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 > Other Discussions > Homebrew Equipment

Notices

Homebrew Equipment A place to show, design and discuss the weird and wonderful electronic creations from the hands of individual members.

Closed Thread
 
Thread Tools
Old 19th Apr 2021, 12:31 pm   #1
David G4EBT
Dekatron
 
David G4EBT's Avatar
 
Join Date: Feb 2003
Location: Cottingham, East Yorkshire, UK.
Posts: 5,736
Default Buffer Amplifier for Digital Frequency Counter

Every now and then as the mood takes me, I allow myself to be distracted and put aside whatever project I'm on with, to have a little homebrew frolic.

When recently trying to check the frequency of a low level signal, my (homebrew) DFM wasn't sensitive enough to provide a reading. Not something I need to do often, but I thought I might have a look around for a circuit for a simple buffer amplifier. Many years back in the mid 90s when I was still interested in amateur radio, I'd built a CM Howes CBA2 Buffer Amplifier from a kit, (to drive a digital display for a homebrew direct conversion receiver), which ended up gathering dust. In 2015, I sold it to a forum member for a nominal sum:

https://www.vintage-radio.net/forum/...d.php?t=118215

That circuit used a BSX960 and a BSX20 transistor, and the spec stated it covered 500 kHz to 30 MHz and the gain was 10dB. (4dB down at 30 MHz).

Rather then emulate Dave Howes CBA2 circuit, I had a poke around on the net and came across a circuit which uses an MPF102 and two 2N2222 transistors, so thought I'd design a little PCB for it to see how it performs. If performed well enough up to 15 MHz then sharply falls off. I thought I'd redesign the PCB to be smaller and instead of having tracks on the underside, would use 'blocks' as I've done with other homebrew designs, and would also create a ground plan on the top surface. This 'MK2' version does have a little more gain, but as with the 'MK1' performance falls off a cliff at 15 MHz.

As I've said before, my practical skills exceed my technical skills by a good measure, so I've no idea whether the frequency limitation stems from the circuit itself or my PCB layouts, which broadly follow the layout of the circuit. I don't think it's down to the choice of the transistors which have a FT well into the VHF region.

I've attached the circuit, some pics of the PCBs and some 'scope traces to illustrate the performance at various frequencies In this post, the pics related to the 'MK1' version. Any thought on the circuit or layout which might up the performance above 15 MHz would be appreciated.

Pic 1: For information - the CM Howes Circuit (1989).
Pic 2: The circuit that I found in internet.
Pic 3: My 'MK1' component layout, tracks and negative UV mask.
Pic 4: The MK1 PCB built and placed in the project box it was designed to fit into.
Pic 5: A 'scope trace of the MK1 PCB at 100kHz showing 200mV in, and 3 V out.

Some more pics to follow.
Attached Thumbnails
Click image for larger version

Name:	CM Howes CBA2 Buffer Amplifier..jpg
Views:	858
Size:	53.3 KB
ID:	232112   Click image for larger version

Name:	Buffer Amp  Circuit V2.jpg
Views:	1029
Size:	54.5 KB
ID:	232113   Click image for larger version

Name:	MK1 PCB AND NEGATIVE UV  MASK.jpg
Views:	391
Size:	50.7 KB
ID:	232114   Click image for larger version

Name:	MK1 in box.jpg
Views:	295
Size:	50.3 KB
ID:	232115   Click image for larger version

Name:	MK1 200mV in 3V out at 100kHz.jpg
Views:	287
Size:	72.8 KB
ID:	232116  

__________________
David.
BVWS Member.
G-QRP Club member 1339.
David G4EBT is online now  
Old 19th Apr 2021, 12:47 pm   #2
David G4EBT
Dekatron
 
David G4EBT's Avatar
 
Join Date: Feb 2003
Location: Cottingham, East Yorkshire, UK.
Posts: 5,736
Default Re: Buffer Amplifier for Digital Frequency Counter

Two further 'scope pics of the MK1 version, and three pics of the MK2 PCB.
(The circuit and component layout is identical in both versions).

The MK1 PCB was 9cms long x 4.5cms high - the MK2 is 6.5CMs long x 3.5cms high.

Pic 1: MK1: 400mV in, 800mV out at 5 MHz.
Pic 2: MK1: 200mV in, 400mV out at 10 MHz.

Pic 3: Component overlay of the MK2 version(same as MK1, but smaller PCB). Two versions of the MK2 PCB design, one with tracks (not used), the other with blocks on the underside and a ground plane on the top surface of the PCB.
Pic 4: The etched underside of the PCB.
Pic 5: The top ground plane, showing clearance around the non-grounded holes.

Some more pics to follow.
Attached Thumbnails
Click image for larger version

Name:	MK1 400mV in 800mV out at 5 MHz.jpg
Views:	198
Size:	69.4 KB
ID:	232117   Click image for larger version

Name:	MK1 200mV in 400 mV out at 10 MHz.jpg
Views:	184
Size:	70.1 KB
ID:	232118   Click image for larger version

Name:	PCB  component overlay, tracks and top layer  MK2 PCB.jpg
Views:	425
Size:	65.3 KB
ID:	232119   Click image for larger version

Name:	MK2 Underside of PCB.jpg
Views:	219
Size:	41.6 KB
ID:	232120   Click image for larger version

Name:	MK2 Top Ground Plane.jpg
Views:	150
Size:	52.5 KB
ID:	232121  

__________________
David.
BVWS Member.
G-QRP Club member 1339.
David G4EBT is online now  
Old 19th Apr 2021, 1:29 pm   #3
David G4EBT
Dekatron
 
David G4EBT's Avatar
 
Join Date: Feb 2003
Location: Cottingham, East Yorkshire, UK.
Posts: 5,736
Default Re: Buffer Amplifier for Digital Frequency Counter

Some more pics.

Pic 1: Top (ground plane) side of the MK2 built PCB.
Pic 2: MK2 100kHz 5mV p-p input, 700mV output.
Pic 3: MK2 1 MHz 5mV p-p input, 300mV output.
Pic 4: MK2 5 MHz 5mV p-p input, 200 mV output.
Pic 5: MK2 10 MHz 50mV p-p input, 200 mV output.

As I said at the outset, at lower frequencies the amplifier works well, but performance falls off sharply beyond 10 MHz, and by 15 MHz, there is no gain. Any thought on how I might be able to improve the HF performance would be appreciated. Just a little homebrew distraction from other things - not having a buffer amplifier hasn't left a gap in my life, but it's been an interesting little project, (for me, at any rate).
Attached Thumbnails
Click image for larger version

Name:	MK2 PCB.jpg
Views:	195
Size:	66.8 KB
ID:	232128   Click image for larger version

Name:	MK2 100kHz 5mV in 700mV out.jpg
Views:	121
Size:	86.4 KB
ID:	232129   Click image for larger version

Name:	Mk2 1 MHz 5mV in 300mV out.jpg
Views:	97
Size:	56.5 KB
ID:	232130   Click image for larger version

Name:	MK2 5 MHz 5mV in 200mV out.jpg
Views:	110
Size:	74.3 KB
ID:	232131   Click image for larger version

Name:	MK2 10 MHz 50mV in 200 mV out.jpg
Views:	98
Size:	59.6 KB
ID:	232132  

__________________
David.
BVWS Member.
G-QRP Club member 1339.
David G4EBT is online now  
Old 19th Apr 2021, 1:52 pm   #4
ms660
Dekatron
 
ms660's Avatar
 
Join Date: Apr 2011
Location: Cornwall, UK.
Posts: 13,454
Default Re: Buffer Amplifier for Digital Frequency Counter

15MHz aside, I suspect your schematic needs altering.

EDIT: Just seen the circuit board, the ground plane will be one plate of various capacitors.

Lawrence.

Last edited by ms660; 19th Apr 2021 at 2:13 pm.
ms660 is offline  
Old 19th Apr 2021, 2:31 pm   #5
Terry_VK5TM
Nonode
 
Terry_VK5TM's Avatar
 
Join Date: Oct 2010
Location: Tintinara, South Australia, Australia
Posts: 2,324
Default Re: Buffer Amplifier for Digital Frequency Counter

Two things I can see, one Lawrence has covered and the second is the 10uF coupling cap to the last stage will be causing problems.

It is common to see a 10 or 100nF cap in parallel with that 10uF to increase the bandwidth.
__________________
Terry VK5TM
https://www.vk5tm.com/
Terry_VK5TM is offline  
Old 19th Apr 2021, 3:13 pm   #6
ms660
Dekatron
 
ms660's Avatar
 
Join Date: Apr 2011
Location: Cornwall, UK.
Posts: 13,454
Default Re: Buffer Amplifier for Digital Frequency Counter

Not to mention the output coupling capacitor.

Lawrence.
ms660 is offline  
Old 19th Apr 2021, 3:18 pm   #7
Radio Wrangler
Moderator
 
Radio Wrangler's Avatar
 
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,798
Default Re: Buffer Amplifier for Digital Frequency Counter

I prefer that circuit to the Howes one. The howes circuit has the input linked straight to the gate of a MOSFET without any protection.

In the circuit you built, the 2N2222 common emitter amplifier has the emitter decoupled in an attempt to screw the last bit of gain out of it, which is not good for bandwidth. A better gain stage would be a cascode pair which should push the roll-off point up quite a bit.

Also the Zout of the Jfet may be a bit high and a limiting factor. An emitter follower to help with the loading of that flat out gain stage may help.

David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done
Radio Wrangler is online now  
Old 19th Apr 2021, 3:55 pm   #8
Bazz4CQJ
Dekatron
 
Bazz4CQJ's Avatar
 
Join Date: Apr 2009
Location: Oxfordshire, UK.
Posts: 4,923
Default Re: Buffer Amplifier for Digital Frequency Counter

Back in 2016, Skywave (Al) spent a good deal of time working on an amplifier circuit for use as an RF probe which used a 2N4416 at the front and two additional transistors, which was flat to about 100Mhz (an evolution of a Marconi design).

See post #22 here http://www.vintage-radio.net/forum/s...977#post896977. That might be worth looking at, and would certainly work well at 30Mhz.

Having spent some time looking at pre-amps for using with DFM's, quite a few incorporate a chip operating as a Schmidt trigger, so the DMF is fed with a square wave, which is said to be useful.

B
__________________
Saturn V had 6 million pounds of fuel. It would take thirty thousand strong men to lift it an inch.
Bazz4CQJ is offline  
Old 19th Apr 2021, 5:31 pm   #9
David G4EBT
Dekatron
 
David G4EBT's Avatar
 
Join Date: Feb 2003
Location: Cottingham, East Yorkshire, UK.
Posts: 5,736
Default Re: Buffer Amplifier for Digital Frequency Counter

Quote:
Originally Posted by ms660 View Post
15MHz aside, I suspect your schematic needs altering.

EDIT: Just seen the circuit board, the ground plane will be one plate of various capacitors.

Lawrence.
Thanks for reading the thread and for your feedback Lawrence, which makes sense. Likewise, the comment on the lack of an output coupling cap which could have been added to the PCB, (albeit the input to my counter has one). I guess the ground plane is both problematic and surplus to requirements.
__________________
David.
BVWS Member.
G-QRP Club member 1339.
David G4EBT is online now  
Old 19th Apr 2021, 5:58 pm   #10
Chrispy57
Hexode
 
Join Date: Mar 2020
Location: Rochdale, Greater Manchester, UK.
Posts: 439
Default Re: Buffer Amplifier for Digital Frequency Counter

Hi David,
lots of more qualified comments already offered, but I wondered if the R4/C4 corner frequency of 12.8kHz may be a limiting factor. Also, whether the two diodes across the gate of the FET may have been mis-drawn too?

Cheers
Chris
Chrispy57 is offline  
Old 19th Apr 2021, 6:49 pm   #11
David G4EBT
Dekatron
 
David G4EBT's Avatar
 
Join Date: Feb 2003
Location: Cottingham, East Yorkshire, UK.
Posts: 5,736
Default Re: Buffer Amplifier for Digital Frequency Counter

Quote:
Originally Posted by Terry_VK5TM View Post
Two things I can see, one Lawrence has covered and the second is the 10uF coupling cap to the last stage will be causing problems.

It is common to see a 10 or 100nF cap in parallel with that 10uF to increase the bandwidth.
Thanks Terry,

That's easy enough to do.

Overall, from the feedback so far, maybe this wasn't the best circuit to have stumbled across?

Something that occupied my time indoors when it was too cold to be outdoors jet-washing my drive and patio.

All done now that the sun has got his hat on! (Good fun - I think I must have been deprived of water pistols as a child).
__________________
David.
BVWS Member.
G-QRP Club member 1339.
David G4EBT is online now  
Old 19th Apr 2021, 7:40 pm   #12
G0HZU_JMR
Dekatron
 
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 3,077
Default Re: Buffer Amplifier for Digital Frequency Counter

I've not played with a 2N2222A in many years but I think this old RF/switching transistor won't perform that well in your circuit.

Looking at the datasheet the feedback capacitance at a collector voltage of about 3V is about 4pF. If I assume 2mA collector current the final stage probably has a (low frequency) voltage gain of about Av = -70. So the Miller capacitance is going to be (Av +1) x 4pF which is nearly 300pF. This stage alone will restrict the bandwidth a lot because of this. The previous 2N2222A stage will also have a fair bit of voltage gain (at low frequencies) and so this will generate a fairly significant Miller capacitance.

I also expect the input JFET stage to produce a lot of negative resistance despite the series 1k resistor at the input. This isn't a great circuit.

Maybe the best thing to do is list what the requirements are for your buffer circuit and then make a circuit that meets those requirements. What you end up with might still be able to be squeezed onto your existing PCB.
__________________
Regards, Jeremy G0HZU
G0HZU_JMR is offline  
Old 19th Apr 2021, 8:29 pm   #13
David G4EBT
Dekatron
 
David G4EBT's Avatar
 
Join Date: Feb 2003
Location: Cottingham, East Yorkshire, UK.
Posts: 5,736
Default Re: Buffer Amplifier for Digital Frequency Counter

Quote:
Originally Posted by Chrispy57 View Post
Hi David,
lots of more qualified comments already offered, but I wondered if the R4/C4 corner frequency of 12.8kHz may be a limiting factor. Also, whether the two diodes across the gate of the FET may have been mis-drawn too?

Cheers
Chris
Thanks for reading the thread Chris and for you comments, and thanks also to everyone else who has responded.

Re the diodes, yes, it makes no sense to have two in parallel. As has been suggested, I've amended the circuit to revere the polarity of D1. I've also added a .01uF cap across C5, and added a .01uF cap on the output. As to R4/C4, given that my main concern is that the amplifier frequency response drops sharply after 10 MHz, do you have any thoughts as what values might improve matters please?

The amended circuit is attached.

If nothing else, it will teach me to not lift random circuits off the net and to expect them to meet my needs.

Just another example if one were needed, of what my end-of-school report prophesised way back in 1954: "He'll be good with his hands".

Oh well: "Use what talents you possess - the woods would be very silent if the only birds that sang were those that sang the best".

Henry Van Dyke.
Attached Thumbnails
Click image for larger version

Name:	Updated and corrected circuit 19 April 2021.jpg
Views:	1873
Size:	68.7 KB
ID:	232175  
__________________
David.
BVWS Member.
G-QRP Club member 1339.
David G4EBT is online now  
Old 19th Apr 2021, 8:45 pm   #14
Radio Wrangler
Moderator
 
Radio Wrangler's Avatar
 
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,798
Default Re: Buffer Amplifier for Digital Frequency Counter

Another thing is that the second transistor is biassed from the first transistor's gate characteristic, which is notoriously uncertain from device to device, s oquiescent currents could be all over the shop. Something a bit more RFy than a 2N2222 would also be a good move and an easy swap. These super-simple bias schemes save a few resistors (which are easily visible) but often add the need to use selected transistors (which is invisible right until it doesn't work)

David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done
Radio Wrangler is online now  
Old 20th Apr 2021, 12:13 am   #15
G0HZU_JMR
Dekatron
 
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 3,077
Default Re: Buffer Amplifier for Digital Frequency Counter

One other thing is that it isn't clear how it is being tested because the photo shows a run of coax going up to a counter. I looked up the circuit for that PW counter and it looks like it has a BJT configured as a common emitter amplifier on the HF port. This will probably have a fairly capacitive input impedance up at HF/VHF.

This will limit the HF performance because there is also the capacitance of the unterminated coax. This is all being driven by a 2N2222A BJT with a 1k collector resistor at the buffer amplifier output. That will create a lowpass filter. Is the circuit being tested in isolation or in the system with the counter?
__________________
Regards, Jeremy G0HZU
G0HZU_JMR is offline  
Old 20th Apr 2021, 12:46 am   #16
G0HZU_JMR
Dekatron
 
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 3,077
Default Re: Buffer Amplifier for Digital Frequency Counter

Looking closer at the HF input port of the PW Robin circuit it might be the case that this lowish input impedance needs to be driven by a fairly low impedance at the buffer end of the coax. This assumes a typical coax run of maybe 60cm feeding into the counter. The input impedance of the coax feeding into the counter could easily be the equivalent of 100R in parallel with 100pF by 30MHz. When this heavy loading gets presented to the collector of the output BJT of the existing buffer circuit I think this will spoil the system gain up at 30MHz.
__________________
Regards, Jeremy G0HZU
G0HZU_JMR is offline  
Old 20th Apr 2021, 3:23 am   #17
ortek_service
Octode
 
ortek_service's Avatar
 
Join Date: May 2018
Location: Northampton, Northamptonshire, UK.
Posts: 1,392
Default Re: Buffer Amplifier for Digital Frequency Counter

I wonder if anyone's tried simulating the circuit in the free LTspice ?
- Or maybe the Micro-Cap Simulator software (made available free > 1yr ago for apparently a limited time, but still there), as mentioned recently here: https://www.vintage-radio.net/forum/...56#post1364956 (Where this Preamp circuit also appears)

And you could then try a variety of different loads on it.

I've done this a lot for OpAmp circuits, but not tried a transistor one yet, so would be nice to have a go.

If the MPF102 (Obsolete?) / 2N2222 (quite commonly used in quite a few HF & VHF circuits, even though much higher frequency ones are are available. But would be perfectly-adequate if being used as an emitter-follower) transistor are not in their libraries, then could probably find something reasonably-close.


I do quite like the CM Howes circuit with an RF dual-gate MOSFET, as can vary gain by altering the DC-biased gate voltage.
And it does appear they've tried to ensure a proper 50R output match, using an emitter follower but also a matching transformer (that could limit B/W & not sure what the spec. of that was if you had to wind your own)

You could probably also do this with an IC, like the old Plessey RF amp ones, or a Video OpAmp. And Analog Devices do DC-150MHz precision differential amps with accurate gain control, but maybe less liable to have these more custom parts lying around.

Last edited by ortek_service; 20th Apr 2021 at 3:49 am.
ortek_service is offline  
Old 20th Apr 2021, 3:38 am   #18
ortek_service
Octode
 
ortek_service's Avatar
 
Join Date: May 2018
Location: Northampton, Northamptonshire, UK.
Posts: 1,392
Default Re: Buffer Amplifier for Digital Frequency Counter

Quote:
Originally Posted by ms660 View Post
15MHz aside, I suspect your schematic needs altering.

EDIT: Just seen the circuit board, the ground plane will be one plate of various capacitors.

Lawrence.
And it looks like on many parts that should have a connection to ground, this ground connection actually only passes through non-clearance holes (and so may only just touch the top groundplane) on the top groundplane to the underside ground track, So they should really be soldered to the top plane, if using a PCB without thru-plating on these holes. Otherwise, ground tracking / planes can be more like a small antenna.
But would probably get away with a lot at HF, as long as transistors can't oscillate at much higher frequencies.
ortek_service is offline  
Old 20th Apr 2021, 12:39 pm   #19
David G4EBT
Dekatron
 
David G4EBT's Avatar
 
Join Date: Feb 2003
Location: Cottingham, East Yorkshire, UK.
Posts: 5,736
Default Re: Buffer Amplifier for Digital Frequency Counter

Quote:
Originally Posted by ortek_service View Post
I wonder if anyone's tried simulating the circuit in the free LTspice ?
- Or maybe the Micro-Cap Simulator software (made available free > 1yr ago for apparently a limited time, but still there), as mentioned recently here: https://www.vintage-radio.net/forum/...56#post1364956 (Where this Preamp circuit also appears)

And you could then try a variety of different loads on it.
A good point - I posted both the information about my long-winded way of using MS PAINT to draw circuits and create PCB artwork and UV masks (post #47), and the link to Micro-Cap in post #49. But there is a steep learning curve in all these things, and with only limited hobby time, the 'law of diminishing returns' soon sets in. That's why I've yet to master KiCad or Micro-Cap, and is what's happening with this little amplifier project. I had all the components to hand, and at first sight it looked a simple little circuit to build, and useful to have, albeit it would be little used.

As to the ground plane, looking back, I don't know what possessed me.

As Lawrence remarked, it's the equivalent of adding capacitors all over the non-grounded pads of the PCB. (I've checked the capacitance per sq cm of the FR4 laminate that I used, and it's 3 pF/sq cm). That said, the second version performs better than the first (maybe because the PCB is smaller?), but as I said earlier, neither of them work above 15 mHz.

The plain fact is that the 'opportunity cost' of spending much more time on this project is the other things that I could otherwise be using my scarce hobby time on, such as restoring a radio. And I mustn't overlook taking up the scarce hobby time of other forum members who have been kind enough to read the thread and to offer their advice and suggestions.

I guess that the 'Dunning Kruger Effect' has set in. I've passed the peak of 'Mount Stupid', not really descended into the 'Valley of Despair, but I'm at the bottom of the 'Slope of Enlightenment' which is long, and has a shallow gradient! As Einstein famously said, 'The more I know, the more I realise how much I don't know". Even a dimwit like me can concur with that observation. I call it 'delusions of adequacy'.

I think the reality is that this little buffer-amp isn't a good design and there's little point in developing it further.

I'm most grateful as ever, for the advice, suggestions and encouragement from everyone.
Attached Thumbnails
Click image for larger version

Name:	Dunning Kruger Effect (learning process).png
Views:	163
Size:	76.7 KB
ID:	232209  
__________________
David.
BVWS Member.
G-QRP Club member 1339.
David G4EBT is online now  
Old 20th Apr 2021, 6:58 pm   #20
buggies
Heptode
 
buggies's Avatar
 
Join Date: Jan 2009
Location: West Lothian, UK.
Posts: 760
Default Re: Buffer Amplifier for Digital Frequency Counter

Quote:
Originally Posted by ortek_service View Post
I wonder if anyone's tried simulating the circuit in the free LTspice ?
- Or maybe the Micro-Cap Simulator software (made available free > 1yr ago for apparently a limited time, but still there), as mentioned recently here: https://www.vintage-radio.net/forum/...56#post1364956 (Where this Preamp circuit also appears)
Well - yes, but - it is the first time I have tried Microcap and have never used LTspice in anger - so have attached what I get as a screendump along with the file for someone more sim-skilled to play with...
The three charts are from the source, the first collector and the output.

(Had to zip the .cir file as not an accepted upload format.)
Attached Thumbnails
Click image for larger version

Name:	buffer.jpg
Views:	233
Size:	100.6 KB
ID:	232230  
Attached Files
File Type: zip bufferamp.zip (6.8 KB, 143 views)
__________________
George
buggies is offline  
Closed Thread

Thread Tools



All times are GMT +1. The time now is 9:49 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.