"The phone is a remarkably complex, simple device,
and very rarely ever needs repairs, once you fix them." - Dan/Panther

Main Menu

Adjusting WE500 dial speed...

Started by twocvbloke, January 29, 2012, 01:19:44 AM

Previous topic - Next topic


Okay, this will be a bit daft asking AFTER I just adjusted my black WE500's dial (not yet braved the red Tenite WE500's dial yet), but, I just thought I'd ask if I did it right... :)

What I did was remove the spring on the governor and push the hooks inwards to I guess tighten up it's tension or strength, which holds the brakes back a bit more, thus increasing the speed at which the dial can turn. To time it I just watched the second hand on the clock on my computer, winding up the dial and then releasing as it ticked onto a new second, and once the dial stopped as the hand ticked over again, then I felt it was at the right rate... :)

It'd be nice having one of them pulse counters like what the GPO used to test & repair dials, but you can't have everything!!!  :D

Anyway, it did work and I'm now getting a proper 10pps rate for the finicky BT line (it worked fine on our old NTL/VirginMedia line which I guess was more forgiving!), it broke dial tone when I dialled 0 (doesn't call the Rotarepo, I mean Operator here), and I dialled my mobile number successfully, so, I'm happy as larry... ;D

It's different to adjusting GPO dials, on those you just gently squeeze the governor brakes together and that sets the spring tension, testing as you go of course... :)


Was just thinking it would be quite easy and cheap to hook up a sound source, such as a small piezo buzzer or other sound producing device, to a 9 volt battery with a couple of alligator clip wires using the dial pile to chirp the piezo when the dial was spun. Placing the piezo next to a computer microphone recording the chirps would result in a fairly accurate poor man's visual pulse counter.

10 pps would sort of look like the following.

Anyone try this?


Now there's a thought, but rather than use buzzer, you could hook it up to the microphone port directly (Mic. sockets usually put out voltage, which should be enough to pick up pulsing, assuming it doesn't short out the mic. hardware causing it to burn out!!), and using something like Audacity to do the recording, then you can get a similar result to that above... ;D

I've got to try doing that now, just need to find an old sacrificial soundcard to try it on...  :D


Ha! I used Audacity to generate that 10 Hz sine wave and used Photoshop to crop out only the wave. I was working on some sound files in Audacity when I read your post late last night here (2 am).

Yeah, I thought about hooking it directly into the mic 1/8 port but with not knowing impedance and power, I didn't think that was a good idea if one wasn't electronically knowledgeable about those things. But just shorting out the mic (open and close) should/may produce a "safe" pulse/click in the recording software. That may work and be cheaper and simpler.

Let us know what you did and how it works.


I like using Audacity for some audio editing and whatnot, I've even created telephone tones (Dial, Ring, Busy, three-tone error, etc.) using the frequency generator built into it for fun, but usually I just use it to try to restore and improve pieces of music I find on youtube that have been badly uploaded or encoded, cos I'm like that...  ;D

But like I say, I need a sacrificial sound card, the ones I have I'm quite fond of, or are integrated and not replaceable, I'll probably just get some cheapo thing off ebay... :D

I'm just wondering if it's easier to plug a whole phone into the Mic port via a home made adaptor (3.5mm mono plug to an RJ11 or BT-slave socket), or to wire the dial directly, how phones work is still somewhat a mystery to me, especially the WE500s with the goo-filled Network blocks... :D


Right, I finally got round to doing this (a combination of listening to music through a GPO phone handset and Hobgoblin beer!!), and have come up with a simple way to test the dial speed, at least on the red WE500... :D

So, all I did was connect the phone's red & green wires to the Ground and Left/Right wires of an audio wire, using a GPO BT52A as a connection block (handy seeing as the WE500 still has it's hard-wired cable attached), and I plugged it into my laptop's microphone socket and using Audacity to record the Microphone source, I dialled away, dialling 0 0 0 0, seeing how 0 should be 10pps, and, er, the red WE500's dial is a tad fast, but seems to operate the BT hardware at the local exchange, so that's okay I guess...  ;D

It works surprisingly well, and as you can see from the pictures, you get a good and clear indication of how well it's pulsing, and with a careful trim of the audio waveform, you can count the pulsing and see what it's rate is... :)

And the obligatory pictures:


Good work.

And, unlike my simulated square wave pulse photo, your actual capture of the 10 pulses for "0" is correct with the next pulse being at the beginning of the next second. Your actual capture is exactly 10 pps where my simulation would be a pulse fast, for what it's worth.

Anyway, nice to see we have a cheap and easy way to test and/or set dialing pulse speed.


I was thinking that the pulses would have spiked on the wave form at each tenth of a second, and they're a tad off, so that made me think it's just that little bit fast as they are spiking at just less than 1/10th, but it dials fine anyway, so it's within tolerances... :)


Starts off fine but increases after 3 pulses. That may just be the physics of a spring turned mechanism starting from no movement to full speed, it overcame the moment of inertia and picked up momentum. Checking a dozen or so phones nay show similar peculiarities. A dirty dial may even slow down as the spring unwinds. Stuff like that.


 :D Great idea.
I tested my 1948 SC 1243.
The dial was stuck, so gave it a drop of oil here and there when I got it, maybe a little to much?

The speed test with Audacity and the phone just plugged in to the line gave a difficult reading.

I dogged out an old tone generator, and tried again,  ;D ;D ;D
The grey fields are the breaks, and 10 pulses finished at 0.925 instead of 1 isn't bad!  :D
(=10.8 pulses pr sec.)


I'm sure it's within tolerances, but it is a handy way to test dials for sure... ;D


 :) In Norway the pulse ratio is 10 pulses pr second +/ 7%
The pulse consist of 60ms break and 40milliseconds make.
This makes the total timing to be from start of first break to 40ms after last break.
Or 10 pulses within 0.96 sec. +/- 7%  Or everything between 892.8 milliseconds and 1027.2 milliseconds is within standard.
The exchanges will accept a wider range, and are designed to tolerate a 67 to 33 ratio to.
Then my SC telephone is 3.65% too fast,  and I am impressed. I do not believe this ever has been tuned since 1948.
My job with the oil can may cause a slowing down of the speed if it collects dust, and happend to be spilled in the drum of the centrifugal regulator. Then a real cleaning has to be done, but I see now reason to repair something working that well.



How did they do it in the old days?
I got this from inside the dial of my Siemens modell 1936.


(If you need the original scan I may send it in some way.  The Orignal BMP Af scan is 101937kB and the JPG cutout is 1575kB  :D)