PDA

View Full Version : Trace width/thickness question.


matt@strat
08-23-2002, 08:58 AM
In reference to the trace width/thickness chart, what voltage is this based on. I am doing a board that will be putting out 28 volts and up to 4 amps. When using the chart do I need to add a fudge factor for the increased voltage?

Tom
08-23-2002, 09:14 AM
Volts are not part of the equation. Makes no difference if it's 5 volts or 500 volts. The fusing point (melting point) of metal is based on AMPS and the ambient temperature.

When calculating trace widths, as a rule of thumb we always round up. Our engineers call it "Insurance".

Also, there are two good calculators that you should be using. The Trace Current.xls and TrAMP.exe are both in the "Site Index" under Calculators.

matt@strat
08-23-2002, 09:45 AM
But the temp rise of the trace is a result of the power being dissipated in the trace. So for a given trace resistance and a given amount of current every time the voltage is doubled the amout of power/heat should double. A 5 to 10 volt jump wouldn't be a significant change, but a 5 to 30 volt jump I would think would be 6 times the amount of power/heat being dissipated by the trace.

randychase
08-23-2002, 09:50 AM
Temp rise is a result of current. I am designing a board right now with 10,000 volts all over it and those traces are really skinny. Now the spacing goes way up.

Interesting question, I will do some more research on it, since power is a function of current and voltage.

matt@strat
08-23-2002, 10:18 AM
OK after crunching the numbers I see the error in my ways.
The voltage rail is irrelavent, the voltage drop across the trace is what counts, and that is a function of trace resistance and current.
Guess I should of had another cup of coffee!!

Lameris
08-27-2002, 05:23 AM
10,000 volts is a new ball game. We had punch thru to the inner layers with 3000 volts. Spacing is critical!

It is also critical on those microvolt traces I use. It doesn't take much contamination or humidity to get a 100 nV signal change.

BPOWERS
08-28-2002, 06:18 AM
10,000 volts is a new ball game but very easy. Spacing is critical 10kv about 600mils min spacing.Our company designs power supplys and 10kv is usually the first stage.100kv-250kv using 100 mil trace is what we use here..(small amps)...some of my pcbs have 60-100amps running thru the etch (6 oz copper min 1" wide on longer traces)..biggest problem we have is the high voltage corona fields at the 200kv point that arcs into the pcb material and tracks along the glass fibers..we are experimenting with flex circuitry to aleviate the fields being set up..if anyone has more info feel free..

svarna
12-03-2008, 05:32 AM
Hello

We are Designing X-Ray generators for Medical Imaging. We are generating +/-60KV (max 120KV) using Diode Capacitor Multiplier. We are using 120 Nos of 1KV diode (SMD BYG21M from Vishay) in series to block 120KV.The Multiplier Boards are kept on transformer oil.Each Diode will carry current of 500mA at 150KHz. For Interconnecting each diode what is the Trace width you recomend ?(Trace Thicknedd is 35uM).As per your earlier discussion it shall be 100 mils, you indicated at small currents, How much is the current?
For high Voltage is smaller Trace width is better or larger trace width.

Pls reply with approximate trace width so that we can proceed with PCB design.

Thanks in advance
Best Regards
B.Sudheendra Varna

Quantum
02-11-2009, 05:49 AM
Hello

We are Designing X-Ray generators for Medical Imaging. We are generating +/-60KV (max 120KV) using Diode Capacitor Multiplier. We are using 120 Nos of 1KV diode (SMD BYG21M from Vishay) in series to block 120KV.The Multiplier Boards are kept on transformer oil.Each Diode will carry current of 500mA at 150KHz. For Interconnecting each diode what is the Trace width you recomend ?(Trace Thicknedd is 35uM).As per your earlier discussion it shall be 100 mils, you indicated at small currents, How much is the current?
For high Voltage is smaller Trace width is better or larger trace width.

Pls reply with approximate trace width so that we can proceed with PCB design.

Thanks in advance
Best Regards
B.Sudheendra Varna

for 500mA assuming 1oz copper tickness on PCB, for external routing layers you should use track width minimum of 0.115mm and for internal routing layers a minimum of 0.3mm

blackstick.co.uk
02-26-2009, 02:04 AM
Hi

If you download the C4 calculator from our website( Blackstick.co.uk/resources.html (http://blackstick.co.uk/resources.html) ).

There is a current carrying capacity calculator for working out how much current a track can carry basd on the copper weight and cross sectional area of the track.

There is also a voltage peak gap calculator for working out track spacing.

All of the calculations are based on the IPC 2221-A, and best of all it's FREE.

Hope this helps
Luke