13th Feb 2014, 11:47 am | #41 |
Octode
Join Date: Jun 2004
Location: Reading/Fakenham, UK.
Posts: 1,324
|
Re: MiniMod pantry transmitter
I've just been reading an article in Feb 2014 Practical Wireless (p53) describing in detail the inductors I specified for use in the MiniMod and there's a point which needs highlighting.
The tapping in the tuned winding (L1) is 25 percent from the earthed end. Although I stated this in my BVWS article (and this is also shown in the circuit diagram) it may be worth stating again for anyone else wishing to build one. I expect anyone who has already built one either measured the resistance to identify the right end to earth, or referred to the Spectrum Communications information from the link in the article for pin identification. I've also learnt recently that the BF256B has better defined characteristics than the 2N3819. I've not tried this substitute yet, but will probably be using it for any MiniMods I make from now on. Regards, Ian |
1st Aug 2014, 2:04 pm | #42 |
Heptode
Join Date: Feb 2008
Location: Leven, Fife, Scotland
Posts: 823
|
Re: MiniMod pantry transmitter
I built one of these on veroboard and after correcting a couple of misplaced links it is working fine. I would like to build a pcb version, did I read about a pcb layout for this or am I imagining this. Best regards, Norman.
|
1st Aug 2014, 3:42 pm | #43 |
Octode
Join Date: Jun 2004
Location: Reading/Fakenham, UK.
Posts: 1,324
|
Re: MiniMod pantry transmitter
Yes, at last I have got around to making a pcb layout.
Attached... Regards, Ian |
1st Aug 2014, 4:53 pm | #44 |
Heptode
Join Date: Feb 2008
Location: Leven, Fife, Scotland
Posts: 823
|
Re: MiniMod pantry transmitter
Hi Ian, Many thanks for the pcb layout. I probably could have worked it out but many thanks for taking the time to share your efforts. All the best, Norman.
|
4th Aug 2014, 9:12 pm | #45 |
Octode
Join Date: Jun 2004
Location: Reading/Fakenham, UK.
Posts: 1,324
|
Re: MiniMod pantry transmitter
Important note!
Beware the lead-outs of the FET TR1. I changed to the BF256B with different pin-outs which I believe is what I posted above in the layout attachments. I changed from 2N3819 (in original design) to BF256B because of their slightly better characteristics, but the leads are different. Please also note the info about the coils and their tappings in my post no 41. Thanks. Ian. |
21st Feb 2015, 3:40 pm | #46 |
Octode
Join Date: Nov 2011
Location: Manchester, UK.
Posts: 1,875
|
Re: MiniMod pantry transmitter
Hope it's OK to resurrect this thread...
A long () while ago, Ian sent me a kit of parts for one of these...and Real Life having intervened, I only just got around to building it. My version uses the BF256B JFET. I wonder if anyone who has built one of these would help me out with a little fault finding? My audio amp works, but I have no carrier. When I connect the drain of the FET, the 9v previously appearing at that point drops to 2.3v. I have checked my wiring and checked again...but I am sure I have done something stupid. Thanks Mark |
21st Feb 2015, 6:33 pm | #47 |
Octode
Join Date: Jun 2004
Location: Reading/Fakenham, UK.
Posts: 1,324
|
Re: MiniMod pantry transmitter
Hi Mark,
Let's see... Unfortunately I don't have one here to check, but I'll make some suggestions. Are you absolutely sure there's no RF and it's not falling outside MW? (That's not likely unless there is something wrong with C1 or C2) Check the continuity of L1 and to all its connections. The BF256B FET has different pin-outs to the 2N3819. A quick Google search will bring up the info. Only 2.3V on the drain seems a bit low, but not having one to hand, I can't be sure. Is R3 1k? Have you a signal generator you can inject at Q2 base and sweep through MW to check that bit is working? Check that the tap on L1 is at the earthy end of the coil and connected to Q1 source/C3. Don't short out the unused winding in the L1 can, but connecting one end of it might help frequency stability. Construction technique is uncritical, so that shouldn't be a problem. Ian |
21st Feb 2015, 10:32 pm | #48 |
Octode
Join Date: Nov 2011
Location: Manchester, UK.
Posts: 1,875
|
Re: MiniMod pantry transmitter
Hi Ian
I baulked this morning at pursuing you by PM for 'after-care' - it's very kind of you to get involved >Are you absolutely sure there's no RF and it's not falling outside MW? I've had a look at the aerial on my ancient 'scope - I can see nothing happening. That said I'm an RF virgin - who knows what the IP impedance of the scope is up there etc etc. >Check the continuity of L1 and to all its connections. OK, I'll try that again. I've been round a couple of times, checking especially that all the earth connections were made on all relevant components. That was perhaps easy to overlook, or miss one, owing to veroboard track cutting elsewhere. Seems OK. >The BF256B FET has different pin-outs to the 2N3819. A quick Google search will bring up the info. Yes, I did that - all agreed with the layout diagram you gave me, which cites BF256B specifically. I had to drag out my 25 yr old copy of Horowitz and Hill (my degree had a little electronics in year 1) to check how d-s-g relate to the symbol for a FET! >Only 2.3V on the drain seems a bit low, but not having one to hand, I can't be sure. Is R3 1k? Yes, but I'll check it again. Seems like the FET is pretty 'on' - unless 2.3v is what my cheapo dig multimeter is making of it switching at RF, while set to DC volts? I'm a bit at sea in this part of the cct - I can see there are Ls, Cs, and feedback, which together suggest oscillation, but beyond that I'm hazy! >Have you a signal generator you can inject at Q2 base and sweep through MW to check that bit is working? That's a good idea. Yes - ages ago I acquired an Advance E2, which worked but is collecting dust under the bench. I'll give that a go. >Check that the tap on L1 is at the earthy end of the coil and connected to Q1 source/C3. Don't short out the unused winding in the L1 can, but connecting one end of it might help frequency stability. OK. You mention this up-thread - I'll have another read so as to work out how to check this. >Construction technique is uncritical, so that shouldn't be a problem. Worsening eyesight, shaky hands, out of practice. If I put a picture up when its done, it will only be to encourage the similarly handicapped! Thanks again Mark |
21st Feb 2015, 11:57 pm | #49 |
Octode
Join Date: Jun 2004
Location: Reading/Fakenham, UK.
Posts: 1,324
|
Re: MiniMod pantry transmitter
If you get really stuck, send it to me. For the postage I'll sort it out.
Ian |
22nd Feb 2015, 12:31 am | #50 |
Octode
Join Date: Nov 2011
Location: Manchester, UK.
Posts: 1,875
|
Re: MiniMod pantry transmitter
When I was teaching (not electronics!) I sometimes got to that point too...'oh for goodness sake, give it here'...
So - I inject RF from the E2 straight into the base of Q2, and it works - if I turn the input audio up too far, the op-amp starts to clip, and not long after, though it's hard to see using a music source, it looks like it's running out of modulation depth. Picked up on my radio nice and loud, with an aerial on it about a meter long. So it looks like my problem is around the generation of the carrier. |
22nd Feb 2015, 9:19 am | #51 |
Moderator
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,900
|
Re: MiniMod pantry transmitter
Don't be too sure about the pinout on the FET, a lot of these devices have become so generic that anything with roughly similar characteristics, and often different pinout gets sold as 2N3819 etc.
I usually give new devices a check with an ohmmeter to look for the gate to channel diode to be sure I've got the right leg as the gate. There are also some JFETs with very high current capabilities like 2N4391 which unless properly biased can pull down a PP3. David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done |
22nd Feb 2015, 7:00 pm | #52 |
Octode
Join Date: Nov 2011
Location: Manchester, UK.
Posts: 1,875
|
Re: MiniMod pantry transmitter
I'm rather glad this didn't work straight away, as I'm learning rather more than I would have!
My FET gives me about 150 ohms d-s both ways around, but g-d and g-s are open cct in both polarities. It seems to me this implies a dead transistor - or is it possible to mess this measurement up? I took the measurements with the transistor in circuit, but not powered up. Perhaps I cooked it as I soldered it in. Oh - edit to add - is D1 there to limit the g-s voltage to 0.7v - does this stop it turning on too hard, or have I got the wrong idea? cheers Mark |
22nd Feb 2015, 9:22 pm | #53 | ||
Octode
Join Date: Jun 2004
Location: Reading/Fakenham, UK.
Posts: 1,324
|
Re: MiniMod pantry transmitter
Quote:
Quote:
Ian |
||
22nd Feb 2015, 11:10 pm | #54 |
Heptode
Join Date: Mar 2009
Location: Portsmouth, Hampshire, UK.
Posts: 674
|
Re: MiniMod pantry transmitter
I believe you said you were using a DMM (in post #48), most of these have a test voltage of less than 500mV on the ohms ranges (I've measured 100-300mV on a few of mine).
This is not enough to turn on a silicon device and would account for your open cct g-d and g-s readings. Try again using the 'diode test' function of your DMM. You should see around 700mV g-s and g-d with g positive and over-range with g negative although in circuit D1 will give a reading for g-s with g -ve (again about 700mV). Jim |
23rd Feb 2015, 10:16 am | #55 |
Octode
Join Date: Nov 2011
Location: Manchester, UK.
Posts: 1,875
|
Re: MiniMod pantry transmitter
OK - new FET on order - I'll be back!
|
23rd Feb 2015, 2:51 pm | #56 |
Hexode
Join Date: Apr 2008
Location: Ramsgate, Kent, UK.
Posts: 252
|
Re: MiniMod pantry transmitter
Don't panic! If you are using a FET then there will be very high/infinite resistance between the Gate (G) terminal and both the Drain (D) and Source (S) terminals, that is the way a FET works. The current between the S & D terminals is controlled by the voltage between the G & S terminals, no current flows into the G terminal under normal operation. The resistance between D & S with the G terminal open or connected to S will read quite low. If it is an N-channel device then applying a negative volatge between G & S will INCREASE the D - S resistance in a similar fashion to the Grid voltage in a thermionic triode to which the FET is parametrically related.
It seems that your measurements confirm a that the device is probably working OK. Some respondants in this thread are confusing the FET with a standard bipolar transistor, the normal operating voltage applied between G & S terminals usually lies between about -10V to +10V for a small signal device. Adrian |
23rd Feb 2015, 3:30 pm | #57 |
Octode
Join Date: Jun 2004
Location: Reading/Fakenham, UK.
Posts: 1,324
|
Re: MiniMod pantry transmitter
Yes, hang on there Mark. I tested a JFET with a moving-coil meter and got sensible readings. I didn't realise you had used a digital meter.
With a digital one they were more or less the same as yours, i.e. apparent o/c between gate and source and drain either meter polarity. The FETs are fairly robust, so it's more likely the fault lies elsewhere. Maybe a Veroboard cut that is not quite a cut, or a whisker of cut track touching another track? Ian |
23rd Feb 2015, 3:34 pm | #58 |
Moderator
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,900
|
Re: MiniMod pantry transmitter
What you have is an N-channel Depletion-mode Junction Field Effect Transistor.
Depletion mode means that it is normally ON, and conducts current from drain to sourece and source to drain... both ways. So that's where your 150 Ohms reading both ways comes from. Drain and source connect to opposite ends of a channel of silicon. N-type doped silicon. so current in it is by electron flow. Around it is a sleeve of P-type silicon, or it can be just done on one side. If you make this side electrode negative, it repels electrons and squeezes them into a narrow flow down the centre line of the channel. Make it very negative and it cuts off the flow completely (Pinch-off it's called) Like common valves, depletion mode JFETs require negative bias to set their quiescent current. Notice that with the gate running negative, there is no gater current, bucause the gate to channel contact area is a reverse biased diode junction. There will be a little leakage, but that's it. If you try making the gate go positive, you discover the gate to channel diode will turn on above about 0.6v, just like any other silicon diode. Makers of DVMs like to keep the ohms range voltages lower than this so you can measure resistors in-circuit without diode junctions in semiconductors turning on. This sort of ohmmeter is useless for checking diodes. Some meters have a diode test function which is useable instead. It fires out a small current and meters the voltage drop across whatever is between the prods. This will let you check the presence of the gate-channel diode junction and find the gate pin. Whhat's difficult is separating drain from source... on some FETs the structure is symmetrical and d and g are interchangeable. In other FETs the drain end is better heat-sunk and the electrical performance can be a bit different if you swap them. Hope this clears things up bit David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done |
23rd Feb 2015, 8:59 pm | #59 |
Octode
Join Date: Nov 2011
Location: Manchester, UK.
Posts: 1,875
|
Re: MiniMod pantry transmitter
Thanks for taking such time to type these responses, folks - they're helpfully and carefully pitched for someone of my limited knowledge, and I'm enjoying learning from such thoughtful tutorials.
Thus armed, I'll take the FET out and test again. In fact I'll plug it into some breadboard with a resistor in series with the channel and play about with the gate bias until I am happy I have passed FET101! Got to start somewhere |
1st Mar 2015, 7:31 pm | #60 |
Octode
Join Date: Nov 2011
Location: Manchester, UK.
Posts: 1,875
|
Re: MiniMod pantry transmitter
Well, the breadboard experiment worked, as does the transistor, and having found my wiring error...it now works! Thanks for your help, folks - great fun, and I know a lot more than I would have had it worked straight away.
A couple of remaining questions, if anyone has the stamina: Why use a FET in this application, and not a normal transistor? Why on AM does the signal seem to 'clean up' (whilst dropping in level somewhat) when one tunes to one side of the strongest signal? I don't mean in terms of distortion - more in terms of AF spectral balance where, tuned for maximum volume, the sound appears rather 'middly'. cheers Mark |