|
Vintage Test Gear and Workshop Equipment For discussions about vintage test gear and workshop equipment such as coil winders. |
|
Thread Tools |
23rd Jun 2020, 12:17 pm | #1 |
Nonode
Join Date: Mar 2013
Location: Leicester, UK
Posts: 2,535
|
Advance type 63A
I have just received this signal generator from a member and although I have another project on the go at the moment, couldn't resist having a quick peak inside. I say quick, but it took a while to remove all of the screws holding two screening covers over the all important bits of electronics. I didn't cound them, but there must have been at least 30-40 of them!
I was informed that there are a number of Hunts caps to be replaced and that appears to be the case. What I am a little puzzled about is that with exception of the electrolytics and one "metalized paper" cap, everything else is specified as "mica". The Hunts caps look like typical paper Mouldseals. Did Hunts manufacture Mica caps in a Mouldseal style package, or are these likely to be paper? Most do not appear to be in critical positions although a couple of them do seem to form part of filtering RC and LC filtering arrangements, although not part of the actual oscillator circuit. Should these be replaced with Mica caps as specified, or can Metalised polester foil types be used instead? Values typically are mostly 0.001, 0.005 and 0.02 with a single odd 0.04. |
24th Jun 2020, 10:49 am | #2 |
Nonode
Join Date: Mar 2013
Location: Leicester, UK
Posts: 2,535
|
Re: Advance type 63A
Bump. Ok, perhaps I have asked a silly question or perhaps some photos will help? Mica caps are usually very distinct from paper and these look like Mouldseals which are usually paper caps, although they are quite small, about 3/4in long and 3/8 in diameter. The black one in the second photo is even smaller.
Paper caps will usually require replacing so if these are Mouldseals then I will be replacing them progressively. However, I would still be interested to know in that case why paper caps have been used when the parts list specifies mica? Also, in the third photo, what are the three round white'ish components? Are these some form of ceramic disc capacitor? |
24th Jun 2020, 12:45 pm | #3 |
Octode
Join Date: Jul 2017
Location: Newcastle Upon Tyne, Tyne & Wear, UK.
Posts: 1,186
|
Re: Advance type 63A
I think they are conventional Hunts 'Mouldseals'. I have seen those values, although uncommon, in valve radios of the era.
Replacements can be found online. Those yellow axial types. The unknown component looks like a thermistor, I could be wrong though. Have you seen the circuit here? I am interested in obtaining a similar FM sweeper/generator so I am curious to see how you get on! |
24th Jun 2020, 1:25 pm | #4 |
Dekatron
Join Date: Apr 2011
Location: Cornwall, UK.
Posts: 13,454
|
Re: Advance type 63A
The light coloured discs in the 3rd photo will be ceramic capacitors.
Lawrence. |
24th Jun 2020, 3:14 pm | #5 | ||
Nonode
Join Date: Mar 2013
Location: Leicester, UK
Posts: 2,535
|
Re: Advance type 63A
Quote:
Quote:
The valve present is a 6BQ7A which again varies from the specified 12AT7 (ECC81). The substitution was mentioned in the linked thread in connection with the later 6 range version, but this generator has 5 ranges although looking at the datasheet the valves are very similar. |
||
26th Jun 2020, 12:53 pm | #6 |
Nonode
Join Date: Mar 2013
Location: Leicester, UK
Posts: 2,535
|
Re: Advance type 63A
While I am waiting for parts to turn up for the radio I am working on, I have investigated the sig gen a bit further. There is good news and there is bad news.
The electrolytic capacitors in the power supply circuit had already been replaced and I replaced the remaining handful of paper caps in the power supply filtering chain with quality high voltage ceramics. The generator was then powered up gradually on and isolating transformer and variac with no problems. At this stage I haven't touched any components in the signal generating circuit nor any of the input/output decoupling caps. The good news is that the generator circuit itself seems to be working. The valve warms up and I am getting output up to and out of the Output Voltage control. I can see modulated AM in the "30% AM" position and a clean sine wave on the CW and Crystal Check positions. Observing on an oscilloscope, the sine wave frequency can be adjusted. Using the built in counter it was possible to determine that the output frequency is about spot on across all frequency bands and a signal is present all the way up to the top of the band 230-100MHz band, although at varying amplitude. So far, so good. Now for the bad news. I was advised that the sig gen had no output and indeed there was no output at the R.F Output terminal, except, as it turns out, when the control is turned to the X1 position in the μV range. There was no output in any other position so It would seem that there is some problem with the resistor chain in the attenuator circuit which will need to be investigated. There is also a mechanical problem with the frequency range control. One of the rivets pops out every now and then from under the spring plate as the control is rotated, with the couplings distorting as it does so. I am concerned that it may give way at some point if not repaired, but unfortunately getting access to this will require a complete dis-assembly. Looking carefully at the unit, I believe that the oscillator can be removed in one complete piece by disconnecting the 3 wires to the coils fed from the power supply, the co-ax from the Output Voltage pot, disconnecting the linkages to the Frequency, Frequency Range and Modulation controls and removing 4 large screws from the rear plate. Disconnecting the tuner shaft looks the trickiest as it uses some rather small grub screws. Hopefully the oscillator circuit will not need to be disturbed in any critical way. If anyone has any notes on how to disassemble the unit, or has done so in the past and provide any insights this would be appreciated. |
27th Jun 2020, 6:47 am | #7 |
Dekatron
Join Date: Jul 2009
Location: Sleaford, Lincs. UK.
Posts: 7,670
|
Re: Advance type 63A
I'd advise giving the grubscrews a squirt of penetrating oil or diesel overnight at least, nothing worse than recalcitrant grubscews.
I've worked on a few Advance sig gens, the resistors in the OP potential divider are often off spec, the highest values at least, same as the rest of the sig gen. I replaced the pot with a 10 turn on mine and tweaked the R value's to get an accurate OP in the ranges I was mostly using. Andy.
__________________
Curiosity hasn't killed this cat...so far. |
27th Jun 2020, 4:23 pm | #8 |
Nonode
Join Date: Mar 2013
Location: Leicester, UK
Posts: 2,535
|
Re: Advance type 63A
I made a bit more progress with this today. I started work on removing the oscillator sub-assembly from the main assembly but ran into a couple of snags. I was able to slacken off the screws on two of the controls easily enough, but despite loosening the grub screw on the variable capacitor shaft it wouldn't slide out of the coupling. PlusGas has been applied and allowed to soak in overnight.
In the meantime, the complete oscillator assembly has been removed from the case. The second snag was the coupling between the two assembly plates and the Output pot chamber. The co-ax was de-soldered and the outer nut securing the potentiometer box to the oscillator assemebly was removed which, after removing the necessary knobs and four large bolts from the front of the case, allowed the complete oscillator assembly to be withdrawn. I am hoping that once the PlusGas does its job, I can separate the two plates and get at the damaged rivet joint. In the meantime I checked out the attenuator. It seems that all resistors present are within the 5% tolerance specified, however there are differences between the hardware and the circuit diagram. The potentiometer measures around 87Ω but should be 100Ω although that is still within 20%. The resistor across it (R9) measures 54.7Ω but also has an inductor in series which is not shown on the circuit diagram. The component list shows R9 as 100Ω. The series resistor between pot and attenuator also reads 54.7Ω but the component list says 120Ω. Within the the attenuator case itself, R12 is missing altogether. Since there is no screw thread to install a screw to tie one end down to the attenuator body (earth) as is the arrangement with the remaining resistors commoned to the case, it would appear that there simply wasn't one installed by the manufacturer. Perhaps these are differences between Type 63 and Type 63A? The switching side of the attenuator seems clean but I am not sure whether there is good contact at all switch positions. The detent mechanism could probably use a clan and a a little Teflon grease and I will clean the switch contacts before re-assembly. There is a dent in the front panel that seems to have been done by an round object of a small diameter, perhaps a metal spindle from something. Whatever is was, the impact must have been pretty severe! The case is at least 1mm thick and an attempt to hammer it out does not seem to have moved it a jot! I have used the flat head of one hammer placed firmly on the dent, and was hitting it quite hard with a sizeable rubber mallet. I am wondering whether this is best left alone? At any rate, progress is being made little by little. |
28th Jun 2020, 5:57 am | #9 |
Dekatron
Join Date: Jul 2009
Location: Sleaford, Lincs. UK.
Posts: 7,670
|
Re: Advance type 63A
The dent might be best drilled out and a bolt and washer used to cover it up, i've found this less of a faff than trying to fill dinks and repaint..
Andy.
__________________
Curiosity hasn't killed this cat...so far. |
28th Jun 2020, 3:00 pm | #10 |
Nonode
Join Date: Mar 2013
Location: Leicester, UK
Posts: 2,535
|
Re: Advance type 63A
Andy, I hadn't considered that, but an additional screw head would probably not look out of place. The total area of the "crater" is probably around 3/4in with the centre dent being maybe 3mm in diameter. I guess if the centre is drilled out it might just make it easier to flatten out.
The good news is that I managed to separate the inner plate from the outer one of the oscillator section. There was an additional grub screw that could now be seen from the top but I was just about able to access from the side once the frequency control had been rotated 180deg and the other controls had been slackened of. This allowed me access to the damaged coupling. Attempting to remove the original rivet might have caused more damage so I first attempted to flatten out the disconnected side. I don't have a rivet tool, but many of the bits of my precision screwdriver set have a narrow shank and widen out in a steeply curved fashion to the width of the bit about halfway up. I was able to use the "shoulder" of one of these bits to gently hammer down and splay out the metal over the plate to make it grip and hold tight in place. However, being concerned that this may work loose again over time as the metal had been stressed, I took a belt and braces approach and added a small recessed screw from my box of bits through the centre of the rivet to make sure that the top would remain splayed out and not disengage. The rivet on the other side looks sound so I have left it as it is. |
29th Jun 2020, 8:58 pm | #11 |
Nonode
Join Date: Mar 2013
Location: Leicester, UK
Posts: 2,535
|
Re: Advance type 63A
The oscillator assembly has been put back together and the couplings and wiring re-connected. The attenuator contacts have been cleaned and checked for good contact. The next step is to see whether I can do something about that dent or whether to leave well alone. Might have a look at that tomorrow.
|
30th Jun 2020, 12:49 pm | #12 |
Octode
Join Date: Dec 2010
Location: St Ives, Cambridgeshire, UK.
Posts: 1,180
|
Re: Advance type 63A
That must have been quite some force to do that knowing how thick the Advance front covers are! I'm not sure you will be able to push it out as, from the photo, it looks like the metal has stretched, like it does with a power press tool. Shame. Andy's method above maybe the most sensible. Just thinking that something like a fly press might get the worst of it out, enough to grind the indentation out and then finish it off to suit.
Thanks for writing your tribulations up, very useful. Andrew
__________________
Invisible airwaves crackle with life. Or they should do. BVWS Member |
30th Jun 2020, 10:00 pm | #13 |
Nonode
Join Date: Mar 2013
Location: Leicester, UK
Posts: 2,535
|
Re: Advance type 63A
I took another look at the dent problem today and think I managed to get the worst of it out with the help of a G clamp. The perspex window had to be removed - which needed doing for cleaning anyway - so that the G clamp could reach and be positioned over the dent. First, the flaking paint was sealed with a bit of superglue to prevent more of it from breaking off. The clamp was then positioned through the dial window. A piece of flat steel was placed over the front to provide a solid flat surface to press against and the flat pad on one end of the G clamp was placed over the bump on the inside surface. The clamp was then gradually and carefully closed by rotating the screw handle. This has pressed out most of the crater and the outer surface is now quite flat except for that small dint. I was also able to straighten the metal collar around the hole for the frequency control spindle so that it sits flush with the case.
I am uncertain whether to press further with the G clamp to see whether that dint can be totally pressed out, or whether to quit while I am ahead? I am somewhat concerned that if the metal has stretched as seems likely, then pressing further might contort the panel in some unexpected manner. Unfortunately I don't have access to a fly press so I am now considering Andy's suggestion as it would probably look better to have a random screw head on the panel than a dint. Tomorrow I will have a look what I have in my box of bits that may be suitable and whether there is sufficient clearance on the inside to accommodate something. I suspect that once a suitably sized knob is fitted, that dent will probably be mostly covered anyway, so it might be best left as is. I will give it some further thought. Once that is dealt with then the oscillator chassis can be re-installed and everything re-connected. |
2nd Jul 2020, 7:22 pm | #14 |
Nonode
Join Date: Mar 2013
Location: Leicester, UK
Posts: 2,535
|
Re: Advance type 63A
Ok, I've gone with Andy's suggestion of fitting a screw so the dent is sorted. I fitted everything back together and tested to make sure the oscillator is working OK and it seems to be. Tomorrow I will reconnect and test the attenuator.
In the meantime I have a couple of questions: 1. What type of RF connector is that for the output and what type of plug do I need to buy to attach to it? 2. How do I test the FM side of things? Need a bit of guidance here. When set to 22.5kc/s, the output is just a plain sine wave. On 75kc/s the output does look modulated and on the Mains FM setting the Mains F.M. Deviation control does vary the level of modulation. I think therefore that the 75kc/s and Mains F.M modes are working although I'm not sure whether the output wave I am seeing is correct. Tomorrow I might try and see whether I can pick anything up on the scanner. I had a look at the manual that I have but it does not mention the X-Sweep Phase control and the X-Sweep Out. Does anyone know what these features are for? The is a high voltage sine wave on that output at present. |
2nd Jul 2020, 9:15 pm | #15 |
Octode
Join Date: Jul 2017
Location: Newcastle Upon Tyne, Tyne & Wear, UK.
Posts: 1,186
|
Re: Advance type 63A
The connector looks like a female Belling-Lee connector.
Could you please post the manual you are using? And some photographs of the front of the unit? I assume that the deviation controls how much the generator sweeps +/- the selected frequency? Did you connect your 'scope to the X out socket? |
2nd Jul 2020, 11:47 pm | #16 |
Dekatron
Join Date: Nov 2003
Location: Heckmondwike, West Yorkshire, UK.
Posts: 9,642
|
Re: Advance type 63A
The connector is just known as the Pye plug: https://www.vintage-radio.net/forum/...7&postcount=13
|
3rd Jul 2020, 9:43 am | #17 | |||
Nonode
Join Date: Mar 2013
Location: Leicester, UK
Posts: 2,535
|
Re: Advance type 63A
Quote:
Quote:
I checked the output voltage and had a look at the waveform starting at the 50v/div setting (with 10x probe) on the scope. The output voltage ranges approximately 100-250vpp depending on the position of the X-Sweep control. I didn't check the frequency but don't want to connect it to anything at the moment until I understand whether the voltage level is as should be and how this feature should be used. Quote:
https://www.vintage-radio.net/forum/...25&postcount=1 Is the one in the photo of the type 63 the male variant versus the female on the type 63A? I checked for example Cricklewood Electronics as well as eBay but couldn't find the male variant. Given how the attenuator module is constructed, fitting a modern N connector might be a difficult proposition. It is safe to insert a Belling Lee plug into this, or, given that it protrudes by 3mm or so, is this likely to cause damage to the centre conductor socket? One other difference with this unit that it does not have a socket labelled "Crystal Check" like the type 63 or "Cal. Output" like I have seen on another type 63A. Instead, the socket on this unit is labelled "Phones". The actual contact is so deep within the unit (about 2.5in) I'm not sure quite what fitted into there. Would the service engineer have had a bespoke probe tool for that? Last edited by WaveyDipole; 3rd Jul 2020 at 9:59 am. |
|||
3rd Jul 2020, 11:06 am | #18 |
Octode
Join Date: Jul 2017
Location: Newcastle Upon Tyne, Tyne & Wear, UK.
Posts: 1,186
|
Re: Advance type 63A
It's interesting how there is a 'XTAL CHECK' position but no obvious input for a crystal. Is there a crystal inside the generator?
I think 'Phones' is for a 1/4" headset. Some generators had a feature where you could listen to the beating between a crystal frequency standard and the RF output of the generator as a way to check calibration. The Taylor 62A MKII is a similar instrument of the period. It was probably a competitor with the Advance 63A you have. https://www.google.com/url?sa=t&rct=...kcyqjGCSXpsrac It might be worth having a look at how the Taylor works as the operating principles could be applied to your Advance. I'm not sure about the mains setting. Is it just an unswept FM wave? With a 50Hz tone? Food for thought. |
3rd Jul 2020, 12:17 pm | #19 |
Nonode
Join Date: Mar 2013
Location: Leicester, UK
Posts: 2,535
|
Re: Advance type 63A
Yes, there is a Crystal inside the generator. Reading the notes in the manual regarding this on the type 63, stets that it is a 5Mc/s crystal and mentions that the cursor is adjustable to the crystal beat, but, the type 63A does not have the adjustment screw which is present on the type 63.
It also states that "the output from the calibrator is available via a socket on the front panel into which may be plugged the probe unit 10670 to connect the output to headphones or to an audio amplifier and speaker.". This answers one of my questions: a special probe is required to reach the contact deep down inside the PHONES socket (as it is labelled on this unit) on this generator. I imagine this is all about shielding. According to the Crystal Check paragraph under Circuit Description, the output is a "beat note", so this would correspond to your note about listening to the beating between a crystal and the generated frequency as a way to check calibration. The manual does stated that the purpose is to "provide[s] accurate checkpoints" so does this provide an audible beat frequency with a null or something which would indicate a precise harmonic of 5Mc/s so one could check that dial marker corresponds to a 5Mc/s multiple? Or is 5mc/s the only checkpoint? I have reconnected the attenuator and the good news is that it seems to be working perfectly fine although reading a signal at x10 and x1 μV is difficult on a scope even with a x1 probe at the 1mV setting, but there is an output. I had a listen with the scanner and you are correct. The Mains reference seems to refer to the FM carrier being modulated with a 50Hz mains signal. On the 75% setting and AM I get a 1kHz tone as indicated. I am considering the possibility of doing the same. Did you do a straight with a 100ohm 10 turn pot or did you choose a different value? Last edited by WaveyDipole; 3rd Jul 2020 at 12:47 pm. |
3rd Jul 2020, 2:32 pm | #20 |
Octode
Join Date: Jul 2017
Location: Newcastle Upon Tyne, Tyne & Wear, UK.
Posts: 1,186
|
Re: Advance type 63A
Thanks for the manual, WD!
I have had a look through the manual for the 63 and have a suspicion that the 63 and the 63A could be more similar than we think. I got the impression from the manual that the crystal can be used across the bands in 5mc/s steps. Is it worth probing the output of the phones socket with your 'scope? Or would you need to strip down the generator again to reach it? |