|
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. |
|
Thread Tools |
18th Jan 2019, 8:23 pm | #1 |
Dekatron
Join Date: Nov 2006
Location: London, UK.
Posts: 3,496
|
A little Friday evening puzzle for light entertainment
Here's a light puzzle for a cold Friday evening.
In the attached circuit - most of a Dansette 50 record player, only tone controls omitted - you have replaced the grid coupling capacitors marked 47K on sight (for any beginners, that's an old fashioned way of saying 0.047uF), but haven't yet powered up the set. You are checking for any out of spec components before you do so. R5, marked as 470K in the anode of the phase splitter triode V1A, measures open circuit. (More accurately, it behaves more like a capacitor than a resistor, fleetingly measuring 15 Megohms and then in an instant, zooming immediately higher towards O/c). So, take it as O/c... R7 (missing label) is its neighbour in the anode of the other triode half of an ECL82, V2A. The value, (the one that is not marked) is 615K. It was originally 560K, so expectedly within tolerance for its age. Suppose (hypothetically) you temporarily forget about the out of spec resistor and the following day, power up the set and put on a record. What happens?
__________________
Al Last edited by Al (astral highway); 18th Jan 2019 at 8:32 pm. |
18th Jan 2019, 8:48 pm | #2 |
Dekatron
Join Date: Apr 2011
Location: Cornwall, UK.
Posts: 13,454
|
Re: A little Friday evening puzzle for light entertainment
I don't think V1a is the phase splitter, I would have thought that the unmarked resistor in the anode circuit of V2a would be a similar value to the one in it's cathode circuit seeing as V2a appears to be the phase splitter......
Lawrence. |
18th Jan 2019, 8:58 pm | #3 |
Dekatron
Join Date: Nov 2006
Location: London, UK.
Posts: 3,496
|
Re: A little Friday evening puzzle for light entertainment
Yes Lawrence , that’s a typo, thanks ... I know V2A is the phase splitter, made a slip while posting.
The puzzle still stands.
__________________
Al |
18th Jan 2019, 9:01 pm | #4 |
Dekatron
Join Date: Feb 2007
Location: Lynton, N. Devon, UK.
Posts: 7,088
|
Re: A little Friday evening puzzle for light entertainment
V2a is a concertina phase splitter, thought the cathode resistor is unusually returned not direct to chassis.
With R5 open-circuit, V1a will pull V2a grid very low, so V2a will pass very little current. At full volume setting, there might be a bit OD signal at V1a cathode (due to grid-cathode of V1a acting as a diode) but even that will be heavily attenuated by the time it feeds through the resistors to V2a cathode. And there'll be nothing at V2a anode. So, you may get the teeniest smidgen of sound, but it'll be pretty distorted. |
18th Jan 2019, 10:21 pm | #5 |
Dekatron
Join Date: May 2008
Location: Derby, UK.
Posts: 7,735
|
Re: A little Friday evening puzzle for light entertainment
Hmmmm ..... If the tiny pulse of current from the stray capacitance charging is enough to get V1a conducting, then it will pull V2a grid low. V2a cathode is at a higher potential than V1a cathode due to R6, so V2a will be cut off, and certainly not doing any phase splitting. V1b grid will be pulled low by R10, so V1b will not be conducting at all. There will be no current in the top half of T1 primary. V1a grid and cathode will act as a diode and pass a half-wave rectified copy of the audio signal, via R6 and the un-numbered 47nF capacitor, to V2b grid. This will modulate V2b anode, so passing a current through the bottom half of T1 primary and inducing a current in the secondary.
So you might just be able to hear a weak, scratchy, distorted version of the record you were trying to play.
__________________
If I have seen further than others, it is because I was standing on a pile of failed experiments. |
18th Jan 2019, 10:32 pm | #6 | ||
Dekatron
Join Date: Nov 2006
Location: London, UK.
Posts: 3,496
|
Re: A little Friday evening puzzle for light entertainment
Hi Peter/ Kalee20
Just to say, in real life, there was no forgetting involved; there was a very different and quite magical scenario! It's described at the top of this thread. The set came to me to turn around in a really short window at excitingly late notice, untested, from an auction win. So an achievable goal was to bring the set to just good and safe enough condition for the owner to gift to someone loved for Christmas. I reformed the electrolytics for a while, checked HT, replaced the two named coupling caps on sight, sniffed around for any potentially destructive or dangerous issues, passed the output transformer as AOK, cleaned/ de-rusted around the valve sockets and residue from the pins and remediated some dodgy re-wiring attempts in a couple of places. I then fired up the set... ...and produced a reasonable volume, albeit with noticeable distortion. As I say, this was under intense time pressure, fitting it around an already very busy schedule, but really wanting to deliver. I was now out of time, so returned the set to its owner so it could be given as a Christmas present, just as promised, before being returned back to me for deeper work in the New Year. Quote:
Even though the resistor in the anode of V1A was already open-circuit, it didn't yet occur to me, because...surprise, surprise, there was reasonable volume from the set, enough to hear from the next room (albeit intermittently because of a separate volume control/ tone control issue.) I didn't have time to measure the output of the cartridge, which was original and had a very worn integral stylus, so I didn't know how much of the signal quality to attribute to this, but I did wonder at the time that it might easily be partly or largely culpable. Quote:
When I got the set back in the New Year, I had unpressured time to get the chassis out again, to look at things more deeply, and to return the set to the highest function possible. Then- measuring around systematically to safeguard original components where possible- I found to my surprise the astonishingly kaput R5. I'd heard of resistors failing in exactly this position before but in those cases, sets were reported as being non-functioning. I did imagine, for want of any other explanation, that V1A was in effect in diode mode. But the relatively good volume was a real surprise. I needed the conviction of measuring the component value again out of circuit, which is when I noted its capacitor-like behaviour. Of course, resistors do go all old and can resemble a capacitor in their death throes, but I'd not previously encountered the way this one was initially (for a very short time) masked like this.
__________________
Al Last edited by Al (astral highway); 18th Jan 2019 at 10:59 pm. Reason: Clarity ...and chain of events in sequence |
||
18th Jan 2019, 10:53 pm | #7 | |
No Longer a Member
Join Date: Oct 2016
Location: Maroochydore, Queensland, Australia.
Posts: 2,679
|
Re: A little Friday evening puzzle for light entertainment
Quote:
This might be one of those cases where a component test on a meter does not give an accurate idea of how the component behaves with a high voltage applied. In that circuit position all that is required to get any reasonable signal output is the junction of the grid and plate, has the small space charge bled away and the plate voltage pulled up probably to only 5 or 10 volts with some tiny resistor current. So at voltage, that resistor was probably not truly O/C as a meter check suggested. One way to check the theory that it might have been the open circuit resistor's capacitance that somehow caused it to produce sound when it shouldn't, would be to replace the resistor with a capacitor, and see what happens. This has always always been the curse of checking capacitors below their rated operating voltage for leakage too and then trying to conclude they are OK in their actual circuit where the behavior can change due to the applied voltage field. I guess it shows that when we test parts a lot of the time with meters the results do not cover every scenario the component encounters in use, especially for higher voltage circuits. Last edited by Argus25; 18th Jan 2019 at 11:00 pm. |
|
18th Jan 2019, 11:19 pm | #8 | |
Dekatron
Join Date: Apr 2011
Location: Cornwall, UK.
Posts: 13,454
|
Re: A little Friday evening puzzle for light entertainment
Quote:
Lawrence. Last edited by ms660; 18th Jan 2019 at 11:26 pm. Reason: addition |
|
18th Jan 2019, 11:20 pm | #9 | |
Dekatron
Join Date: Nov 2006
Location: London, UK.
Posts: 3,496
|
Re: A little Friday evening puzzle for light entertainment
Quote:
And interesting also your note about the inconsistent/ misleading behaviour of such a resistor at lower test voltages. That's a really good take-away. I like it. I certainly learned something about the deeper nuances of measurement - and its relationship to diagnostics - in this case and it's an area of practice that does fascinate me.
__________________
Al |
|
19th Jan 2019, 12:23 am | #10 |
Dekatron
Join Date: Nov 2006
Location: London, UK.
Posts: 3,496
|
Re: A little Friday evening puzzle for light entertainment
Julie, thank you for taking the time to post your very detailed, crystal-clear chain of events /analysis! That is exactly the expected outcome and you can imagine , I was so surprised to find the set defying all common reason.
Hugo’s theory makes great sense. Lawrence, that’s a very interesting application, within a valve TV, to find the same underlying problem with a different outcome... thanks!
__________________
Al |
19th Jan 2019, 3:10 pm | #11 |
Tetrode
Join Date: Jan 2015
Location: St Jean d'Angely, Charente-Maritime, France
Posts: 81
|
Re: A little Friday evening puzzle for light entertainment
Do R4 and R6 constitute bootstrapping to increase the gain?
Bob |
19th Jan 2019, 8:44 pm | #12 |
Dekatron
Join Date: Apr 2012
Location: Wiltshire, UK.
Posts: 14,007
|
Re: A little Friday evening puzzle for light entertainment
My scenario:
R5 going open-circuit would have meant the grid of V2A dragged down-to-ground by V1A and so V2A's anode would have soared-away-to-HT+ In turn this would have overwhelmed the insulation-resistance of an elderly C8 and a high DC voltage would then have arrived at the grid of V1B, which would have turned hard on and either suicidally-died or have chosen to take a section of T1-primary down with it. |
20th Jan 2019, 11:04 am | #13 |
Heptode
Join Date: Feb 2010
Location: Heysham, Lancashire, UK.
Posts: 669
|
Re: A little Friday evening puzzle for light entertainment
C8 gets full HT every time the amp is switched on (until V2 warms up), or if V2 failed or was unplugged. It needs to be checked for leakage at full HT just like any other anode - grid coupling cap (aka “that cap”).
Stuart |
21st Jan 2019, 10:31 am | #14 | |
Dekatron
Join Date: Feb 2007
Location: Lynton, N. Devon, UK.
Posts: 7,088
|
Re: A little Friday evening puzzle for light entertainment
Quote:
In your original post, you described the resistor as looking like a capacitor when metered, I assumed you were measuring in-circuit and were actually seeing the effects of a following coupling capacitor - particularly as you said 'So, take it as O/C.' But looking again at the circuit, I see that there's no coupling capacitor as it's direct coupled - and I took this on board for part of my post #4 anyway, so at this point I'm giving myself a rap over the knuckles for lack of consistency in thought processes! If R5 is not truly O/C then it's true that not very many microamps worth of pull-up will get some sort of life from it. Maybe there's parallel leakage due to the valveholder. Maybe there is electronic coupling within the valve, between triode and pentode section! Maybe the resistor 'comes a bit good' with significant volts across it. As you have measured the resistor out of circuit, it does point towards the resistor doing strange things (I've yet to see a resistor doing this, but I distrust many carbon composition types on sight). It does show, in this game you can't take much for granted! |
|
23rd Jan 2019, 7:26 pm | #15 | |
Dekatron
Join Date: Nov 2006
Location: London, UK.
Posts: 3,496
|
Re: A little Friday evening puzzle for light entertainment
Quote:
Kalee20/ Peter... It certainly does show that not much can be taken for granted!!! I have now replaced the ‘ capacitor in resistor’s clothing’ in the anode of the first triode . There were some other out-of-tolerance candidates, including the -ve feedback resistor, but pretty much as expected for a set 54 years old , and nothing compared to the extraordinary behaviour of R5! The unmarked resistor in the anode of V2A was 250K, design is 180K, 10%, so fixed that. The volume available now is astonishing, even well below max, and the sound is really bright and clear, with minimal distortion. I had already replaced the vol control, and the native bass is wonderful without any additional correction mods. I’m getting a good impression of what this sounded like from the factory gate. The owner seems delighted . Now I’m just waiting for a nice new cartridge so I can work on restoring the add-on companion amplifier for stereo!
__________________
Al Last edited by Al (astral highway); 23rd Jan 2019 at 7:54 pm. |
|