Has anyone any thoughts about how one make an LCR meter that works at low frequencies (below 20 Hz)? I need to measure the complex impedance of a capacitor of a few hundred pF. This presents a pretty high impedance at low frequencies. My HP 4284A goes down to 20 Hz, but a few hundred pF presents an impedance in the M¦¸ range which is higher than the optimal range. But ideally I would like to make a measurement at 10 Hz.?
I do have an EG&G DSP based lock in amplifier which could possibly be bought into a setup.
Dave
|
Re: Marconi Q meter exciter - update
My original intention was to make an exciter for the TF1245A that would not require any modifications to it. I had some success in the 10 MHz to 30 MHz range by modifying the design of a ham radio broadband power amplifier with a pair of RD16HHF1 MOSFETS. I was able to run the Q meter from my signal generator with full excitation. Difficulties arose at the lower frequencies where a bigger output transformer was required and at higher frequencies where the MOSFETS were pushed to their limit. I was working on improvements including a built in frequency display when a TF1246 exciter became available as a result of a deceased estate sale. So now I have the Q meter working very nicely with both the TF1246? and TF1247 exciters. I also have made a copy of the TJ230 series loss test jig which was not difficult at all. For completeness I would now like to make a copy of the dielectric test jig TJ155 B/1 or C/1. All I have are a few suboptimal photos so if anyone has one of these jigs and can provide detailed photos of both sides, preferably with a scale in the frame, I would be extremely grateful.
73, Morris VK3DOC
|
My feed shows 10 days without a post
Is That true?
Nothing like the Q meter thread days! :-)
|
Re: How to measure input resistance of an amp with limited equipment and a caveat.
You could wind a 10:1? RF transformer on a toroid as long as it covers your desired frequency. WE used a fixed?3dB pad to provide some isolation from RF souces? Amidon was always my source for toroids. Google?Groups no longer updates any group, it is now just a staic archive up until a few weeks ago.
On Tue, Feb 13, 2024 at 7:32?AM Mikek < amdx@...> wrote: Hi Micheal. I remember you from the SED days. I don't go there much anymore, my mail program
will no longer take me there and I don't like the Google access.
? The amp I'm discussing is an RF amp not an audio amp so the 120 to 12VAC transformer will not work,
? Someone mentioned the Q meter thread on this group, a while back., in it we had a couple of methods to build a
very low impedance RF transformer. So, yes it is possible. However, I like this method, measure how much the voltage drops when the
unknown load, (amps input impedance) is attached to a 50¦¸ source and calculate the answer. I had some odd health issue on Sunday
and some lingering effects on Monday. I'm not especially good at Math, but I could not seem to get calculations right,
even when it was right in front of me using a calculator. Hopefully in a day or two, I'll be back to my normal math ability,
it may not be good but it's functional. :-)
??????????????????????? Mikek
?
??
|
Re: How to measure input resistance of an amp with limited equipment and a caveat.
Hi Micheal. I remember you from the SED days. I don't go there much anymore, my mail program
will no longer take me there and I don't like the Google access.
? The amp I'm discussing is an RF amp not an audio amp so the 120 to 12VAC transformer will not work,
? Someone mentioned the Q meter thread on this group, a while back., in it we had a couple of methods to build a
very low impedance RF transformer. So, yes it is possible. However, I like this method, measure how much the voltage drops when the
unknown load, (amps input impedance) is attached to a 50¦¸ source and calculate the answer. I had some odd health issue on Sunday
and some lingering effects on Monday. I'm not especially good at Math, but I could not seem to get calculations right,
even when it was right in front of me using a calculator. Hopefully in a day or two, I'll be back to my normal math ability,
it may not be good but it's functional. :-)
??????????????????????? Mikek
?
??
|
Re: How to measure input resistance of an amp with limited equipment and a caveat.
The automated test for this on the PRC77 Manpack radio used a low impedance audio output? signal source. It then used a set of relays to drop the signal to 50% at the input. The computer added the individual resistors? to display the impedance. You can replace the relays with a good variable resistor . Once you are at 50%, disconnect it and measure the resistance. I use a quality 120 to 12VAC transformer to convert the 50 Ohm output of my function generator to 0.5 Ohms so that the test doesn't change the test signal level. If you need to do this fairly often, build a fixture with the transformer and resistor, and? use a SPDT switch to take the reading? without disconnecting everything. How fine of a reading do you need? (Percentage) A single turn pot (Using as a Rheostat) will get you down to a few percent easily. If you want more, used a 20 turn precision pot. This eliminates all math and works quite well.
I use the same transformer as a signal to locate shorts on PC boards. I use 1KHz at 0.1VAC so that it doesn't cause semiconductors to start conducting.
I agree with Dave - you need to be sure of the source resistance. If the RF source is a fairly high level, then divided down quite a bit to your 8 mV with say a wide-band resistive step attenuator, then it should be a good 50 ohms already. If it's not so nice, but still big to start with, you can pad it locally. The more attenuation the better to smooth things out.
Once you get the source R right, you can make matching pads locally to change out the net source R right there, while keeping it such that the 50 ohm source "always" sees 50 ohms load. The changing DUT input R will of course reflect back and upset things, if it changes a lot, so some iteration would be needed.
Another option is to forget about 50 ohms and make the source R nearly zero. Remember the HP4342A that you studied in depth? Picture the same kind of deal, where you have a leveled source matched into a wide-band step-down transformer. It would be tricky to replicate that whole thing, but how about say a 10:1 voltage ratio, so a 1/100 Z ratio, getting it to a pretty good voltage source from 50 ohms to about 0.5 ohm, so in a one percent error ballpark. Then you can just pick your desired source R for each experiment. Also monitor the voltage on each side of the Rs to be sure of (or compensate for) what's going on.
Or, maybe you could just use the HP4342A (if you still have it) as the source, and maybe even the level indicator too. If its ranges are suitable, you could maybe run it up to part way to the max frequency, with the tuning cap set to minimum, and use Rs as the DUT, with the Q voltmeter looking at the input of the Norton amplifier. You would then have to interpret the Q reading back to the desired voltage at the input, to figure the resistance versus Rs. I think.
Ed
|
Re: How to measure input resistance of an amp with limited equipment and a caveat.
I looked online and reviewed the schematic for the 651B and its 50 ohm output on all attenuator steps, but the highest step comes straight from the output amplifier, which could be a little off from 50. But on the zero dBm step or lower there are precision resistor attenuator pads that will make the output a very good 50 ohms. So that 50 ohms is the 50 ohms shown at the generator in your first diagram. For diagrams with a 100 ohm resistor near the generator, you need to add an additional 50 ohms external to the generator.
To check your algebra (or mine from my last comment), you can make up a simple case on paper where you can figure out the expected voltages and then use your equation to see if you get the correct answer. For example: substitute a 50 ohm resistor for the input to the amplifier. With a 50 ohm source impedance from the generator, half the generator voltage will appear across the 50 ohm substitute for the amplifier input. Pick a random 1 V for the generator output (Vin) and that means Vout will be 0.5 V. Put that into your equation and see if it checks. Just to be sure, try a different set of conditions, like using a 100 ohm resistor to subsitute for the input of the amp. A little simple math shows that Vout will be 2/3 of Vin.?
|
Re: How to measure input resistance of an amp with limited equipment and a caveat.
On Mon, Feb 12, 2024 at 01:43 PM, Ed Breya wrote:
I agree with Dave - you need to be sure of the source resistance. If the RF source is a fairly high level, then divided down quite a bit to your 8 mV with say a wide-band resistive step attenuator, then it should be a good 50 ohms already. If it's not so nice, but still big to start with, you can pad it locally. The more attenuation the better to smooth things out.
?I don't see how I have any problem with the my signal generator, it is an HP 651B. It's output is a very 'low output impedance amplifier' with a 50¦¸ resistor in series. It has internal attenuators to set the output for precisely the voltage I want. I am monitoring the voltage using a Tektronix 2465CTS oscilloscope.
Once you get the source R right, you can make matching pads locally to change out the net source R right there, while keeping it such that the 50 ohm source "always" sees 50 ohms load. The changing DUT input R will of course reflect back and upset things, if it changes a lot, so some iteration would be needed.
?I think I have the source R right. ?BTW, I have been putting a 50¦¸ resistor as a load, just to check and it does measure as 50¦¸. ?Yes, but I think I'm beyond the error I made in the initial set of measurements. I got wrong answers because I had picked the wrong resistor to solve for. R1 instead of R2. And maybe an algebra mistake. ?It's still a work in progress for me. ?????? Mikek
|
Re: How to measure input resistance of an amp with limited equipment and a caveat.
I agree with Dave - you need to be sure of the source resistance. If the RF source is a fairly high level, then divided down quite a bit to your 8 mV with say a wide-band resistive step attenuator, then it should be a good 50 ohms already. If it's not so nice, but still big to start with, you can pad it locally. The more attenuation the better to smooth things out.
Once you get the source R right, you can make matching pads locally to change out the net source R right there, while keeping it such that the 50 ohm source "always" sees 50 ohms load. The changing DUT input R will of course reflect back and upset things, if it changes a lot, so some iteration would be needed.
Another option is to forget about 50 ohms and make the source R nearly zero. Remember the HP4342A that you studied in depth? Picture the same kind of deal, where you have a leveled source matched into a wide-band step-down transformer. It would be tricky to replicate that whole thing, but how about say a 10:1 voltage ratio, so a 1/100 Z ratio, getting it to a pretty good voltage source from 50 ohms to about 0.5 ohm, so in a one percent error ballpark. Then you can just pick your desired source R for each experiment. Also monitor the voltage on each side of the Rs to be sure of (or compensate for) what's going on.
Or, maybe you could just use the HP4342A (if you still have it) as the source, and maybe even the level indicator too. If its ranges are suitable, you could maybe run it up to part way to the max frequency, with the tuning cap set to minimum, and use Rs as the DUT, with the Q voltmeter looking at the input of the Norton amplifier. You would then have to interpret the Q reading back to the desired voltage at the input, to figure the resistance versus Rs. I think.
Ed
|
Re: How to measure input resistance of an amp with limited equipment and a caveat.
Let me just start with, I have something wrong. I think it is in the algebraic manipulation to Solve for R1. But, I don't know.
The 8mV is open circuit voltage of the signal generator. Yes, if you load the generator with 50¦¸ the output voltage will drop by 1/2.
That is the basis on which I thought rearrangement of the Voltage divider formula would be useful to find the value of R1---AHHH
I think I just realized I should have solved for R2!? Not R1. Here is a link to the Voltage divider formula. Page down past the calculator.
Yes, I did have my scope available and found very little phase shift, actually none with the a 50¦¸ load and just a tiny shift with 100¦¸ load.
Thanks, I think you may have solved where my error comes from.
??????????????????????? Mikek
|
Re: How to measure input resistance of an amp with limited equipment and a caveat.
Also, it appears that your equation is wrong, it should be R2 times Vout divided by Vin minus Vout.? ?(R2 * Vout)/(Vin - Vout)
|
Re: How to measure input resistance of an amp with limited equipment and a caveat.
What is your 8 mV voltage source? RF signal generators with a 50 ohm output port have a 50 ohm internal source resistance and if there is a calibrated output level, that output is with a 50 ohm external load, so open circuit it will be twice the indicated output. If there is internal source impedance, it is in series with the external resistance of? test circuit, making your R2 value in the calculations 100 and 150 ohms. Assuming a high impedance measuring device for the voltages, the internal and external resistances won't cause significant voltage drop, making your Vin voltage measurement OK. Audio and function generators may have various output impedances, including very low internal source impedances; you need to know what is going on in your source generator to get the math correct.
If you want to include an evaluation of reactance, you could use an oscilloscope to synchronize to the source generator and then measure the phase shift on the other channel. The math gets trickier to add phase angles to your equations, though.
|
How to measure input resistance of an amp with limited equipment and a caveat.
I may be a little wordy, but I want the concept to reach the measurement. ?I want to measure the input impedance of a Norton Noiseless Feedback amp. It has been proposed that the input impedance of the amp is a function of the load on the output of the amp. Specifically the input is 60% to 80% of the output resistance. ?I made a crude attempt outlined in the picture below.  However, I have been told thatthis is inaccurate, especially if the input has reactance. With the next measurement I do, I want to be able to change the source resistance just to verify that my first test is wrong. It shows the input resistance changes when the source resistance changes. Even though that doesn't make sense, I want verify that doesn't happen. ? How do I make an input impedance measurement and then change the source resistance, for a second test and still have accuracy?
? ? ? ? ? ? ? ? ? ? ? ? Thanks, Mikek
?PS, for those that want to see a schematic or details of the article, it is here.
|
Re: Fw: [projects-in-metal] A $5 2 axis DRO for a small drill press
https://www.temu.com/goods.html?_bg_fs=1&goods_id=601099515105514
On Friday, January 19, 2024 at 01:52:34 PM CST, N2MS <mstangelo@...> wrote:
Reg, I would like to experiment with these digital calipers. Where did you buy them? 73 Mike N2MS > On 01/18/2024 12:26 PM EST Reginald Beardsley via groups.io <pulaskite@...> wrote: > > > That's a pretty clever solution as a normal microscope XY stage is *very* poorly suited for such a task > > The zero and mm/inch features are very useful. I'd much rather use metric than inches. For drilling enclosures and such zeroing on one corner lets me move to each point very simply and accurately. I'm going to make small dimples in the table at the limits of travel to assist in centering the work on the table properly. > > I spent 9 months looking through a polarizing microscope and stepping 1000 points per thin section identifying the mineral under the crosshairs for my MS thesis. I did 54 thin sections and then recounted the first ones until I had repeatable results. In my case I didn't care where I was. > I simply stepped so many times in the X and Y directions and punched the appropriate button in a row of tabulators. Then fed the data to an ANOVA program. > > I've been testing a micron mechanical dial indicator with surface plate and gauge block. I picked a 0.1001" block for convenience. In the process I put a HF 1" digital mike on it. It read 0.1001". Not as sexy as a Mitutoyu, but just as accurate and 1/5th the price. All of this is fallout from the precision required for PCB manufacture. 0.1mm/0.005" accuracy for $2? Wow! > > Have Fun! > Reg
|
Re: Fw: [projects-in-metal] A $5 2 axis DRO for a small drill press
Reg,
I would like to experiment with these digital calipers. Where did you buy them?
73 Mike N2MS
toggle quoted message
Show quoted text
On 01/18/2024 12:26 PM EST Reginald Beardsley via groups.io <pulaskite@...> wrote:
That's a pretty clever solution as a normal microscope XY stage is *very* poorly suited for such a task
The zero and mm/inch features are very useful. I'd much rather use metric than inches. For drilling enclosures and such zeroing on one corner lets me move to each point very simply and accurately. I'm going to make small dimples in the table at the limits of travel to assist in centering the work on the table properly.
I spent 9 months looking through a polarizing microscope and stepping 1000 points per thin section identifying the mineral under the crosshairs for my MS thesis. I did 54 thin sections and then recounted the first ones until I had repeatable results. In my case I didn't care where I was.
I simply stepped so many times in the X and Y directions and punched the appropriate button in a row of tabulators. Then fed the data to an ANOVA program.
I've been testing a micron mechanical dial indicator with surface plate and gauge block. I picked a 0.1001" block for convenience. In the process I put a HF 1" digital mike on it. It read 0.1001". Not as sexy as a Mitutoyu, but just as accurate and 1/5th the price. All of this is fallout from the precision required for PCB manufacture. 0.1mm/0.005" accuracy for $2? Wow!
Have Fun!
Reg
|
Re: Fw: [projects-in-metal] A $5 2 axis DRO for a small drill press
It doesn't lose its zero when it shuts off. Zero it, open it and turn it off. As soon as you move it, it will turn back on and you will see that it still retains the zero.
I use a 4" HF SS caliper constantly and have never lost the zero unless I pressed the button by mistake.
On Friday, January 19, 2024 at 09:52:20 AM CST, g steinback via groups.io <gsteinba52@...> wrote:
Has anyone had issues with the 'auto off' feature of these digital calipers?
I can see working on a particular project, stepping away for a few minutes (coffee?), and coming back to a powered off caliper - losing its initial reference point.
There doesn't appear any way to override the function on my inexpensive Harbor Freight calipers.
JerryS
|
Re: Fw: [projects-in-metal] A $5 2 axis DRO for a small drill press
Has anyone had issues with the 'auto off' feature of these digital calipers?
I can see working on a particular project, stepping away for a few minutes (coffee?), and coming back to a powered off caliper - losing its initial reference point.
There doesn't appear any way to override the function on my inexpensive Harbor Freight calipers.
JerryS
|
Re: Fw: [projects-in-metal] A $5 2 axis DRO for a small drill press
That's a pretty clever solution as a normal microscope XY stage is *very* poorly suited for such a task
The zero and mm/inch features are very useful. I'd much rather use metric than inches. For drilling enclosures and such zeroing on one corner lets me move to each point very simply and accurately. I'm going to make small dimples in the table at the limits of travel to assist in centering the work on the table properly.
I spent 9 months looking through a polarizing microscope and stepping 1000 points per thin section identifying the mineral under the crosshairs for my MS thesis. I did 54 thin sections and then recounted the first ones until I had repeatable results. In my case I didn't care where I was.
I simply stepped so many times in the X and Y directions and punched the appropriate button in a row of tabulators. Then fed the data to an ANOVA program.
I've been testing a micron mechanical dial indicator with surface plate and gauge block. I picked a 0.1001" block for convenience. In the process I put a HF 1" digital mike on it. It read 0.1001". Not as sexy as a Mitutoyu, but just as accurate and 1/5th the price. All of this is fallout from the precision required for PCB manufacture. 0.1mm/0.005" accuracy for $2? Wow!
Have Fun!
Reg
On Thursday, January 18, 2024 at 08:55:22 AM CST, Steven Greenfield AE7HD via groups.io <alienrelics@...> wrote:
On Thu, Jan 18, 2024 at 03:50 AM, Dennis Zabawa wrote: I have used the commercial version of the device you described.? I found that the drill bit popping up in my vision caused me to actually get nauseous and dizzy so, no more using it. As someone who gets motion sickness easily, I feel your pain. Back to the topic of a cheap DRO: A friend of mine used cheap HF stainless digital calipers to set up someone's microscope XY table. They would saw a block in half, find a micrometeorite track in one half, then find it in the other side. So with the digital calipers, set the XY table to the halfway point between the samples. Zero. Now if you find something at, say, 11.05mm, just spin it the other way to -11.05mm and you are at the same point in the other half of the sawn block. I don't know how well I described that... ? -- Steve Greenfield AE7HD WRWU703 CN87oa?? http://www.ae7hd.com
|
Re: Fw: [projects-in-metal] A $5 2 axis DRO for a small drill press
On Thu, Jan 18, 2024 at 03:50 AM, Dennis Zabawa wrote:
I have used the commercial version of the device you described.? I found that the drill bit popping up in my vision caused me to actually get nauseous and dizzy so, no more using it.
As someone who gets motion sickness easily, I feel your pain. Back to the topic of a cheap DRO: A friend of mine used cheap HF stainless digital calipers to set up someone's microscope XY table. They would saw a block in half, find a micrometeorite track in one half, then find it in the other side. So with the digital calipers, set the XY table to the halfway point between the samples. Zero. Now if you find something at, say, 11.05mm, just spin it the other way to -11.05mm and you are at the same point in the other half of the sawn block. I don't know how well I described that... ? --
Steve Greenfield AE7HD WRWU703 CN87oa??
http://www.ae7hd.com
|
Re: Fw: [projects-in-metal] A $5 2 axis DRO for a small drill press
I have used the commercial version of the device you described.? I found that the drill bit popping up in my vision caused me to actually get nauseous and dizzy so, no more using it.
|
Dr. David Kirkby, Kirkby Microwave Ltd
Michael A. Terrell