|
Vintage Audio (record players, hi-fi etc) Amplifiers, speakers, gramophones and other audio equipment. |
|
Thread Tools |
12th Apr 2018, 11:50 am | #1 |
Moderator
Join Date: Feb 2006
Location: Wembley, Middlesex
Posts: 7,225
|
Hacker GP15 Cavalier
I've repaired loads of these, with no problems except now.
I originally serviced this two years replacing nearly all the resistors and capacitors. It came back because there was a loud howling sound, to cut a long story short, I found that tapping the chassis made the howling come and go, so I removed the valves, sprayed switch cleaner into the holders, let it evaporate then refitted the valves and switched on. And got absolutely nothing. So I've checked all the resistors which are Vishay PR02 series and they are in spec. So I've carried out checks and found: The first ECL86 triode stage does not have any gain, there is no voltage drop across the anode load resistor 220K but there is approx 5V on the cathode. From the service sheet, here are the recorded voltages in red against the published voltages in black HT1 C12 260V 285V HT2 C11 250V 265V HT3 C8 250V 265v V1a pin 9 Anode 145V 259V V2a pin 9 Anode 145V 259V V1b pin 2 Cathode 1.3V 1.27V V2b pin 2 cathode 1.2V 4.8V V1b pin 6 Anode 260V 277V V2b pin 6 Anode 260V 277V V1b pin 7 Cathode 7.8V 9.0V V2b pin 7 Cathode 7.8V 7.8V Voltages taken using the original valves. The same goes for the second triode, there is no voltage drop across the 220K anode load resistor but there is 1.3V volts at the cathode. If you remove the valve, the cathode voltage (measured on the cathode resistor/capacitor) falls to 0V. I've tried new ECL86's which make no difference Now by my reckoning if there are volts on the cathodes, the valves must be conducting, if they are conducting there should be a voltage across the anode load resistors. Apart from the valve holders, everything else is new. Does anyone know what might be going on? |
12th Apr 2018, 12:08 pm | #2 |
Dekatron
Join Date: Sep 2007
Location: Oxfordshire, UK.
Posts: 4,311
|
Re: Hacker GP15 Cavalier
Assuming the triode grids are at 0V (DC) then +5V and +11V on the cathodes will have turned the triodes hard off. So I'm not surprised there's no anode current flowing in them, nor any gain. It's as if the cathode resistors have gone open circuit and the cathode voltages are being held high by the electrolytic caps which would have charged up quickly with an initial burst of current and are now being maintained at volts by an absolutely tiny current (just enough to cope with the caps' leakage) through the triodes.
EDIT: Or, I suppose, current could be leaking from the pentode g2 (pin 3) into the triode cathode resistor (pin 2) and switching the triode off. Cheers, GJ
__________________
http://www.ampregen.com Last edited by GrimJosef; 12th Apr 2018 at 12:23 pm. |
12th Apr 2018, 12:09 pm | #3 |
Dekatron
Join Date: Apr 2011
Location: Cornwall, UK.
Posts: 13,454
|
Re: Hacker GP15 Cavalier
Part of the HT -ve rail floating?
Lawrence. |
12th Apr 2018, 12:35 pm | #4 |
Dekatron
Join Date: Sep 2007
Location: Oxfordshire, UK.
Posts: 4,311
|
Re: Hacker GP15 Cavalier
That's a possibility too. There's a pic of the layout in post #14 here https://www.vintage-radio.net/forum/...ad.php?t=18508. I believe the small 2k2s wired in parallel with the electolytics are the cathode resistors/capacitors in question. If you pull the ECL86s out how many volts do you get on that 8uF 275V capacitor ?
Cheers, GJ
__________________
http://www.ampregen.com |
12th Apr 2018, 12:45 pm | #5 |
Dekatron
Join Date: Sep 2007
Location: Oxfordshire, UK.
Posts: 4,311
|
Re: Hacker GP15 Cavalier
Hang on. Some voltages have appeared in the first post. Now one of the triode cathodes (V1b) looks right - not 11V but 1.27V. But the other one, V2b, is still cut off (4.8V, which is close to 5V). Is there something intermittent ?
Now there is an anomaly between the V1b anode voltage and cathode voltage. Cheers, GJ
__________________
http://www.ampregen.com |
12th Apr 2018, 2:02 pm | #6 |
Moderator
Join Date: Feb 2006
Location: Wembley, Middlesex
Posts: 7,225
|
Re: Hacker GP15 Cavalier
Sorry GJ, I've amended post 1, because I wrote it before putting in the voltage values.
No I loosened and tightened the screws on the tag strips and for a time, the voltage on V1's anode dropped to 145V, so I used some wire to strap all the ground tags together and to the -ve of the smoothing capacitor, but afterwards V1a's anode rose back to 260V With the two ECL86's pulled the HT on all 3 capacitors is around 360V |
12th Apr 2018, 2:24 pm | #7 |
Dekatron
Join Date: Sep 2007
Location: Oxfordshire, UK.
Posts: 4,311
|
Re: Hacker GP15 Cavalier
It sounds like there is something intermittent somewhere (the most painful kind of fault to track down, as I'm sure you know). If measurements don't show it up then I find I have to break the circuit down into blocks and test them individually e.g. use a separate HT supply just to power the triodes, maybe one at a time, and see if they work fine when fed with a dedicated supply. If so then there is something wrong with the normal supply, or the feeds from it to the triodes. It can be a slow old process I'm afraid.
If it's any consolation I have been trying for the last 5 months (on and off, admittedly) to find the source of some hiss in a valve-regulated HT supply for an audio pre-amp. I finally discovered yesterday that the hiss is not coming from the HT supply at all but from a solid-state 12V regulated supply which feeds the heaters in the pre-amp and shares a common ground wire along the umbilical with the HT. The common ground was coupling the LT hiss into the audio. 10 minutes fitting a small RC filter and all was then quiet. The feeling of relief and satisfaction was almost worth the wait ! 360V on a 275V cap ... there's brave . Cheers, GJ
__________________
http://www.ampregen.com |
12th Apr 2018, 2:42 pm | #8 |
Moderator
Join Date: Feb 2006
Location: Wembley, Middlesex
Posts: 7,225
|
Re: Hacker GP15 Cavalier
The caps are now rated at 500V, C8 is rated at 450V. I'm confident that the supply is OK.
The circuit can be broken into stages, I was wondering if the valve holders themselves have failed but if that was the case then wriggling the valves should make the signal come and go, but it doesn't. I'm working on the first stage which is a simple voltage amplifier. |
12th Apr 2018, 5:29 pm | #9 |
Triode
Join Date: Dec 2011
Location: Perth, Scotland, UK.
Posts: 24
|
Re: Hacker GP15 Cavalier
I had either a GP42 or GP15 that howled a couple of minutes after the valves starting conducting.
It was caused by one of the internal signal cables being microphonic. By a process of elimination I traced it to the cable, I didn't know cables were capable of being microphonic to that extent but replacing the cable fixed it. |
13th Apr 2018, 7:27 am | #10 |
Moderator
Join Date: Feb 2006
Location: Wembley, Middlesex
Posts: 7,225
|
Re: Hacker GP15 Cavalier
It's not a wiring problem, unit is being tested out of the record player, connected to an oscillator.
|
18th Apr 2018, 10:54 pm | #11 |
Moderator
Join Date: Feb 2006
Location: Wembley, Middlesex
Posts: 7,225
|
Re: Hacker GP15 Cavalier
I decided to replace the valve holders but the result surprisingly is the same.
It looks like the fault is with both triode sections. There is virtually no virtually any voltage drop across the 220K anode load resistors yet the triode's cathodes have about 5V across them. So if the valve isn't conducting, where is the cathode voltage coming from. Both triodes are behaving in the same way. |
19th Apr 2018, 9:30 am | #12 | |
Dekatron
Join Date: Sep 2007
Location: Oxfordshire, UK.
Posts: 4,311
|
Re: Hacker GP15 Cavalier
Quote:
Cheers, GJ
__________________
http://www.ampregen.com |
|
19th Apr 2018, 9:36 am | #13 |
Dekatron
Join Date: Apr 2011
Location: Cornwall, UK.
Posts: 13,454
|
Re: Hacker GP15 Cavalier
Try measuring the voltage both sides of the cathode resistors with respect to HT winding center tap to eliminate the possibility of a -ve rail break.
Lawrence. |
19th Apr 2018, 12:07 pm | #14 | |
Moderator
Join Date: Feb 2006
Location: Wembley, Middlesex
Posts: 7,225
|
Quote:
Again no change |
|
19th Apr 2018, 12:15 pm | #15 |
Dekatron
Join Date: Sep 2007
Location: Oxfordshire, UK.
Posts: 4,311
|
Re: Hacker GP15 Cavalier
At this stage I'd croc-clip the triode grid hard to the bottom wire of the cathode resistor/capacitor and then use another croc lead to the negative terminal on the main smoothing cap. That way I'd be 101% certain of the grounding. If that didn't resolve things then I'd copy just the triode circuit using a new valveholder, resistors and capacitor and repeat the test on the suspect ECL86 in that setup. In the end a large enough scientific hammer will crack the nut.
EDIT: I'd also try and use another voltmeter, just in case Cheers, GJ
__________________
http://www.ampregen.com |
19th Apr 2018, 12:22 pm | #16 |
Moderator
Join Date: Feb 2006
Location: Wembley, Middlesex
Posts: 7,225
|
Re: Hacker GP15 Cavalier
I have done that now but no change.
Valveholders and valves are new |
23rd Apr 2018, 3:28 pm | #17 |
Moderator
Join Date: Feb 2006
Location: Wembley, Middlesex
Posts: 7,225
|
Re: Hacker GP15 Cavalier
Well I'm stuck. i have even changed the anode load resistors just in case they were breaking down, but again no change. I have tried both new and old valves, I have replaced just about every component in the amplifier, including the valve holders.
I cannot offer any explanation as to what is going on in the triode circuits of both valves. It is seriously doing my head in. HELP |
23rd Apr 2018, 8:15 pm | #18 |
Pentode
Join Date: Nov 2015
Location: Oxfordshire, UK.
Posts: 171
|
Re: Hacker GP15 Cavalier
Hi Michael
Might be worth a look at the two fuseholders/Fuses I have known them to corrode and do very strange things to the HT supply. A quick clean might be worth a try. Regards RT |
24th Apr 2018, 8:29 am | #19 |
Dekatron
Join Date: Jan 2005
Location: Oxford, UK.
Posts: 17,866
|
Re: Hacker GP15 Cavalier
What a good idea. The fuses themselves are also not beyond suspicion.
|
24th Apr 2018, 3:14 pm | #20 |
Moderator
Join Date: Feb 2006
Location: Wembley, Middlesex
Posts: 7,225
|
Re: Hacker GP15 Cavalier
Cleaned up fuseholders replaced fuses, no change.
HT is rock steady. On both triode sections of the valve there appears to be 2.2mA flowing through the cathode resistor but only 0.022 mA flowing through the anode resistor. Grids are at 0V. I've also checked to see if there is any spurious oscillation, none found. HT is rock steady with no ripple or very little ripple on it. I should have perhaps said from the outset that when I took the player in, it would make a loud howling noise just after it was switched on which appeared to come and go if you knocked the valves, but as I said both valves and valveholders have been replaced |