|
Vintage Test Gear and Workshop Equipment For discussions about vintage test gear and workshop equipment such as coil winders. |
|
Thread Tools |
21st Mar 2017, 7:36 pm | #21 |
Dekatron
Join Date: Nov 2011
Location: Surrey, UK.
Posts: 4,400
|
Re: Frequency counters
The issue of "light touch" pick-off buffers is both interesting and tricky- I've tried a few lash-ups, but even tiny capacitance can make a difference at the HF ends where there is a wide frequency span and reliance on minimum strays.
This may be heretical to some, but the next rainy-day step is to try one of the cheaply and widely available tiny wire-ended battery pentodes as fractional-pF ("gimmick", etc) pick-off, low anode load initial buffer into an otherwise semiconductor slight gain/low-Z output stage. Even a cathode-follower EC90 or low anode load EF95 would give space, power, heat and stray capacitance issues in many oscillator boxes. |
21st Mar 2017, 11:56 pm | #22 | |
Dekatron
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 3,077
|
Re: Frequency counters
Quote:
It uses feedback to minimise the input capacitance but as a side effect of this it will typically introduce a lot of negative resistance at the input across the HF band and also up into VHF. So it can be unstable with certain (inductive?) input terminations and become an oscillator. But it is a classic buffer circuit The simpler JFET source follower from Nuts and Volts will have a high input impedance but I think (at high frequencies) it will struggle to drive a reasonable (30cm?) length of coax with a 1Megohm//20pF scope or counter at the other end. So I think the frequency response will be poor if the coax and counter are connected to it.
__________________
Regards, Jeremy G0HZU |
|
22nd Mar 2017, 1:05 am | #23 |
Triode
Join Date: Mar 2017
Location: Chelmsford, Essex, UK.
Posts: 38
|
Re: Frequency counters
If you're correct about the gain of this, could you point in the direction of a suitable buffer and amplification circuit or something that is already built that I ought to consider to provide sufficient oomph to drive a counter and scope without burdening the the already overstretched output stage of the signal generator?
|
22nd Mar 2017, 11:12 pm | #24 |
Dekatron
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 3,077
|
Re: Frequency counters
Ideally, I think a few more tests are needed first. I had a quick look at the circuit for this sig gen and my knowledge of valves is very limited.
However, I'm going to assume the valve that feeds the output is some sort of cathode follower and therefore it will have a reasonably low source impedance at the cathode. I'm also going to assume that there are several volts DC here normally and a fairly large RF waveform in terms of Vpkpk. Maybe several volts pkpk? This is then tapped down via a 1k cathode resistor and a 1k pot // 500R. So I'm going to assume that the RF voltage at the top of the variable resistor is about a quarter of that at the cathode. Also, any probe capacitance here is going to cause RC rolloff up at VHF because the RC timeconstant of those 1k resistors will be equivalent to several hundred ohms and maybe 20pF for a x10 scope probe and even more for a x1 probe. So probing here is going to give confusing/droopy results up at VHF even before you try fitting a buffer. I suspect that the classic JFET+ BJT unity buffer (suggested by Karl) can be made to work here because it has low input capacitance. Much lower capacitance than a scope probe. I also suspect that the signal level at that 1k variable resistor is much higher than the spec in the manual for the sig gen. So you might not need any more than unity gain. Just don't probe the input of the buffer with a x1 or x10 scope probe or the probe capacitance will cause RC rolloff and much confusion. You also need a fairly fast PNP transistor in that unity buffer if you want to preserve the performance up at VHF. Finally, I would suggest that once a suitable buffer is found (that works) that you share the DC power between the the buffer and the counter via the coax. This way you will only need one DC power feed to power both the buffer and the counter. The buffer has to live inside the sig gen, very close to the 1K variable resistor.
__________________
Regards, Jeremy G0HZU Last edited by G0HZU_JMR; 22nd Mar 2017 at 11:31 pm. |
22nd Mar 2017, 11:44 pm | #25 |
Dekatron
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 3,077
|
Re: Frequency counters
If you don't want any active parts inside the sig gen then the other way to do it would be a series resistor of maybe 4k7 ohms and a DC blocking capacitor fed from the cathode feeding into 50R coax that then exits the sig gen.
At the external counter end of this coaxyou could put a wideband 50R gain block to drive the counter because the level out of the coax will be quite small if you try this option. But the frequency response of this will be very good as long as the counter end of the coax is terminated in 50R by the wideband amplifier. This passive solution will load the cathode a bit but I'm not sure it will matter much. This method is worth considering if there is a fairly big RF waveform at the cathode. It would be viable if there is maybe 5Vpkpk of RF at the cathode? The 4k7 series resistor and the 50R coax will act as a very lossy potential divider but the wideband gain block can make up for this and drive the counter. I would use a couple of small 2k2 resistors in series to make up the series 4k7 resistor in order to minimise the overall through capacitance here. But 0.25pF package capacitance across the 4k7 will act as a pre emphasis up at VHF so maybe the capacitance doesn't matter here?
__________________
Regards, Jeremy G0HZU |
23rd Mar 2017, 1:33 am | #26 |
Triode
Join Date: Mar 2017
Location: Chelmsford, Essex, UK.
Posts: 38
|
Re: Frequency counters
I'm beginning to lean towards waiting to see how the standard set up works with the Blackstar Meteor 100 when it arrives in a few days time.
I might then perhaps invest in a better signal generator as the results with this one seem to be a bit inconsistent as today, I used 2 different oscilloscopes to try and set the frequency to 100Mhz (sig gen reads 80Mhz) and at that frequency it is giving out 140mv measured by connecting directly to the generators high output sockets, at 1Mhz it gives 3v. I was surprised to note that my Watsu 60Mhz scope actually gave the same readings as my Goldstar 100Mhz did, despite it being pressed into working with a 66% increase in its designed frequency range? |
23rd Mar 2017, 1:50 am | #27 |
Dekatron
Join Date: Apr 2009
Location: Oxfordshire, UK.
Posts: 4,935
|
Re: Frequency counters
I’ve looked at a number of web articles about using basic DFC modules and it seems to be fairly standard practice to put a buffer amp in front of them, and some users also like to add a Schmitt trigger between the amp and the DFC. I think that is there to get a clean, square waveform. Does that sound like a good idea?
B
__________________
Saturn V had 6 million pounds of fuel. It would take thirty thousand strong men to lift it an inch. |
23rd Mar 2017, 2:05 am | #28 | |
Dekatron
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 3,077
|
Re: Frequency counters
Quote:
If the source impedance of the sig gen is somewhere approaching 500R and the scope input is 1Meg ohm in parallel with 20pF and you use 50cm of (unterminated) coax then theory suggests that the amplitude will drop by about 18dB at 100MHz even if the sig gen emf is consistent in output level. If you then add another 6dB of droop to allow for rolloff in the 66MHz BW scope when fed with a 100MHz signal then you get 24dB droop or a factor of 16 in droop overall at 100MHz. So a 3V signal appearing on the scope at 1MHz could be expected to fall to 3/16 = 187mV on the scope display when the sig gen is changed to 100MHz if you use the above coax and scope method to measure the sig gen performance. Hope I've got that right, it's late and I'm a bit tired Try repeating the test with a 50R termination at the scope end of the coax. The level will drop a lot at both 1MHz and 100MHz but you should find there is less droop when comparing 1MHz and 100MHz and maybe this droop will be dominated by the droop in the scope at 100MHz?
__________________
Regards, Jeremy G0HZU Last edited by G0HZU_JMR; 23rd Mar 2017 at 2:12 am. |
|
23rd Mar 2017, 2:45 am | #29 |
Triode
Join Date: Mar 2017
Location: Chelmsford, Essex, UK.
Posts: 38
|
Re: Frequency counters
Problem with doing that, I don't have a 50 ohm terminator as yet, I just learning my way around again after many years away from electronics. How I measured it was by using the standard 100Mhz probes clipped directly onto the output terminations of the generator by removing the screw clamps so the ground clip was connected directly to the ground and the probe tip just inserted inside the banana socket on the high out socket. So the only lead, coax or other was the probe lead itself.
Is that not the best method then? Sorry if I might seem to be a little slow on these things, it was around 50 years ago when I was playing around with electronics, so I have a lot to catch up on and re-learn as I go. |
23rd Mar 2017, 2:04 pm | #30 |
Dekatron
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 3,077
|
Re: Frequency counters
If you are doing this with a switchable x1/x10 scope probe set to x1 then I think you will see the level on the 1Meg //20pF scope drop dramatically as you increase the frequency from 1MHz to 100MHz. I'd expect to see it droop by a factor of about 20 if the sig gen source impedance is in the ballpark of 400 ohms.
The other thing worth mentioning is that a scope probe set to x1 isn't the same as a piece of regular 50R coax. The coax used in a scope probe is usually a special lossy/resistive type of coax and it won't have a characteristic impedance of 50R. So my suggested 50R termination test at the scope end won't work very well with a x1 scope probe. So be prepared for disappointment if you do get a 50R termination and try using it with a x1 scope probe rather than regular 50R coax
__________________
Regards, Jeremy G0HZU |
23rd Mar 2017, 8:11 pm | #31 |
Triode
Join Date: Mar 2017
Location: Chelmsford, Essex, UK.
Posts: 38
|
Re: Frequency counters
Everything you and others have said so far does kind of make sense to me but all I can do is use the equipment I have to hand. The real strange thing about the switchable probes is that, with them switched in either 1x or 10x mode, after making the required adjustments on the scopes input selectors, the final measured values were almost identical.
I have today striped down the scopes and carried out re-calibration on the two in question and everything apart from the timebase has been confirmed as being spot on. I can't check the timebase as I don't have access to the required scope calibrator, but the vertical side is fine. Oh well, as I said, lets hold on for the Black Star counter to arrive, it might just work fine with that. |
24th Mar 2017, 2:26 am | #32 |
Dekatron
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 3,077
|
Re: Frequency counters
If this helps have a look at the specs for a typical Tektronix 200MHz x10 scope probe.
http://uk.rs-online.com/web/p/oscill...robes/4810823/ In x1 mode the claimed bandwidth is typically only 6MHz and this will be dependent on the source resistance of the circuit being tested. So in some cases where you are probing a high impedance source, the bandwidth will be even less than 6MHz in x1 mode. This is because the probe has about 100pF input capacitance in x1 mode. Take a look at the specs in the user guide below on page 4. In case you haven't done this already, you have to compensate the probe when in x10 mode as per page 3 of the pdf below. If this isn't done properly then the probe won't have a flat frequency response up into the RF region in x10 mode. It could be that it becomes peaky or droopy at HF and VHF. Get this wrong and the results you measure with the probe in x10 could be very inaccurate/confusing. http://docs-europe.electrocomponents...6b80652196.pdf
__________________
Regards, Jeremy G0HZU |
24th Mar 2017, 12:58 pm | #33 |
Triode
Join Date: Mar 2017
Location: Chelmsford, Essex, UK.
Posts: 38
|
Re: Frequency counters
The probes were fully compensated in 10x mode and they were directly connected to the output sockets of the signal generator, no other length of co-ax or other cable was involved. This was done to minimise the chances of errors creeping in to contaminate the information I give to you guys.
You have so much more experience in this field that I have or ever likely to get. My expertise is in electrical engineering having been in that field all my working life. Now I've effectively it seems, retired I have decided to resurrect my interest in electronics so I'm sort of relying on you guys for advise and support when my knowledge is not sufficient. The interesting point here is that there was discernable difference in the wave form displayed and its measurement when the probes were switched between 1x and 10x once the appropriate volts/div correction had been made on the range switch. |
24th Mar 2017, 1:46 pm | #34 |
Dekatron
Join Date: Jun 2016
Location: London, UK.
Posts: 3,687
|
Re: Frequency counters
Probes and oscilloscopes are mysterious beasts.
This is the best reference on probes, bandwidth, loading etc I am aware of: http://ecee.colorado.edu/~mcclurel/tabcprobes.pdf Also see Linear AN47 which contains some additional information including some of the above excerpted and is a thoroughly good read anyway: https://pdfs.semanticscholar.org/dd0...b5c33c5902.pdf - the portion AN47-7 onwards is absolutely brilliant and in some places quite funny. |
27th Mar 2017, 11:47 pm | #35 | |
Triode
Join Date: Mar 2017
Location: Chelmsford, Essex, UK.
Posts: 38
|
Re: Frequency counters
Quote:
It does show up the errors in the old vernier type of signal generators to be of alarming magnitudes. Thank you. |
|
28th Mar 2017, 9:24 am | #36 |
Dekatron
Join Date: Jun 2016
Location: London, UK.
Posts: 3,687
|
Re: Frequency counters
Excellent. Glad it hear it works. I had one of those counters many years ago and they're very pretty good.
Just a word of caution though: don't leave them the sun - the front panel goes a funny orange colour after a bit. I learned this the hard way! |
29th Mar 2017, 9:38 pm | #37 |
Triode
Join Date: Mar 2017
Location: Chelmsford, Essex, UK.
Posts: 38
|
Re: Frequency counters
Bit late for that I'm afraid the previous owner already did that. Still it works well enough for me and allows me to now get the IF sections correctly aligned by ensuring that I use the dead space between broadcast stations, something that was impossible to do with the Tech because the drift on it would have encroached into useable broadcast frequencies.
|
5th Apr 2017, 10:31 pm | #38 |
Triode
Join Date: Mar 2017
Location: Chelmsford, Essex, UK.
Posts: 38
|
Re: Frequency counters
Update on the project so far, after playing around with the Black Star frequency counter in conjunction with the sig gen, I discovered that switching between ext and int modulation the frequency on int mod was miles out of calibration. This was checked on a scope and the waveform was nothing at all like an AM signal should look like either. So I stripped the sig gen down and discovered that the lead from a cap was almost grounded before it reached the range switch. This was re-routed and the frequency checked again. Result, no difference between int and ext modulation setting and the scope showed something resembling an AM signal now. To try and increase the output voltage available, I tried a 390 ohm resistor between the 1k pot and ground thinking that it should help to provide something in the order of a 40% increase in the voltage available, this seems to be the case so far.
The chinese frequency counter also arrived so I quickly coupled that up and it works very well and the accuracy against the Black Star is also very good indeed so I'm still thinking of connecting this directly to the sig gen and mounting it on the top, freeing the Black Star for other uses. Now I just need to think about a power source for the counter, I don't think the 6.3v heater winding is man enough to power it as well, seeing as the 6.3v is already down to 6.15v and after rectifying and smoothing, this is going to be in the realms of 7.3vdc. Counter needs 9v to 15v @160ma, so maybe an external power brick is the way to go, any thoughts? |
6th Apr 2017, 9:24 am | #39 |
Dekatron
Join Date: Jun 2016
Location: London, UK.
Posts: 3,687
|
Re: Frequency counters
Makes sense. I'd make sure it's explicitly not a switching power supply. Once those get inside anything RF, they are nothing but trouble.
TBH I'd probably grab an AC wall wart (the things are like swiss army knives!), add a jack to the back of the unit and stick a 7812+bridge+smoothing cap arrangement inside the unit. Or another small 6va transformer if there's enough room in it. |