Avometer Model 40 Special (7.5mA FSD)

[pmmunro]

Thanks for the interesting pictures Dave. This will interest Andy I expect as someone must have drawn up the general arrangement of the meter to make it all fit in.

It would be good to see any mechanical drawings from the period which I imagine will be done to a high standard, more than can be said for the circuit diagrams I have seen.

PMM

[The Philpott]

Thanks Peter. I have had a much easier time with picture focus since setting my camera at '5metres' rather than VGA.

{Any issue with misalignment of contact spikes and battery box leaves is more problematic if it occurs with Models 8, 8/II, 8/III and 9/II, due to the spikes being mounted on a superstructure rather than tapped direct into the facia. As you have observed previously, if the meter is being worked on (or the case is replaced whilst the leaves are misaligned) the spikes are prone to getting knocked off and fracturing the platform on which they're mounted.}

As regards the Model 40(S) the resistance ranges are now working. The leaf switch contacts did ultimately require cleaning, although their poor contact was exacerbated by an unusual and excessive amount of endfloat and side play in the Rotary AC Selector Switch. I do have a decent replacement for it which might reduce the side play by up to 50%, but the cause of the endfloat is TBA. It feels as though a washer is missing, although if the hole for the lateral locking pin is drilled out of spec this would/could have the same effect. (I am accustomed to finding the occasional badly jigged component in an Avo of this era.)

Dave

[AndyGilham]

Originally Posted by pmmunro

It would be good to see any mechanical drawings from the period which I imagine will be done to a high standard, more than can be said for the circuit diagrams I have seen.

I shall have a hunt for some assembly drawings, sure I have seen a few! Not sure if it’s for the model 40 though!

Andy

[The Philpott]

On further assessment of the facia i notice that the rotary selectors are mismatched- the DC selector is of the taller type (knurled section nominally 7/32" deep) and the AC selector is of the shorter type (knurled section nominally 5/32" deep.)

On discovering this i looked at a couple of Admiralty meters of WWII vintage, which all have the taller type selector- which is more ergonomic to use.

I can't draw any definite conclusion from this anomaly, but if i remove the AC selector to establish whether it's excess clearance and end-float is curable i will be able to assess whether the 5/32" and 7/32" knobs have any other differences apart from the height of the bakelite section.

Also, on consolidating a fractured section of the (rear of the) facia with Araldite i noticed that the consistency of the sub-structure is quite unlike that which you would encounter on chipping the FRONT of the meter. I am accustomed to a granular appearance like that seen when snapping a piece of cheddar in half, but the section that i repaired on the rear had a layered, flaky almost crystalline look to it's sub-structure- more like a piece of coal (or bitumen that has cooled too rapidly.)

The facia being an abnormally thick (and intricate) moulding i can only assume that it doesn't necessarily cure/cool evenly. This tends to be supported by the warpage that can sometimes be clearly seen at the edges of early meters facias- some are slightly convex from top to bottom, some from side to side. I find it interesting that the company seem to have been pushing right to the limits of the material technology available to them.

Dave

[The Philpott]

As we have seen, if the scaleplate did not sit neatly against the mirror plate the factory would glue them together using a blob of black composition at the top corners. This also made the scaleplate more rigid, and ensured it did not move too close to the needle. In this case they had separated and i reglued them, using anything to hand as a heath-robinson jig to hold them together while curing. (interesting that an old recording on the dictaphone tape remained incorrupt.)

Blu-Tak on a lolly stick- the best way to get the last nut fitted when installing the battery box. I don't own BA box-spanner screwdrivers but i doubt they would be slim enough for this job.

Another anomaly- the '4' digit on the battery box lid is effectively the wrong way round. When the meter arrived here this lid was on back-to-front.

Dave

[The Philpott]

The two types of (AC) rotary selector switch:

The thicker type started life with deeper detents- in this case they are still in very good condition. Use of this type of switch might dictate a slightly longer coil spring to maintain full precision. The wear evident on the thinner type switch pictured has more of an effect on the precision of the selection than i thought- that little piece of grooving as the ball enters or leaves the crater renders the selection very mushy. This is the worn AC selector from the Model 40(S).

Note that the cam rotor is about 20 thou thicker on one of these assemblies. I have not done precise measurements yet but i suspect the rotor cannot be swapped to the other type of selector.

Dave

[The Philpott]

(Correction to previous post- the cam rotor could be swapped between selector assemblies on this occasion. The thicker cam rotor pictured would not have fitted beneath the leaf switch stack.)

I have fitted (visually identical) rotary selector knob/shaft assys. from a donor Mod.48A as they function much more positively than the worn originals. The low resistance range will now zero without wandering and it also manages to correctly measure a 20ohm resistor correctly on the (high)100,000ohm range (ie only 2 gradations on the scale!)

Note the four electrical connexions to the movement- the black insulation is more modern but the lime green and ref are old woven type. (The fourth one is hard to see- the swamp wire.)
There is a 'minimum' stop for the needle, provided with a spring coil to reduce shock (shaped like the back end of a safety pin, just visible in the pic.)
I made a 'maximum' stop from tinned copper wire, as this part was absent.

There was a certain amount of dirt on the glass and the scaleplate, so i decided the movement was coming out (back in now and meter assembled)
There are a few little defects on the glass that look like stray weld spatter or grinder spark damage. Something has melted a mark into the facia in the past, tending to support being in a workshop which included hot work.

A small arc blob which i had to remove from the 480vAC radial contact -which prevented smooth rotation from 1200vAC to 480vAC ,had me thinking of an end user report which someone mentioned a while ago- i believe it may have been written by the GPO and criticised the 1,200v range/s on the 40 series meters without specifying why they were less than satisfactory.. I believe the answer may lie with both these AC ranges being capable of handling UK 3 phase voltage- so toggling between the ranges may have reached and exceeded the capabilities of some meters when measuring, say, 400vAC.

If users were going by the book they would feel comfortable starting on the 1200v range then switching to 480v range, as the manual says you can do this- however i do wonder if this toggling without shutting off the power supply has damaged some meters. I certainly noticed an arb blob (and had to smooth it off with 600 wet and dry) on a Model 48A- in exactly the same place. One generally knows when an arc ball has been encountered in this way, as the selector is easier to rotate one way than it is the other.

Fortunately the scale plate paint is quite stable on this meter- there are a few aluminium oxide 'worms' underneath but i am not expecting them to spread.

Dave

[The Philpott]

I took my time out of deference to rarity, but i think i can call it finished now. The serial number is repeated on the rear of the facia and also the rear of the magnet. The 'No.11' legend from the scaleplate is repeated on the brass L section frame that supports the magnet. The divide by 2 leaf spring had to be removed and it's kink had to be accentuated- symptom of a feeble spring is the meter enters DB2 mode and will not return to normal ranges. Annoying if you have just replaced the movement!

No more info from the previous owner's daughter at this point.

Dave

[The Philpott]

The owner's daughter now confirms that both her Father and Grandfather worked for Lucas, the Grandfather started there just after WWII, as a tool setter, and left in the '70s. As positive a provenance as we're going to get... perhaps he skip-delved after hours, to his and our advantage.

Out of curiosity i took resistance measurements across the swamp and the movement when they were disconnected, accuracy of these measurements won't be fantastic as they were quite low...but it does mean i can compare with a standard movement in the donor 48A meter.

Dave

[The Philpott]

It appears that the movement is unique to this meter type. It occurs to me that if it were ever to go open circuit, a movement from a standard Model 40 might be used (should the scale shape match with a reasonable degree of accuracy) however as analogue aficionados are in relatively short supply i will ask at this point- how would i do this? A resistance across the movement i assume- but how to calculate the value? What i would effectively be trying to achieve is 20% less deflection for a given current.


Thankyou
Dave

[pmmunro]

It might be possible to use a standard Model 40 moving coil asssembly but that would depend on the moving coil characteristics in your special meter, which I don't think you've told us.

As you know, the next problem would be the scale shape. I don't know what factors affect the scale shape of an Avometer but I would expect that the uniformity of the magnetic field would be dominant and that the winding of the moving coil would be of secondary significance.

At the time when your meter was made, AVO would have addressed the problem, if necessary, by simply writing a new scaleplate to suit the repaired movement.

It may well be that your meter was set up by adjusting the "artificial ageing" of the magnet to give the required full scale deflection. This would have allowed a standard Model 40 moving coil assembly to be used.

PMM

[The Philpott]

Thanks Peter, para.2 is particularly thought provoking.

In case it was relevant i took the following measurements from all the windings associated with the movement of the special meter, and also of a normal Mod. 48A (in ohms):

Model 40(S) {Model 48A}
Winding on card- 8.5 {9.4}
Winding on bobbin- 13.3 {13.6}
Moving coil- 7.9 {11.7}

Dave

[micheal]

My guess is the movement is shunted with a resistor, are there any extra shunts
or bobbins when compared to model 40/48.
Mike

[The Philpott]

I couldn't see any obvious tacked on windings Mike, but i must admit i was focusing on the movement rather than the wiring of the meter overall. If i ever have it apart again i will compare it to a Model 47A(S) from 1944 which is effectively a Model 40.

I did notice a winding of 194ohm on a board piggy-backed onto the lower resistance board, but as it was in good condition i didn't investigate what it was for. (It is shown in the pictures in post No.14.)

Dave