself so you could argue that this will introduce a tiny error and that the 5.6pF cap should really be a bit smaller to accommodate this self capacitance correctly. But it depends how accurate you want the final filter design to be when compared to an ideal filter response.

I'll try and post up some stuff about the oscillator later today if I get the chance. It should work fairly well although it won't be the best method to use for this type of inductor. I find that it's best to check the inductor close to the frequency of interest. A typical LCR meter or bridge will do it at a very low frequency and the oscillator method will probably measure it at a frequency that is too high. You would also have to be careful with the oscillator method if you wanted to measure the inductance per metre of a transmission line. I don't think it would work well here because it would probably measure at a high frequency leading to a false result. You could also have problems with an inductor that had a very high self capacitance but this is probably not going to happen very often with regular RF inductors in the 56nH to 3uH range.
__________________
Regards, Jeremy G0HZU
G0HZU_JMR is offline  
Old 28th Oct 2017, 7:03 pm   #28
Wendymott
Octode
 
Wendymott's Avatar
 
Join Date: Jun 2013
Location: Huddersfield, West Yorkshire, UK.
Posts: 1,782
Default Re: SMD Inductors

Hi Jeremy... I rejigged the RF Ocsillator and added a 1uH coil in series with the SMD Pads...In a tin can it works very well, with the expected frequencies on my Thandar 1Ghz counter. I feel I can now measure even smaller inductances... I have been down to 22nH with expected results.
Besides going "Down" in inductance values.. I went UP as well, measuring 10uH again with the expected frequency.
I was even able to trim the T37-2 torroids with the same number of turns = 6 one spaced = 22nh.. one not spaced = 270 nH...... I think this will be most useful.
I will keep the TF2700 to one side for measuring LF parts.
I think I am now happy with the DDS VCO and will progress with the rest of the SSB Generator..... I need to make an AGC module for the whole system now, and a "Post Attenuator". I recently bought a RF Attenuator from our Chinese friends with 0.25 DB steps 0 - 4 Ghz.
Thank you for all your efforts and help.
Attached Thumbnails
Click image for larger version

Name:	FET VCO Ver2.jpg
Views:	44
Size:	29.2 KB
ID:	151541  
__________________
Should get out more.

Regards
Wendy G8BZY
Wendymott is offline  
Old 28th Oct 2017, 10:04 pm   #29
G0HZU_JMR
Dekatron
 
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 3,077
Default Re: SMD Inductors

Yes, I think you are better off with the oscillator approach rather than trying two inductors on the TF2700. I just think you could end up getting frustrated as you could be chasing for poorly defined nulls on the bridge. I think this is because the bridge would be trying to measure just a few milliohms change in reactance in series with a significant change in ESR once the SMD inductor is added. You are right down at the low end of the inductance range the 2700 can handle and I think you could easily get lost in a sea of measurement uncertainty as you tried various combos of reference inductor and test SMD inductor.

With the two inductors in series in the oscillator circuit the big change in ESR when the SMD inductor is added won't be an issue as long as there is sufficient reserve in terms of negative resistance looking in to the JFET. Otherwise it would stop oscillating. However, I guess you would have to make sure the 1uH reference inductor is mounted in a stable manner where it can't move as you add/remove the SMD inductor during the test process. Otherwise a 1% shift in the 1uH inductor as you add the SMD inductor will introduce a 10nH error. The same goes for stray capacitance effects if you move your hand near the 1uH inductor.

The alternative would be to reduce the 470pF caps and measure the test SMD inductor directly. This would be quicker for each test because you only have to measure one frequency each time. But the test frequency will be much higher and I think the circuit layout would become more critical if you want to minimise any stray offset inductances in the oscillator components themselves.

I think I'd prefer to use the tracking gen (trap) method over the oscillator but then again I like to use a VNA for stuff like this as it can give a swept measurement of inductance vs frequency very quickly over a large frequency range and it can often be used to indicate a Q curve for low to medium Q inductors. Very powerful!
__________________
Regards, Jeremy G0HZU
G0HZU_JMR is offline  
Old 29th Oct 2017, 10:08 am   #30
Radio Wrangler
Moderator
 
Radio Wrangler's Avatar
 
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,800
Default Re: SMD Inductors

to see the true impedance of a small RF part, a physically small resonant circuit is probably the best you can get, so the resonant frequency is least disturbed by strays and the Q shows the loss factor of the L-C pair.

I most often use a VNA, but that does the job by measuring the strays and subtracting them. My set of APC3.5 impedance standards don't do the job, I have to solder zero and 50 Ohm terminations onto the connector I use to get the inductor into coax. Just the length of an SMA pin is a problem. At lower frequencies I used to use an HP Q-meter. VNA measurements of Q were not good enough for the inductors I was making.

David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done
Radio Wrangler is offline  
Old 29th Oct 2017, 6:13 pm   #31
G0HZU_JMR
Dekatron
 
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 3,077
Default Re: SMD Inductors

One thing to keep in mind is that there will still be about 2dB droop in the DDS output by 39MHz because of the 1.5dB droop due to the sinx/x response and the 0.5dB in the filter. For a mixer LO I guess this doesn't matter much but I did try adding the basic amplitude equaliser circuit to the LPF output. See below for the DDS output response with the equaliser. This has a scale of 1dB/div. This shows the expected level of about -11dBm and the overall ripple across 0-39MHz is now only about +/- 0.3dB.

With a bit of tweaking or with a slightly better equaliser filter this could be improved further but this would be good enough for many applications that need a flatter response.

The 125-(2*RF) spurious at 47MHz when the DDS is tuned to 39MHz was -60dBc on my DDS with the filter fitted. This is about the same level as the 35MHz spurious at 125-(3*RF) when the DDS is tuned to 30MHz.
Attached Thumbnails
Click image for larger version

Name:	dds_eq.gif
Views:	40
Size:	23.8 KB
ID:	151576  
__________________
Regards, Jeremy G0HZU

Last edited by G0HZU_JMR; 29th Oct 2017 at 6:37 pm.
G0HZU_JMR is offline  
Old 29th Oct 2017, 10:41 pm   #32
Wendymott
Octode
 
Wendymott's Avatar
 
Join Date: Jun 2013
Location: Huddersfield, West Yorkshire, UK.
Posts: 1,782
Default Re: SMD Inductors

Hi Jeremy and David. Yes Ok understood about all points...I will try the equalising filter and if ok I will add to the pcb....
Due to the close spacing of the pins of the 9850, I think my laser drum is now showing its time limit, thus I am not getting sharp track/gap resolution. I will try my replacement toner cart and see if I get better resolution, otherwise its a trip to the print shop who have a decent Zerox printer..... I used them for my prototype 9950 pcb before I ordered from China.
Onward and upward.
__________________
Should get out more.

Regards
Wendy G8BZY
Wendymott is offline  
Old 30th Oct 2017, 12:30 am   #33
G0HZU_JMR
Dekatron
 
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 3,077
Default Re: SMD Inductors

It's probably not worth using the equaliser just for an LO signal for the MC1496. So it's really not worth making another PCB just for this. But maybe keep the equaliser circuit handy in case you ever do need a flat output from the DDS?

With the original Minikits 30MHz LPF you probably had over 20dB droop in the filter at 39MHz. Now it's down to about 2dB droop which is a lot better.
__________________
Regards, Jeremy G0HZU
G0HZU_JMR is offline  
Old 30th Oct 2017, 12:57 pm   #34
Wendymott
Octode
 
Wendymott's Avatar
 
Join Date: Jun 2013
Location: Huddersfield, West Yorkshire, UK.
Posts: 1,782
Default Re: SMD Inductors

Hi Jeremy..... Thanks for that... I will keep you posted..
__________________
Should get out more.

Regards
Wendy G8BZY
Wendymott is offline  
Closed Thread

Thread Tools



All times are GMT +1. The time now is 9:50 pm.


All information and advice on this forum is subject to the WARNING AND DISCLAIMER located at https://www.vintage-radio.net/rules.html.
Failure to heed this warning may result in death or serious injury to yourself and/or others.


Powered by vBulletin®
Copyright ©2000 - 2024, vBulletin Solutions, Inc.
Copyright ©2002 - 2023, Paul Stenning.