Attenuation of a splice made with Signal Fire splicer – test

The market of fiber optic splicers has changed significantly over the last few years. Japanese splicers of renown brands (Fujikura, Sumitomo, Fitel) and Korean equipment (e.g. Inno), keeping pace in many aspects, have been joined by cheap Chinese ones. Splicers made in China were available on the market much earlier, but the product range was not as extensive as today and, above all, the price difference between the Japanese and Chinese devices was not as significant as today. This situation, of course, forced the Japanese manufacturers to adapt to the market and they have released simple models, which, although their prices still differ from the Chinese models, are much closer to them. The choice remains whether it is better to buy a Japanese splicer, almost twice as expensive, with a basic positioning method, or whether to buy a Chinese one, which looks better in many aspects "on the paper".
As expected, Chinese splicers such as Signal Fire AI-8C L5870 or AI-9 L5875 dominate selected areas of the fiber optics installation industry. As far as operator applications are concerned, they are still second category equipment – possibly as the backup devices, but in the field of ISP installations, FTTH networks, multifamily houses, TV/SAT or CCTV systems the cheaper splicers prove to be an excellent choice.
A lot can be said about the parameters, but from the point of view of an installer investing in this type of equipment two things seem to be of critical importance:
  • reliability (and quick service in case of a fault),
  • good quality splices (i.e. durable and with low attenuation).
The failure rate of the AI-8C and AI-9 models is negligible. Service statistics from 2020 show less than 1% for the more popular AI-9 model, extremely significant given the huge number of units of this type of equipment sold, and the fact that much of this small number of service calls was due to lack of proper user training. Servicing of the AI-7 or AI-8 models, which have been in use by installers for at least 2 years (at which time the models were replaced by the current ones), is also rare. For this reason, splicers of this brand coming from the official distribution network of Dipol are covered by a 4-year warranty.
A good quality splice is a slightly more complicated subject and, above all, it depends on a number of factors the crucial ones of which are unrelated to the splicer itself. These include a well-adjusted fibre cutter, cleanliness while splicing and the knowledge of the installer. Many forget about the basic calibrating the splicer arc, and they fail to notice or do not want to see the irregularities during the splicing resulting from a fibre that has not been fully cleaned or has been badly cut.
Splice attenuation test
The remainder of this article will explain the procedure for testing splices made with the Signal Fire splicer. The splice attenuation measurement is performed using the transmission method, i.e. using a 1310/1550 nm light source and an optical power meter.
For the test, special boxes have been designed that contain about 120 m of G.652D standard optical fiber terminated with SC/APC connectors on one side. The other end of the fiber remains free – it can be freely unwound and repeatedly alternately spliced and broken. In addition, the boxes are equipped with a splice tray.
Test bench with dedicated fiber boxes and Signal Fire splicer
The procedure includes measurement of splice attenuation using the so-called transmission method. The attenuation of a single splice is determined from a reference measurement. The following steps include:
1. Making a splice so that the measuring equipment can be connected and a reference measurement can be made.
2. Connecting the measuring equipment. Saving the power level value in the meter's memory and resetting the settings (saving as a reference for future measurement).
3. Fiber breakage.
4. Splicing the fiber again.
5. Validation of the attenuation indicated by the optical power meter. This attenuation is directly related to the repeated splicing.
6. Repeating the procedure to verify the repeatability of the meter readings.
First fiber splicing for the reference measurement. The attenuation of the connection is not important for the procedure - the one indicated by the splicer was 0.01 dB.
Connecting a group of measuring equipment: a 1310/1550 nm light source (left side) and an optical power meter (right side). Equipment of this type can be found here. Since the measurement fibers are terminated with SC/APC connector and the light source requires SC/UPC connector, it is necessary to use SC/APC - SC/UPC L3412 patchcord. To keep it in order, similar setup has been made at the meter side, although in this case it is not necessary (meter connector allows use of both types of plugs).
Measurements for the wavelength of 1550 nm were performed first. This transmission window highlights the imperfections of the transmission path more. The signal level generated by the light source is -5 dBm. The signal strength recorded by the meter on the other side of the fiber is -7.11 dBm. It means the attenuation of -2.11 dB (all connectors, fiber, splice). This attenuation value is not significant, since it is the attenuation of a single splice that will be measured. The current reading will only be a reference for the actual measurement. The reference reading is shown in the second line and is currently 0.00 dB (the meter has been "reset" and has saved the reference value). Any further attenuation (added e.g. by another splice) will be visible here.
After breaking the fiber, as expected, the power meter indicated values typical of no signal (-50 dBm on a linear scale i.e. the lower value of the meter's measurement range and -42.87 dB with respect to the saved value).
When splicing the fibers again, while watching the power meter readings, interesting phenomena can be noticed:
/The initial positioning of the fibers alone in the splicer causes the meter to start detecting the signal again. What is particularly interesting is that the power of this signal is lower than the reference power by only 0.76 dB. Such an attenuation value is not uncommon in mechanical splices or in poor quality detachable connections. If, at this point, both sides of the fiber were connected, for example, media converters will successfully provide data transmission, despite the fact that the fibers are not physically connected.
After the splice was made, the final value of the additional attenuation, which is in fact the attenuation of the splice made, was recorded. This value was 0.05 dB (the splicer indication is 0.01 dB attenuation). The generally accepted standard is to allow splices with attenuation not exceeding 0.1 dB. However, there are factory standards and European standards that allow 0.15 dB or even 0.3 dB of attenuation. The splice tested is therefore very good in terms of signal attenuation.
For the sake of order, the splice was protected with a heat-shrink sheath to verify whether the attenuation value changes, which actually did not happen. The durability of the splice was verified by pressing the splice along with the shield into a dedicated spot in the splice tray. This did not change anything in the context of the measurement.
The above procedure was repeated several times for the 1550 nm wavelength, each time obtaining attenuation results in the range of 0.03 - 0.06 dB. For the 1310 nm wavelength, the indications were lower and ranged from 0.00 dB - 0.03 dB.