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 > Specific Vintage Equipment > Vintage Amateur and Military Radio

Notices

Vintage Amateur and Military Radio Amateur/military receivers and transmitters, morse, and any other related vintage comms equipment.

Closed Thread
 
Thread Tools
Old 6th Dec 2015, 11:43 pm   #61
Radio Wrangler
Moderator
 
Radio Wrangler's Avatar
 
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,801
Default Re: Eddystone 888A.

There were articles all over the place about substituting FETs in place of thermionics. One ploy was to use a cascade structure with a high voltage device on top (FET or bipolar) this was manufactured for a while as the 'Fetron'

I read about them in 'Ham Radio' (American) magazine and in Pat Hawker's column in Radcom as well as in WW

Another ploy was to use FETs in all places, and to reduce the HT voltage to something better suited to them.

Unfortunately these ploys foundered on the issue of dynamic range. They just got known about around the same era that interned problems in a variety of classic receivers came to light and few people wanted to make things worse.

David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done

Last edited by Radio Wrangler; 6th Dec 2015 at 11:53 pm.
Radio Wrangler is offline  
Old 7th Dec 2015, 12:43 am   #62
Wendymott
Octode
 
Wendymott's Avatar
 
Join Date: Jun 2013
Location: Huddersfield, West Yorkshire, UK.
Posts: 1,782
Default Re: Eddystone 888A.

Hi Tanuki and David. Many thanks for your trouble....I will make a suitable mini pcb and mount on a 7 pin valve plug...... David,, its only the VFO and BFO I am interested in "Fetising"...... If successful I may well attend to the 750 as well.. AS it happens the 888 has now its case on.... but that was just to tidy up the work bench... tomorrow, as they say... is another day.
__________________
Should get out more.

Regards
Wendy G8BZY
Wendymott is offline  
Old 7th Dec 2015, 6:27 am   #63
Bazz4CQJ
Dekatron
 
Bazz4CQJ's Avatar
 
Join Date: Apr 2009
Location: Oxfordshire, UK.
Posts: 4,924
Default Re: Eddystone 888A.

Mmmm.... I wouldn't be averse to Fetising my HRO (which is in the middle of a "mid-life update", which has been going on for about 4 years now). However, if you put a FET in to the LO aren't you risking problems restoring the tracking? My BFO has long since gone and definitely will be fetised and product detectorised too...eventually.

B
Bazz4CQJ is offline  
Old 7th Dec 2015, 7:53 am   #64
Radio Wrangler
Moderator
 
Radio Wrangler's Avatar
 
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,801
Default Re: Eddystone 888A.

You may have problems getting as much output from FETised LO and BFO circuits, as well as LO/RF tracking.

In the HRO, the LO runs at rather a high level and qualifies as a QRP transmitter in its own right.

Some of the FET substitution articles used unusual high voltage high power J-FETs, that never caught on and are essentially unobtainable. Most used period low power parts like 2N3819, 2N5245 and 2N4416 with greatly reduced HT. The plug-in-the-valve-socket ones used a low power J-FET cascaded by a high voltage bipolar.

This is one area which soon fell out of favour, simply because it didn't work very well at all. Everyone at the time decided to stick with valves.

Some hybrid receiver designs did use transistor VFOs for low heat, low drift reasons, with buffer stages into the (valved) mixer. The G3PDM rx design (RSGB handbooks late 1960s into 1970s) is a good example. It uses a very stable Vackar circuit and the article is excellent on explaining how to make a stable VFO.

David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done
Radio Wrangler is offline  
Old 7th Dec 2015, 11:12 am   #65
G6Tanuki
Dekatron
 
G6Tanuki's Avatar
 
Join Date: Apr 2012
Location: Wiltshire, UK.
Posts: 13,953
Default Re: Eddystone 888A.

Yes the G3PDM FET Vackar circuit is really rather good: I used a slightly-different-frequency version of it (with the tuning range reduced quite a bit, followed by a frequency-multiplier chain) to convert a Pye Westminster to continuous tuning on 2 metres! The key to getting stability was to mechanically restrain all the components: from memory I even used a blob of Evo-Stik to glue the FET to a metal cross-brace in the diecast box.
G6Tanuki is offline  
Old 7th Dec 2015, 12:42 pm   #66
Wendymott
Octode
 
Wendymott's Avatar
 
Join Date: Jun 2013
Location: Huddersfield, West Yorkshire, UK.
Posts: 1,782
Default Re: Eddystone 888A.

Hi peeps.... Early days yet...... I take all your comments with interest and see what happens. The 888 uses the triode grid of the ECH42 to inject the LO so I am hoping the level required is not too high, I will measure this and see if it is feasible. As for tracking, with the 888 being bandspread, I think this may not be a problem... but as I say.... I have to "Suck it and see"..... I will report back in the next few days....
__________________
Should get out more.

Regards
Wendy G8BZY
Wendymott is offline  
Old 7th Dec 2015, 3:29 pm   #67
Skywave
Rest in Peace
 
Join Date: Jun 2006
Location: Chard, South Somerset, UK.
Posts: 7,457
Arrow Re: Eddystone 888A.

On the topic of modifications to an 888 / 888A, in mine, the 6BA6 R.F. amp. has been replaced by an ECC85 in a cascode arrangement. (That mod. was made by the previous owner; I have chosen to retain it). The idea is that the cascode cct. gives a better over-load characteristic and lower noise. (If I recall correctly). That mod. has never given any problems; it was fitted prior to 1968. Recently, I had another, unmodified, 888A here and did a few qualitative comparison tests: I did not find any 'negative' results with that mod.

Al.
Skywave is offline  
Old 7th Dec 2015, 5:24 pm   #68
Bazz4CQJ
Dekatron
 
Bazz4CQJ's Avatar
 
Join Date: Apr 2009
Location: Oxfordshire, UK.
Posts: 4,924
Default Re: Eddystone 888A.

Quote:
Originally Posted by Skywave View Post
On the topic of modifications to an 888 / 888A, in mine, the 6BA6 R.F. amp. has been replaced by an ECC85 in a cascode arrangement. Al.
A similar ECC85 cascode circuit appeared in SWM for the HRO replacing the original 6D6 1st RF, which claimed to be an improvement. Interestingly, it does not make use of the AGC line. Just a few days ago, I found an article in a 1959 copy of PW with a mod for the HRO 1st RF which kept the original 6D6 but disconnected the AGC from it, replacing it with a toggle switch for Hi/Lo signals. In an era dominated by SSB for amateur use, I'm not quite sure about what changes to a 1st RF stage are most appropriate?

There have so many mods published for vintage receivers including various Eddystones, HRO, AR88, RA17 etc, etc. T'is a pity that some kind soul(s) has not attempted to collate them in to a single entity.

B

Last edited by Bazz4CQJ; 7th Dec 2015 at 5:35 pm.
Bazz4CQJ is offline  
Old 7th Dec 2015, 6:51 pm   #69
G3PIJpeter
Hexode
 
Join Date: Nov 2011
Location: Chippenham, Wiltshire, UK.
Posts: 319
Default Re: Eddystone 888A.

Quote:
Originally Posted by Bazz4CQJ View Post
A similar ECC85 cascode circuit appeared in SWM for the HRO replacing the original 6D6 1st RF B
The 1960s SWM circuit I built utilised a plug-in adapter arrangement for substituting the ECC85 into the 6D6 1st RF amplifier socket. It certainly gave consistent gain up through 21 and 28 Mc/s, whereas the original valve tailed off a bit at higher frequencies.
- Peter
Attached Thumbnails
Click image for larger version

Name:	06FEB2006_072A.jpg
Views:	110
Size:	27.7 KB
ID:	116565  
G3PIJpeter is offline  
Old 7th Dec 2015, 7:06 pm   #70
Skywave
Rest in Peace
 
Join Date: Jun 2006
Location: Chard, South Somerset, UK.
Posts: 7,457
Arrow Re: Eddystone 888A.

With regard to the AGC in a typical multi-valve superhet, it's not uncommon to find that the last I.F. stage receives the greatest amount of AGC voltage; previous I.F. stages receive progressively less; the R.F amplifier receiving the least of all. (The mixer usually is not controlled from the AGC line. I believe that that is to preserve the designed non-linearity in the mixer). The general idea of that AGC distribution is simply to maximize the S/N ratio. In the event of a medium strength signal, it's the gain in the last I.F. amp. that is reduced first; on substantially strong signals, the gain of the R.F. amp. gets turned down as well. All that auto gain control (if switched in) is often supplemented with the manual 'R.F. gain' control * - which sometimes operates on the R.F. and I.F. stages as well (or some of the I.F. stages.) Sometimes separate R.F. and I.F. gain controls are used, giving maximum flexibility in manual control - the 888A being a good (and an uncommon) example of that technique. That arrangement is particularly suitable for the reception of SSB: the R.F. gain control can be adjusted for optimum S/N, or handling really strong signals; the I.F. gain control can then be used for optimum signal level for SSB demodulation via the product detector; the A.F. gain setting is then simply a question of how loud you want the audio to be!

*It's interesting to note that in early versions of the Racal RA-17, when the AGC was switched in, the I.F. gain control was disabled. (There is no 'R.F. gain' control as such - instead, a resistive switched attenuator is featured). Subsequently, a mod was issued enabling that I.F. gain control to be functional, if the AGC was switched in - or not. An increase in 'user operability' was thus established.

Al.
Skywave is offline  
Old 7th Dec 2015, 7:31 pm   #71
G6Tanuki
Dekatron
 
G6Tanuki's Avatar
 
Join Date: Apr 2012
Location: Wiltshire, UK.
Posts: 13,953
Default Re: Eddystone 888A.

IF AGC and where it's applied is subjefct to much debate: many receivers don't apply AGC to the final IF stage - so it acts as a 'virtual AGC amplifier' and - allegedly - gives better overall AGC action.

The usual justifications for not AGCing the mixer in a 'communications' receiver are:

[1] it minimises 'pulling' of the local-oscillator due to changes in the capacitance the mixer presents to the LO.

[2] It means you can keep the mixer biased correctly for best S/N ratio and/or least susceptibility to crossmodulation.

In a dual-conversion receiver it's often the case that the first mixer is non-AGC-controlled but the second mixer does get to see (some degree of) the AGC voltage: 'pulling' of a LO is less of an issue at lower frequencies like converting a 1.6MHz first-IF to 85 or 100KHz for the second IF.

Many 'professional' receivers (usually dual-conversion) have two AGC loops - one deriving its input from ahead of the main selectivity-elements of the radio, the other working like a normal AGC setup. The idea is that the AGC derived from before the main selectivity-elements can 'see' and respond to strong signals that are close-to the signal you're receiving but not close-enough to actually come through to the traditional AGC detector after the selectivity. A very strong signal, say, 30KHz off what you're trying to receive (which would cause crossmodulation in the first mixer) is 'seen' by the outer AGC loop and so throttles-back the first RF amp to stop the mixer being nailed. The reduced sensitivity is made-up for by the main AGC loop turning up the gain of the main IF strip to compensate.

As can be imagined, getting the time-constants of the various stages right is somewhat of a compromise - but the idea was well-enough appreciated that the likes of the Hallicrafters SX115/SX117 used it.

Other receivers I've seen actually have the first RF-amp valve designed to have a relatively low maximum gain (the RF amp stage actually getting a lot of its signal-gain by way of the transformer-action of the tuned circuits and the valve doing load/impedance matching) - the valve's maximum gain may be only 5, and actually negative when faced with strong signals when the AGC turns things down. All in the interests of minimising crossmodulation [the Navy have some odd requirements - a ship has serious practical limits on how much physical separation's possible between the various transmitter and receiver antennas]

Last edited by G6Tanuki; 7th Dec 2015 at 7:55 pm.
G6Tanuki is offline  
Old 7th Dec 2015, 9:23 pm   #72
Wendymott
Octode
 
Wendymott's Avatar
 
Join Date: Jun 2013
Location: Huddersfield, West Yorkshire, UK.
Posts: 1,782
Default Re: Eddystone 888A.

The technical aspects of the thread are going way beyond what I envisaged when I first posted the item, but MOST welcome. However... I need to stretch the grey matter again please...
The 15 Metre range has been "deaf" from getting the receiver. I have done quite a bit of work on it and still not reached a conclusion.
The tracking of the mixer tuning seems to be poor. If I set the tuning at 21 Mhz, and peak using "L", then tune to 21.5 Mhz, peaking "C" should be the end of the matter. However... returning to 21 Mhz the sensitivity is way down, thus retuning "L" "Core inwards", I can then repeak, but then returning to 21.5, I have to reduce "C" to peak again. Ad infinitum, until there is no more core adjust.
I removed the coil and trimmer to find additional "C's" not shown on the schematic... one @ 40pf +/-1%, across the trimmer cap "Beehive", and another 35pf +/- 1% from V Cap to deck.
I replaced both of these with "near enough" caps to get an idea what is going on...and even added wires to enable additional caps "outboard", to try and get an idea whether caps are needed to be added or not. I measured the removed caps, hoping one may be faulty, but no both read spot on.
All other bands are tuning correctly.
By the way.... trying to replace a 6AM6 with a ECC85 in the 888 is NOT an option...you cannot get to the base, its virtually impossible without removing the wave change switch, and I am not going there.
__________________
Should get out more.

Regards
Wendy G8BZY
Wendymott is offline  
Old 8th Dec 2015, 12:11 am   #73
Skywave
Rest in Peace
 
Join Date: Jun 2006
Location: Chard, South Somerset, UK.
Posts: 7,457
Arrow Re: Eddystone 888A.

Thanks G6T for your follow-on to my previous post, especially your reminder of why AGC is often not applied to the mixer. Most of the other points you raise I am familiar with. I deliberately kept my remarks as brief as I could on this topic (of which so much has been written and researched) to try to keep my post appropriate to the Eddy. 888A - and similar receivers.

Al.
Skywave is offline  
Old 8th Dec 2015, 12:51 am   #74
Skywave
Rest in Peace
 
Join Date: Jun 2006
Location: Chard, South Somerset, UK.
Posts: 7,457
Arrow Re: Eddystone 888A.

Wendy: two remarks from me.

1. Many moons ago, I had a fault on my 888A on 21 MHz where I could not get the local osc. to track. Like you, I went back and forth until I eventually either ran out of adjustment or gave up, realising that something was seriously wrong and serious intervention was thus required. All of this was such a long time ago that I cannot recall the degree of analysis and proof that I undertook which led me to substantially alter the value of a small value capacitor in the L.O. tuning. As I said, all of that was many years ago - but all these years later, with this receiver having undergone several re-alignment checks since, without any problems making any necessary adjustments (trimmers and coil cores only), that impromptu capacitor is still there. The reason for that substantial change was - and remains - a mystery, especially as all the other caps. in the L.O tuning measured the correct values. The most likely suspect - since it is the one item that wasn't measured nor replaced - is the coil itself, but a much later analysis of the cct. operation seems to rule that component out of the list of suspects.

When I get obscure faults like that, I like to fully understand why - but sometimes with such faults, I simply have to settle for "Ah! Now it's working correctly: I'll settle for that!" Fortunately, that approach doesn't keep me awake at night worrying about it!

One further idea: the tuning coil may have two positions for resonance: one with the core well 'down', one with the core 'up'. You may have had the core in the wrong position: always worth a try.

2. Re the ECC85 modification. There was one little detail I inadvertently overlooked: sorry, but easily remedied. The previous owner had made a module to house a few extra components. It consists of a B7G paxolin valve plug and a B9A chassis-mount socket. The B9A socket has a hand-made cylindrical metal skirt made of tin-plated thin steel sheet which is soldered to the lugs of the B9A socket, thus acting as an earthed screen. Mechanical support between the B7G plug and the B9A socket is accomplished by stout wiring (where no additional components are required - e.g. heater wiring and ground connections). That mod. is thus readily reversible, but the removal of the entire ass'y. from the B7G socket and the subsequent fitment of a 6BA6 does not noticeably affect the set in any readily appreciable manner * - after a few minor compensatory adjustments to the R.F. trimmer in the coil box.
Hopefully, that description will enable you to construct similar - if indeed you so choose.

* One day, I really should perform the 'two signal generators intermodulation test' for that modification. But I have enough things here that aren't working, or not working properly, (plus various design projects, and some of those have nothing to do with electronics & radios) to keep me going for years!

Al.
Skywave is offline  
Old 8th Dec 2015, 11:50 am   #75
Wendymott
Octode
 
Wendymott's Avatar
 
Join Date: Jun 2013
Location: Huddersfield, West Yorkshire, UK.
Posts: 1,782
Default Re: Eddystone 888A.

Hi Al. That confirms my suspicions in as much as..... maybe it was a..."its nearly ok" let it go. The Core has only one peak, and that is fully or nearly in. The beehive is nearly out.
I thought about the osc not tracking and looked at the coil and components there too. It is a very "Techy" adjustment for either. I am getting to the feeling that as I will not be bothering with 15, I may let it pass.
Ref the ECC85 mod, I think, for the moment I will leave till I have nothing to do... which is a long way off....
One other thing... the "Top Band" beehive trimmer in the L/O was broken, the ceramic post was broken, thus the adjust element was floating about and shifting the upper tuning point. I replaced with the RF coil trimmer and retro fitted another style on that coil.
"NEXT" ?
__________________
Should get out more.

Regards
Wendy G8BZY
Wendymott is offline  
Old 8th Dec 2015, 12:37 pm   #76
turretslug
Dekatron
 
turretslug's Avatar
 
Join Date: Nov 2011
Location: Surrey, UK.
Posts: 4,385
Default Re: Eddystone 888A.

I wonder if the fact that both of these sets wouldn't align properly on 21MHz points towards an underlying systematic error, reminiscent of the way that some R1155s had an incorrectly calculated padder value on one range. I can imagine that development and production schedules sometimes (often?!) get pushed beyond deadlines and errors creep through.

When I acquired my 670A, I did my usual quick ohm-meter scan through all the cathode resistors before powering, just to be sure- the AF amp's read zero to chassis rather than 2.7k. That'll be that TCC elko, won't it?- nope, the capacitor lead-out went from cathode to g3 pins, through the central B8a spigot and on to chassis unjointed. It all looked original, the soldering had the same look and degree of tarnish as everything else, so I assume that it was a production line error, with who knows how many other sets afflicted. The UAF42 pentode and detector sections would have been under-biased, but I dare say it worked happily for years with anode volts a bit low and distortion a bit high.

You're not kidding about RF box valve socket accessibility! I thought, if I have to change a base in a 750, how the heck would I do it? I decided that one might have to withdraw the band-change shaft, carefully undo and withdraw the long, thin wafer asembly/retaining screws (and their assortment of spacer, fibre and lock-washers...), remove all the screws retaining the coils for one section (fortunately, they go into a tapped brass bar on the other, top of the coil-box to facilitate bonding to the variable capacitor- I do like the mechanical detailing on these sets) and finally lift out wafers and coil-set as a complete, slightly wobbly assembly, having unsoldered its various connections as it lifted clear. Hooray!- valve sockets would now be clear and unobstructed. Actually, there would be a bit more to do. At that point, I decided only to worry/investigate if it actually need doing! Crossing bridges and coming to them etc etc....
turretslug is offline  
Old 8th Dec 2015, 12:40 pm   #77
Skywave
Rest in Peace
 
Join Date: Jun 2006
Location: Chard, South Somerset, UK.
Posts: 7,457
Arrow Re: Eddystone 888A.

So, in summary: the coil is set for max. inductance and the trimmer is set for min. capacitance - and presumably any other combinations of those two fail to give even a near calibration and all your investigations lead you to the conclusion that there are no defective capacitors, variable nor fixed. Hmm. Time for a different way of looking at this problem: two things spring to mind.

1. I know this seems very unlikely, but is the beehive trimmer defective? Can you remove that trimmer and check its capacity on a capacitance meter whose calibration you can trust or independently verify?

2. Turn the problem 'around'. You want the tracking to be correct - which at present, with the capacitors fitted, it isn't. So, change the value of the fixed capacitor which is in parallel with the trimmer until the tracking is correct. (Never mind what its 'correct value' is; never mind about originality. You simply want the set to 'work' - as was originally intended). The easiest way of doing that, assuming that you have the kit for accurate capacitance measurement, is to adjust the trimmer for half max. value of its capacity, refit it at that setting and then experiment with new values for the fixed capacitor which is in parallel with the trimmer. Since that trimmer, at present, is at min. capacity, then you know that the fixed cap. is too high in value. Therefore you know which 'direction' to go when trying replacement values for the fixed capacitor in parallel with it, i.e. less capacity.

It's very rare that I find myself having to resort to such measures to fix a fault, but for me, the old adage of "When the Devil drives . . . " calls the shots. But, most importantly, I always make a careful note of such a change, including why, when and how, etc. That way, if a later date, another strange and similar problem occurs in the same area, I have an accurate record of what I did.

And the very best of luck with it!

Al.
Skywave is offline  
Old 8th Dec 2015, 1:23 pm   #78
G6Tanuki
Dekatron
 
G6Tanuki's Avatar
 
Join Date: Apr 2012
Location: Wiltshire, UK.
Posts: 13,953
Default Re: Eddystone 888A.

Can you stick a frequency-counter on the LO and check you've actually got it tuned the right side? If you've inadvertently aligned it the wrong side, it'll be 3.2MHz out, which is something like 15% - which would make the resultant tracking/padding a bit unpredictable.

[Don't laugh - I once did something similarly silly and wasted an afternoon over the frequency-multipliers in a PMR radio: I had the tripler doubling and the doubler tripling and so wondered why - though things peaked - there was a significant lack of drive to the PA valve].
G6Tanuki is offline  
Old 8th Dec 2015, 2:50 pm   #79
Skywave
Rest in Peace
 
Join Date: Jun 2006
Location: Chard, South Somerset, UK.
Posts: 7,457
Arrow Re: Eddystone 888A.

Quote:
Originally Posted by G6Tanuki View Post
Can you stick a frequency-counter on the LO and check you've actually got it tuned the right side?
You know, I never thought of that! Perhaps that's because it's a bit too obvious. Moreover, it has reminded me that I did have that trouble with my 888A soon after I acquired it: 28 - 30 MHz. But that was a very long time ago & my memory isn't what it used to be . . .

However, a quick way of checking for an incorrect L.O freq., where the L.O. is on the wrong 'side' of the incoming R.F. signal is this: If the L.O is on the 'high' side, then the image is on the 'high' side, and ditto for 'low' side tuning (which is rare, anyway). In the absence of a freq. counter, a signal generator will soon tell you if the L.O. is on the wrong 'side' by simply comparing the signal strengths at the required incoming R.F freq. and at the image R.F. freq. That does assume, of course, that an R.F. stage is present and that it is correctly tuned.

Al.
Skywave is offline  
Old 8th Dec 2015, 4:43 pm   #80
G4XWDJim
Octode
 
G4XWDJim's Avatar
 
Join Date: Oct 2008
Location: Littlehampton, West Sussex, UK.
Posts: 1,465
Default Re: Eddystone 888A.

Eddystone 888s seem to suffer commonly with this problem. I wonder if it's because each lower frequency band only covers a small amount like half a meg. I had the same problem many years ago with the two middle bands although I've forgotten exactly what the frequencies were and whether it was 888 or 888A.

I tried fitting silver micas of different values, even tiny variable ones to experiment and get the swept coverage correct. I lost interest and got fed up with it before I succeeded and moved it on.

The problem as I remember it was that the oscillator wouldn't change freqency enough from one end of a band to another. If the bandspread band covered say 500 kHz change from end to end the local oscillator had to change its frequency by the same amount albeit between two points eg 3.5 to 4.00mHz. I recall that it would only change by 460 kHz or something like that. I even began to think that some of the moving vanes on the oscillator tuning capacitor were making a bad connection and instead of varying between zero ish and 500 pf the maximum closed capacity was perhaps only 450 ish. At that time I didn't have an accurate capacitor measuring device and had better things to do with my life which was gradually ebbing away staring at Mr Eddystone.

I used a digital receiver to listen to the local oscillator which is a very easy way to get the end points spot on - on most receivers anyway. Even easier than a frequency counter apart from the mental maths involved.

I look forward to hearing the answer.

Jim
G4XWDJim is offline  
Closed Thread

Thread Tools



All times are GMT +1. The time now is 7:00 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.