My analysis of this movement is similar to the one I did with for the Raketa 2609.HA (see https://www.watchuseek.com/f10/anatomy-raketa-2609-ha-4781423.html). Still, for completeness and to keep this document self contained I will repeat part of my previous text.

A comparison between this two movements focusing on the dial side as already been done by schnurrp here: https://www.watchuseek.com/f10/analysis-raketa-2609-2623-24-hour-movements-894141.html.

Some of the Raketa 2623.H technical specifications can be found in: «Raketa 2623 H» | ??? «??????».

According to Raketa, the Raketa 2623.H is based on the Raketa 2609.HA. In fact, as we will see, the movements are very closely related.

Like the Raketa 2609.HA, the Raketa 2623.H is a movement with a center second hand. The following picture shows the bridge side of both of this movements, the 2623.H on the left and 2609.HA on the right.

Fom the bidge side perspective the movements seem identical.

The next picture shows the dial side of both this movements

As it can be seen in the previous picture, to accommodate for the lager minute and hour wheels the minute wheel was moved slightly up. In respect to the rest of the dial side, the movements are almost identical.

I will do the disassembly in the following order:

1. motions works

2. keyless works

3. ratchet wheel, crown wheel, click.

4. balance assembly and pallet

5. train of wheels and barrel

The next figure shows the movements’ train of wheels.

It can be seen in the previous figure that the movements were designed such as to the fourth wheel being placed in the center and going through a hollow second wheel.

It also can be seen that there is almost no difference between the two except for hole in between the barrel and the third wheel.

The next figure shows the barrel, second wheel and its bridge for the two movements.

The next figure shows the main plate’s bridge sides.

The next figure shows the main plate’s dial sides.

**Jewels**

We have:

- 2 balance jewels
- 1 inpulse jewel
- 2 balance beaing jewels
- 2 pallet jewels
- 2 pallet bearing jewels
- 2 second wheel bearing jewels
- 2 third wheel beaing jewels
- 1 fourth wheel beaing jewel
- 2 escape wheel beaing jewels

The jewel grand total is then 16.

**The train of wheels**

The next figure shows the wheels of the train of wheels.

The wheel theeth and pinion leaves are as follows:

Barrel | Second | Third | Fourth | Escape | |

Teeth | 72 | 80 | 75 | 80 | 15 |

Leaves | 12 | 10 | 10 | 8 |

This movement’s balance wheel frequency is 18000 vibrations per hour (vph) which is euivalent to 18000/3600=5 vibations per second (vps). So 2.5 oscilations of the balance wheel occur every second, since an oscilations corresponds to a complete cycle of the balance wheel or 2 vibrations.

One cycle of the balance wheel takes 1/2.5 = 0.4 seconds. Noting that one teeth of the escape wheel advances per cycle and that the escape wheel has 15 teeth it can be concluded that one revolution of the escape wheel takes 15*0.4 = 6 seconds.

Because the escape wheel's pinion has 8 leaves, for every full rotation of the escape wheel the fourth wheel advances 8 teeth. Hence, every 6 seconds the fourth wheel advances 8 teeth. The fourth wheel has 80 teeth and so it takes 60 seconds (6*80/8=60) for one revolution of the fourth wheel.

The fourth wheel pinion has 10 leaves and rotates once every 60 seconds, so 10 teeth of the third wheel advance each minute. This wheel has 75 teeth, so it does a full rotation once every 7.5 minutes.

The third wheel's pinion has 10 leaves, consequently the second wheel advances 10 teeth every 7.5 minutes. As the second wheel has 80 teeth it can be concluded that the second wheel rotates once every 60 minutes (7.5*80/10=60).

Finally, one revolution of the barrel takes 72/12=6 hours, so depending on how many full rotations the barrel can do from fully wound to fully unwound the power researve can be worked out. If, for example, it can have 7 full rotations then the power reserve will be 42 hours. The power reseve of this movement is 40 hours, so 7 full rotations should be about right.

Note also that the frequency of the balance wheel, assuming that a full rotation of the second wheel takes 1 hour, can be calculated as (80/10)*(75/10)*(80/8)*(2*15/8) = 18000 vph.

**The motion works**

The motion works are made by the cannon pinion which has 12 leaves, the minute wheel with 48 teeth, minute pinion which has 8 leaves, and the hour wheel with 48 teeth. This wheels are shown in the next picture.

Applying the same logic as before it follows that since the cannon pinion rotates fully once per hour (it is attached to the second wheel which does, as we havee seen, a complete revolution per hour) that the minute wheel takes 48/12=4 hours to rotate and takes 4*48/8=24 hours for one revolution of the hour wheel.

**Keyless works**

The next figure shows the keyless works in the winding position and in the hand setting position.

In the winding position, where the stem is totally pressed in, the slidding pinion is in a position such that the winding pinion is engaged with the crown wheel so the movement can be wound.

In this position, when the stem is turned clockwise the slidding pinion makes the crown wheel rotate which in its turn engages the ratchet wheel and wounds the mainspring.

The click pevents the ratchet wheel to turn backwards which makes the mainspring to correctly unwind. The crown wheel, click and ratchet can be seen in the next figure.

If the stem is turned anticlockwise the slidding pinion dos not move thanks to its Breguet theeth.

When the stem is pulled out, the setting lever turns on its pivot and pushes the yoke which also pivots and moves the sliding pinion to the hand setting position. In this position the slidding pinion engages with the intermediate setting wheel.

The intemediate setting wheel then engages the minute wheel which engages with the cannon pinion and hour wheel.

When the stem is pessed in and goes fom the hand setting position to the winding position, the slidding pinion is returned the position, where it engages with crown wheel, by the yoke spring which puts pressure on the yoke.

**The winding train**

The winding pinion which has 16 teeth, the crown wheel which has 28 teeth and ratchet wheel which has 53 teeth make the winding train. This 3 wheels are shown in the next picture.

For each 28/16=1.75 turns of the winding pinion the crown wheel has 1 revolution and for 53/28 turns of the crown wheel the ratchet wheel has 1 revolution. So for each (28/16)*(53/28)=(53/16)=3.3125 turns of the winding pinion the ratchet wheel has a full rotation.

If it is assumed as before that takes 7 full rotations for the main spring to go from fully unwound to fully wound, then about 23 to 24 full turns on the winding stem are needed to do so.

**The hand setting train**

The hand setting train is made of the sliding pinion which has 13 teeth, intermediate setting wheel wich has 15 teeth, cannon pinion which has 12 leaves, minute wheel, minute pinion and hour wheel.

One revolution of the cannon pinion means the minute wheel advanced 12 teeth, the intermediate setting wheel advanced 12 theeth and the sliding pinion advanced also 12 teeth.

So, since the sliding pinion has 13 teeth, when in the setting position an almost full turn of the stem makes the hands of the watch advance 1 hour.

**Conclusion**

From the comparasion between the Raketa 2623.H and 2609.HA results that this movements are vey similar.

On the bridge side the movements are almost equal: the respective wheels are identical; the second wheel bridges are identical; the wheels bridges are identical, except for the fact that in one case says 2623.H and in the other 2609.HA; and the main plates (bridge side) are almost identical.

On the dial side the keyless works are equal but the rest is different, for the main plate to be able to accommodate the bigger minute and hour wheels.

As we know there plenty of Raketa watches for sale in ebay claiming to be 24 hours while using the 2609.HA. I don’t have or seen one but I must say that now I am curious to see what were the modifications made.

I am no specialist but it seems to me that is possible to convert the 2609.HA in the 2623.H, but, like noted by schnurp in his post, that requires expertise and machinary, as this author says “Not exactly a ‘do-it-yourself’ job”.

And this ends the post. Hope you enjoyed it.