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 19th Oct 2017, 10:15 pm   #1
Wendymott
Octode
 
Wendymott's Avatar
 
Join Date: Jun 2013
Location: Huddersfield, West Yorkshire, UK.
Posts: 1,782
Default SMD Inductors

Hi Peeps...... Im in a bit of a quandry.. I Bought some smd chokes / inductors for my SSD VFO filters.....Some are 1210 form and some are 0806 form...

I decided to check the values .. as Jeremy G0HZU sent me another set of values to modify the filter.

I have a Marconi LCR Bridge TF2700.... which seems to work pretty well,,, but I am getting unexpected measurements...i.e purchased 330 nH measured 190nH.... Purchased 56 nH measured 240 nH.

I expect the problem is.. at such low values..there will be "self Inductance".
.
To gain contact I soldered TC wire to the pads and inserted to the sensor terminals via croc clips.

Has anyone got a method of measuring such items please...

I was hoping to make some 220 nH and 270nH inductors for the Go HZU filter, but unless I can measure them.. it could be impossible.
Thanks in advance
__________________
Should get out more.

Regards
Wendy G8BZY
Wendymott is offline  
Old 19th Oct 2017, 11:13 pm   #2
kalee20
Dekatron
 
Join Date: Feb 2007
Location: Lynton, N. Devon, UK.
Posts: 7,059
Default Re: SMD Inductors

It's all self-inductance. You mean, stray inductance!

The secret is to make a little jig of some sort, with a couple of pieces of metal (copper foil) a millimetre apart. Fix it to your bench so it can't move. Then short between them, and measure the inductance. This is your stray. Remove the short, put in your SM inductor, holding it in place with something like a cocktail stick, and measure again. Subtract the stray value, and you're left with what you wanted to measure.

I have to say though, 56nH is a pretty small inductance to measure.
kalee20 is offline  
Old 19th Oct 2017, 11:40 pm   #3
Wendymott
Octode
 
Wendymott's Avatar
 
Join Date: Jun 2013
Location: Huddersfield, West Yorkshire, UK.
Posts: 1,782
Default Re: SMD Inductors

Yep[ Kalee.. Stray inductance.... Good plan..... I will let you know the outcome.
__________________
Should get out more.

Regards
Wendy G8BZY
Wendymott is offline  
Old 20th Oct 2017, 6:54 am   #4
Craig Sawyers
Dekatron
 
Craig Sawyers's Avatar
 
Join Date: Nov 2010
Location: Oxford, UK.
Posts: 4,941
Default Re: SMD Inductors

HP LCR meters have SM jigs for measuring just that sort of component - and they cost a fortune. The hold the part in without soldering.

But as Kalee says, it is easy to make a little jig. I made one out of a small circuit board that fit the terminals on my ancient HP 250B RX meter to measure SM capacitors. Connect to the terminals, measure the jig capacitance. Then solder on the SM capacitor and renull the bridge. Subtract the two.

I would send a pic - but I've lost the darned thing! Which means I'll have to make another.
Craig Sawyers is online now  
Old 21st Oct 2017, 7:09 pm   #5
Wendymott
Octode
 
Wendymott's Avatar
 
Join Date: Jun 2013
Location: Huddersfield, West Yorkshire, UK.
Posts: 1,782
Default Re: SMD Inductors

Just an update... I Made a pcb jig to accept SMD devices... I have come to the conclusion that such small values cannot be reliably measured...
I finished up getting the data for the 37-2 torrroids.. 8 T = 270nH 7T - 220nH (ish) depending upon winding space and possibly wire gauge , but measuring on the TF2700 was to say the least arbitrary.
I did think about an L/C oscillator and buffer, with a reference capacitor...... measure the frequency and do the sums..
Oh yes... I forgot..... I bought some SMD Inductors from a supplier ............. looks like they had been "binned" incorrectly at the suppliers.... were 0R33 resistors instead of 270nH inductors.... trouble is.. I cannot find the packing note re which supplier it was.
__________________
Should get out more.

Regards
Wendy G8BZY

Last edited by Wendymott; 21st Oct 2017 at 7:12 pm. Reason: Additional info
Wendymott is offline  
Old 21st Oct 2017, 7:18 pm   #6
Ed_Dinning
Dekatron
 
Join Date: Jan 2004
Location: Newcastle upon Tyne, Tyne & Wear, UK.
Posts: 8,171
Default Re: SMD Inductors

Hi Wendy, it is possible to operate the TF2700 with an external oscillator at a higher frequency and it can also be modified to have fine and course "loss balance" controls that will allow a better null.
As Kalee says, biggest problem can be the jig at these frequencies. You could also try measurements using a Q meter if you have one.

Ed
Ed_Dinning is offline  
Old 21st Oct 2017, 8:21 pm   #7
Guest
Guest
 
Posts: n/a
Default Re: SMD Inductors

A known capacitor in parallel and a GDO?
 
Old 22nd Oct 2017, 11:51 am   #8
Wendymott
Octode
 
Wendymott's Avatar
 
Join Date: Jun 2013
Location: Huddersfield, West Yorkshire, UK.
Posts: 1,782
Default Re: SMD Inductors

Hi Ed...... As it happens... I was thinking about modifying the "loss balance" pot,,, as it is very nearly at "0" but not quite. I dont have a "Q" meter or a GDO,,, but I can easily knock up an oscillator and counter...
Re the loss balance pot mod... are there any notes on this mod on the forum please ?? I was thinking it may be useful to put in a 10T pot, as I am not too worried about the scale.
__________________
Should get out more.

Regards
Wendy G8BZY
Wendymott is offline  
Old 22nd Oct 2017, 12:18 pm   #9
Argus25
No Longer a Member
 
Join Date: Oct 2016
Location: Maroochydore, Queensland, Australia.
Posts: 2,679
Default Re: SMD Inductors

Over the years I have had much more success measuring capacitors accurately than inductors. I think there are a lot of reasons for this that I won't go into here.

So what I would do to measure an inductor in this range is to place a known (checked) value of capacitor in parallel with it, either a surface mount part or a radial part with short leads to make it say resonate roughly around 5MHz. So turn it into a parallel resonant circuit for the purpose of measurement.

I'd place about a 10k and 250k resistor in series, the 10K connected to the generator ground connection and 250k in series with that to one resonant circuit leg and the hot generator lead connected to the other leg of the resonant circuit and place the scope across the 10k resistor. The ground of the generator and ground of the scope being common.

(The high value resistance 250k isolates the capacitance of the scope probe from the resonant circuit and doesn't cause too much damping due to the generator's low output Z)

Then simply adjust the generator to find when the voltage across the 10k resistor drops at parallel resonance, then calculate the inductance from the frequency and known capacitor value.
Argus25 is offline  
Old 22nd Oct 2017, 2:02 pm   #10
G0HZU_JMR
Dekatron
 
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 3,077
Default Re: SMD Inductors

The other way to get the inductance right is to exploit the parallel LC trap circuits that are already in the elliptic lowpass filter circuit. These will put deep notches in the filter stopband at about the resonant frequency of the trap. In this case they are at 80MHz and 130MHz as can be seen in the simulation below. So if your spectrum analyser and tracking gen can be set up to see these notches you could trim the inductors this way. But you may find your analyser can't do this efficiently enough compared to other methods like the GDO or scope methods already suggested. But it's the way I have always done it with elliptic filter designs and I've done this many times over the years. If your analyser can display these deep notches it can be very rewarding to watch the filter passband take the required shape and cutoff response as you steer the traps to the right frequency

I've not used a meter like the 2700 for maybe 25 years but it may struggle with small SMD inductors. Usually, a 220nH SMD inductor from someone like Coilcraft will be made with wire that is very fine... like a hair for example. So the ESR of the inductor could be typically 0.5 ohm even when measured at DC. So you may be asking the 2700 to measure an inductive reactance of maybe 10-20milliohms at a few kHz that is in series with about 500 milliohms of resistance.

A 220nH inductor made from a T37-2 toroid will probably be OK at 39MHz in terms of Qu but you may find the Qu is only about 70 up at 39MHz. But I think I used 70 in the design anyway. The 6 or 17 mix would be better here in terms of Qu but this won't have that much impact on the filter droop at 39MHz. i.e. a negligible improvement I think.
Attached Thumbnails
Click image for larger version

Name:	minikits39.jpg
Views:	71
Size:	91.6 KB
ID:	151163  
__________________
Regards, Jeremy G0HZU

Last edited by G0HZU_JMR; 22nd Oct 2017 at 2:10 pm.
G0HZU_JMR is offline  
Old 22nd Oct 2017, 8:59 pm   #11
Ed_Dinning
Dekatron
 
Join Date: Jan 2004
Location: Newcastle upon Tyne, Tyne & Wear, UK.
Posts: 8,171
Default Re: SMD Inductors

Hi Wendy, no reason why a 10T pot will not work unless the winding then becomes more inductive and gives problems.
I have the schematics for the bridge somewhere so I'll try and look up the pot value.

Ed
Ed_Dinning is offline  
Old 23rd Oct 2017, 12:52 am   #12
Wendymott
Octode
 
Wendymott's Avatar
 
Join Date: Jun 2013
Location: Huddersfield, West Yorkshire, UK.
Posts: 1,782
Default Re: SMD Inductors

Hi peeps, Ed, Jeremy and Argus. Interesting comments from you all... Jeremy.. I will disconnect the filter from the DDS O/p to see the filter response...
Ed, I have the schematic of the 2700 thanks.......
Argus.... thanks for the idea....... but today I knocked up an L/C fet oscillator, and found that if I calculated the resonant frequency of say a 1 uH choke, and with a "C" Fixed, I would get a frequency of say... 15.915 Mhz..... but replacing the Inductor with a supposedly identical inductor, there was a tolerance spread of say 100 Khz, this was the same for 2uH2 and 3uH3.....but it will give me an idea of the actual value... especially if I test 3 of each value... get an average.... and go from there.
Attached Thumbnails
Click image for larger version

Name:	FET VCO.jpg
Views:	52
Size:	41.6 KB
ID:	151218  
__________________
Should get out more.

Regards
Wendy G8BZY
Wendymott is offline  
Old 23rd Oct 2017, 10:45 pm   #13
G0HZU_JMR
Dekatron
 
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 3,077
Default Re: SMD Inductors

In terms of the capacitors, it's best to stick with small SMD caps rather than attempt to use leaded caps for this type of filter. This is because any parasitic inductance in the caps will cause the elliptic lowpass filter to look like a highpass filter up at UHF.

Up at UHF you can visualise this filter where the toroids are no longer behaving as simple lumped parts and they also have the 5.6pF or the 18pF bypass cap across them. These caps will pass UHF signals quite well unless the shunt caps can prevent this. So the simplest view of this LPF at UHF is to imagine it as a HPF consisting of shunt (cap) inductances and series 5.6pF and 18pF caps. This is now a fairly tidy highpass filter at UHF. So the lowpass filter stopband can easily degrade to virtually nothing up at UHF if the shunt caps look like inductors of just a few nH up at UHF. The PCB layout tracks can make this much worse so the shunt caps need to have the shortest path to ground possible.

The raw output of the DDS chip will typically be quite fast and can churn out image spurs up into the GHz region. So anything you can do to reduce the inductance of the shunt caps (eg using SMD parts) will bring a reward in terms of keeping the DDS spurs low up at UHF. Using two SMD caps in parallel to make up each shunt cap will improve things even more but maybe this is overdoing it a bit.

But definitely try and use SMD parts and keep the connections and via holes really snug on your PCB layout. If you made this filter with leaded parts and a fairly casual layout the LPF would become a HPF with virtually no attenuation up above about 300-400MHz for example. It might be OK for your application but the DDS output would look quite dirty up at UHF on a 0-1GHz span on your spectrum analyser.

Are the JFETs the classic J310 in your osc circuit? You don't have much/any source resistance so I'd expect them to run close to IDSS meaning they could be at 40mA or more and get quite warm.
__________________
Regards, Jeremy G0HZU
G0HZU_JMR is offline  
Old 24th Oct 2017, 10:28 am   #14
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 reverted to All SMD Components now....In one of your earlier posts you suggested the inductors should have a "Q" of 70..... I could find no such animals so thats why I played with torroids ...Using 330 nH chokes... " Am waiting for the 220/270 nH inductors...I can get a flat response up to 35 Mhz and it drops off at 45 Mhz by approx 10 db... and it is NOT a steep drop.......... were the characteristic drawings theoretical or actual ?
The J310's ......... they are running at 5v..... and I could not feel too much of a temp rise...the unit will only be used sporadically so I am not going to panic too much.
In fact after this DDS filter has been made.. I doubt that the osc will be used much at all... but it gives me an alternative measuring "stick" for small inductors.
__________________
Should get out more.

Regards
Wendy G8BZY
Wendymott is offline  
Old 24th Oct 2017, 11:05 pm   #15
G0HZU_JMR
Dekatron
 
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 3,077
Default Re: SMD Inductors

The plots are all simulations but they should be quite reliable in terms of what to expect when the filter is built.

This filter is only a 5th order design so it should only give about 7dB rejection at 47MHz. The loss at 39MHz should be about 0.5dB and the notch frequencies and the stopband profile should look just like the simulation as long as you use good quality parts and a tight layout. It really should agree very closely.

I think it will be fine with the T37-2 toroids although they are physically very big. I'd be nervous to use SMD 220/270nH inductors here because they typically have very low Q at these frequencies. So this could cause extra droop at 39MHz.

If you want more rejection at 47MHz I can show you a 7th order design that uses an extra inductor section? But this will need decent inductors and tight tolerance caps to realise the full performance without problems. eg T30-17 toroids would be a good choice here for high Q up at VHF. A 7th order filter would get closer to 20dB rejection at 47MHz. But the Qu of the inductors will need to be up around 150 at 39MHz to prevent excessive droop at 39MHz.

If built with the right quality of parts with a decent layout it should agree extremely well with the simulation. Normally I would measure the inductors on a VNA to export a two port model to get better accuracy but for simple filters like this a basic model with a simple Q model will be good enough.

I mentioned the J310 Idss issue because I'd expect the oscillator to be a bit drifty during warmup if it runs at a high drain current.
__________________
Regards, Jeremy G0HZU
G0HZU_JMR is offline  
Old 25th Oct 2017, 10:43 am   #16
Wendymott
Octode
 
Wendymott's Avatar
 
Join Date: Jun 2013
Location: Huddersfield, West Yorkshire, UK.
Posts: 1,782
Default Re: SMD Inductors

Thanks for the extra info...due to a "messup" by the supplier.. my SMD chokes will not arrive till thursday so I cannot compare yet. I will photo the actual responses and post them here. Ok about the J310's....
__________________
Should get out more.

Regards
Wendy G8BZY
Wendymott is offline  
Old 25th Oct 2017, 10:57 pm   #17
G0HZU_JMR
Dekatron
 
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 3,077
Default Re: SMD Inductors

I had a go at lashing together the 5th order filter on a bit of scrap PCB and measured it this evening. You can see below that the results are very similar to the simulation. The droop at 39MHz is 0.5dB as predicted, the loss at 47MHz is 7dB as predicted and the stopband notches are at 80MHz and 130MHz.

This was built with SMD caps and T37-2 toroids with a tight layout. The 139pF cap was made up of 2 x 68pF caps in parallel. The stopband is slightly lower at 200MHz on the real filter because of the parasitic inductance of the SMD caps as this makes the caps appear a bit bigger than they really are at >150MHz. Even a couple of nH series inductance can make the capacitance of the shunt caps 'grow' >10% up at 150-200MHz.
Attached Thumbnails
Click image for larger version

Name:	Lpf39MHz.gif
Views:	51
Size:	18.3 KB
ID:	151413  
__________________
Regards, Jeremy G0HZU
G0HZU_JMR is offline  
Old 25th Oct 2017, 11:10 pm   #18
G0HZU_JMR
Dekatron
 
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 3,077
Default Re: SMD Inductors

I simulated the original Minikits 30MHz LPF and the plot is shown below. At first this plot looks bad because of the big dip/ripple in the passband and the poor VSWR in band.

But maybe they did this deliberately to try and suppress the 35MHz term that appears at 30MHz. i.e. the DDS will produce a spur at 125 -(30x3) = 35MHz when at 30MHz and this could be 55dB down from the raw DDS output. So the filter adds an extra 15dB to this?

But this seems a clumsy way to do it. A better way would be to use a 7th order filter but maybe the component/tolerance/Q/cost rules this out for them.
Attached Thumbnails
Click image for larger version

Name:	MinikitsLPF30.jpg
Views:	41
Size:	70.2 KB
ID:	151414  
__________________
Regards, Jeremy G0HZU

Last edited by G0HZU_JMR; 25th Oct 2017 at 11:15 pm.
G0HZU_JMR is offline  
Old 25th Oct 2017, 11:51 pm   #19
Wendymott
Octode
 
Wendymott's Avatar
 
Join Date: Jun 2013
Location: Huddersfield, West Yorkshire, UK.
Posts: 1,782
Default Re: SMD Inductors

Jeremy.. Thats the answer.. And when I need 39 Mhz its dropped through the floor....10 Metres + 9 mhz I.F.
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.
Using the T37-2 the turns ratio is quite coarse 8T = 470nH 7T = 390 nH. but that was with 0.75mm wire. Its all a matter of sucking and seeing.
__________________
Should get out more.

Regards
Wendy G8BZY
Wendymott is offline  
Old 25th Oct 2017, 11:58 pm   #20
Wendymott
Octode
 
Wendymott's Avatar
 
Join Date: Jun 2013
Location: Huddersfield, West Yorkshire, UK.
Posts: 1,782
Default Re: SMD Inductors

Hi Jeremy.... Sorry I didnt see your previous posting... your "lashup" seems to correlate with mine..... I am waiting for the SMD inductors.. and I already have the T37-s... Have you winding details of the T37-2's please.. I used 8T and 7T with 0.5mm copper
__________________
Should get out more.

Regards
Wendy G8BZY
Wendymott is offline  
Closed Thread

Thread Tools



All times are GMT +1. The time now is 11:23 am.


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.