Thermal printer head made out of PCB

Question

I'm interested in only the thermal printer head, the part the gets hot. This question is not about the motor, gears, etc.

First, I would like to know as much as possible about a standard thermal printer head: how these are manufactured (materials, processes, etc.), the temperature profile for each one of the "pixels", their size (resolution), how they're distributed, how they're driven, how much power they need, how fast they can go from cold to hot and vice versa, and also how thermal paper reacts to temperature and time.

However, the main question is: would it be possible to manufacture a basic thermal printer head using only/mostly standard PCB manufacturing processes?

Answer 1

Modern thermal print heads are typically made with a thin-film process, on a ceramic substrate with glaze. I can't show you the details in a diagram because the document I have is marked "Kyocera Confidential".

In order from the base to the contact surface there is:

• Ceramic substrate
• Glaze layer (makes the heater convex to contact the paper)
• electrodes and heater elements deposited on the glaze and etched
• protective overcoat

Could you make a print head with a PCB process? A very crude, slow and unreliable one, perhaps. All that is required to make the paper turn black is to locally raise the temperature sufficiently. Since rigid PCBs are flat, limiting the spread of heat in the paper movement direction would be difficult (consider the contact area created by a compliant platen roller), but resolution in the orthogonal direction would also be pretty bad due to PCB manufacturing limitations and the limitations of having one material for both electrodes and heating elements. Maybe you could get results similar to mechanical television sets for a limited number of prints.

Answer 2

To add to the answer:

1. As for the likely temperature, Optimal Operating Temperature for Thermal Paper contains:

   The optimal temperature for thermal paper can vary, but a common range is between 150 to 185 degrees Celsius.

   Which is above the 130°C Tg or glass transition temperature at which What Is The Maximum Temperature of FR4 PCB? says:

   Where it changes from solid to a soft and rubbery state.

2. As for abrasion, looking at a sample datasheet for a thermal print head the Abrasion Life is specified as 150 km which is presumably the length of material which can be printed before the print head is worn out. The datasheet doesn't specify the surface material to achieve the abrasion life which is unlikely to be a design parameter for a normal FR4 PCB.

Answer 3

standard PCB manufacturing processes?

Define "standard" here. I remember you can very solidly burn your finger on a thermal printer head, so temperatures above 120 °C are likely – and that's not OK for most "standard" PCBs (made of cheap FR4 fiberglass-epoxy laminate). But higher-temperature media are a thing, so generally there's nothing wrong with printing an array of thin traces on a ceramic PCB (many PCB manufacturers offer different substrates).

The problem here becomes the trace material: Can't really put soldermask atop of something that's getting drawn by paper – soldermask isn't designed to withstand constant abrasive action.

Can't leave the copper bare – same problem.

My suspicion here is that in general, thermal printer heads are indeed made of "usual" materials (ceramic substrate) and the conductive stuff is printed on, but it's not copper as in standard products.

You can, by all means, try to just order PCBs (pcbway has substrate that has a glass temperature of 150 °C; that should be enough if you actually do control the current through these traces individually – not that hard; all you want to do is make sure your trace resistance doesn't increase from its cold resistance more than is thermally OK for the substrate); you'll want to thermally attach a big cooler to the backside of the PCB.

Will this work forever? Probably not. Will this work for a prototype? Probably!