I want to use my Graflite flash bulb flash on my Canon 5D Mk III.

  1. I know that modern DLSRs have a limit to the voltage and current that a flash can use when triggering, before it damages the camera.
  2. I know that most similar questions concern using older electronic strobes / flashes which can have too high a trigger voltage. My Canon for example can handle a maximum of 6 volts trigger voltage. That info is freely available.
  3. Camera flashes that use actual flash bulbs however, mostly have the opposite problem. In the case of my Graflite flash, the voltage is only 3 volts, BUT the triggering amperage / current is much higher than electronic flashes.
  4. I have built a circuit (No the Wein Safe-Sync does not work for this situation) to isolate the Graflite triggering current from the Canon's PC port.

My problem is that I can find nowhere, what the Canon 5D Mk III current limit in milliamps. I have called Canon more than once and they say that information is proprietary, and refuse to provide it.

So here is the question: I need to know if someone can tell me what the maximum milliamps a Canon DLSR can handle during the flash triggering sequence from the PC port.

I have my triggering circuit (That part that will connect into the Canon PC port) down to 4.85V & 70mA. But is that low enough?

I assume that someone with an electronic oscilloscope could connect a later model Canon flash to that scope and see just what triggering current is output from the flash. So, if anyone has such a scope and a Canon flash, I would really like to have that answer also.

  • 1
    Actually, a 5DMkIII can withstand 250V (not 6V) on the hotshoe/PC sync port.
    – inkista
    Oct 23, 2017 at 23:17
  • @B.Varner Only the older original Digital Rebels (300D), 10D, D60, D30, and the lower tiered film EOS cameras were limited to 6V tolerance. Many of the upper tier film EOS cameras and all of the Canon DSLRs other than those listed above can tolerate up to 250V. What EOS digital camera are you using?
    – Michael C
    Oct 24, 2017 at 0:03
  • @MichaelClark, OP says 5DMkIII in the text.
    – inkista
    Oct 24, 2017 at 1:32
  • Then I wonder where the "freely available" source is that says the 5D Mark III can only tolerate 6V?
    – Michael C
    Oct 24, 2017 at 1:41
  • @MichaelClark I'd guess it's that damn oft-quoted botzilla page. Which was created for Powershot G users back in 2008.
    – inkista
    Oct 24, 2017 at 18:13

3 Answers 3


The Canon engineers are probably correct in saying that the PC port is a low-voltage port. I would not recommend putting much current at all through that port.

There are a couple of ways of solving your problem:

  • Add a single transistor on the input to your flash. Then, your flash trigger power will become negligible, because the transistor will be triggered through the camera, and all the high-current flow will be within the device itself.

  • Use the hotshoe. The center pin of the hotshoe can be used to trigger most flashes. I can't say with certainty what Canon uses for the hotshoe mount in the 5D Mark III, either, but according to a discussion thread on Electronics point, the XT uses a 2SK2715 MOSFET for the flash trigger, and it's probably a good bet that they use the same part on all their cameras. The 2SK2715 is made by ROHM Semiconductor, and according to the spec sheet, it is rated at 500V at up to 2A. Canon downrated their maximum spec to 250V, either as a hedge against slightly substandard parts or to allow them to source other equivalent parts that may have looser tolerances. And, of course, a part designed to handle up to 1 kilowatt triggering power laughs at your 0.3395 Watt triggering current requirements, and any even-semi-plausible replacement MOSFET that handles such high voltages will have similar specs.

If you're worried about it, use one of those approaches.

  • Well maybe you are correct, but. My information was from 2 calls directly to Canon. First call was to a support person. The second I got through to a person who said they were an engineer. Both stated that the max voltage was 6V (For 5D Mk III). Maybe they were just covering themselves because of what I was planning to do, but that is an large difference..... Would be nice if someone actually had some written documentation to this affect.......
    – B. Varner
    Oct 24, 2017 at 2:50
  • Anyway, I would love it if someone else wanted to use their $3,000 DLSR to test this theory so I did not have to use my own.......
    – B. Varner
    Oct 24, 2017 at 2:53
  • Oh, wait... you're talking about a side port. I was describing the hot shoe, which you should be using.
    – dgatwood
    Oct 24, 2017 at 4:34
  • I have updated my answer accordingly.
    – dgatwood
    Oct 24, 2017 at 4:47
  • PC and hot shoe sync contact should be the same point, simply wired together. Both have to wait for same shutter shorting signal, and then do the same trigger thing.
    – WayneF
    Oct 24, 2017 at 14:54

My Canon 50D says the PC Sync port is good for 250V, any polarity.

If the current of the flash unit sync input does not fall below about 3-5 mA the output will remain asserted forever. Sounds like it is driven by a bidirectional thyristor, triac or some such thing which has a recovery threshold.

The electrical characteristic is different for + & - polarities, even though they say either is OK. It pulls down to about 650 mV for + polarity, and about -140 mV for - polarity.

The hot shoe sync seems more robust for pulling heavier loads and pulling closer to zero V. It also recovers and turns off independently of whatever the load current is. The manual says don't use a high voltage flash on the hot shoe since it may not fire, but no warning about any potential damage if you do.

No indication of allowed current is given for the hot shoe either. Seemed to work fine up to about 34 mA, but I didn't try to go beyond that. I have an MT-24EX flash I could check the current on. That would reveal a lower limit of allowable current, but not anything about what the max might be.

Here is the measured hot shoe current for my MT-24EX at 2mA/div and 50mA/div:

Here is the measured hot shoe current for my MT-24EX at 2mA/div and 50mA/div:

The polarity should be understood to be current going into the camera hot shoe (2mA steady state, 120mA x 1usec initial transient). I should have flipped the orientation of my current probe.


This is not an answer you might want, call it comments, but I needed more bytes.

I always wondered if Canon's 6V sync limit was real? No designer would allow that vulnerability. The same paragraph in the manuals also said "Buy only Canon flashes", but I think it bit them instead.

A Wein SafeSync is just an optical isolator, two separate circuits, input rated to 400 volts that provides safe 6 volt out. I don't know how it might affect flash bulbs, or flash bulbs it.

Speedlights are 4 AA cells, 6 volts. This sync voltage goes to point I'll call "PC" through a high value resistor to limit current to a harmless value. PC is the PC connector, and also the flashes trigger circuit connects internally to PC. The camera shutter sync simply shorts PC to ground (mechanical switch in old days, SCR or similar today). This short drops open 6V at PC to 0 volts (harmless through the resistor). The trigger circuit watches PC sync voltage and triggers flash when it drops to zero. This 6V trigger current is NOT the flash voltage, which is a separate 300 or 500 volt circuit at the flash tube.

Some early electronic flash was different, with dangerous sync voltage, exceeding 200 volts, dangerous to electronics today (the market for Safe-Sync). The old BC flash just charged a capacitor to provide more current for a flash bulb, often with a 22 volt battery.

I think the application simply needs a modern flash.

How is that going to work? Modern cameras have no M or F sync for flash bulbs, which must be triggered several milliseconds early so they can become bright before the shutter opens. If it is an old camera with M or F sync, then no electronics, and the current won't be any issue. But flash bulbs have no use for modern X sync for electronic flash, which is all modern DSLR offer.

You might use a real slow shutter to wait open for the bulb to light up. M sync at 25 milliseconds is an extra 1/40 second waiting for the bulb. That's what M sync is, a 25 millisecond delay. F sync was about 10 or 15 milliseconds. M medium delay, F fast delay, for faster bulbs. But modern electronic X sync is instantaneous. Does your planning include that?

  • Ok, this is the problem I get when I ask this question of anyone. The responses given above are mostly not an answer to the specific question I asked. Let me explain. I am not trying to attach an old electronic flash to my DLSR. I am trying to attach a flash bulb flash. Totally different. Old electronic flashes have high voltage. Flash bulb flashes have high amperage or current, not necessarily high voltage. My flash bulb flash is only 3 volt, I have measured it. No capacitor in it to boost power. It does have very high amperage to light the flash. Too high for my DLSR.
    – B. Varner
    Oct 25, 2017 at 4:01
  • Too high for Wein Safe-Sync which will not trigger this flash. It was designed for old electronic flashes, not old flash bulb flashes. I called the Wein reps here in the US and that is what they told me. Graflite flashes have no way of connecting to a flash shoe. Only via PC cord. So, since it appears no one knows what AMPERAGE the camera can handle, I was hoping someone had an oscilloscope & could test a newer Canon flash & tell me how many AMPS come across when triggered. I would then use that info as a guideline in the circuit I am creating for this purpose.
    – B. Varner
    Oct 25, 2017 at 4:02
  • I was speaking of flash bulb M or F sync. M sync is a 25 millisecond shutter delay waiting for the flashbulb to fully ignite and be bright. DSLR offer only X sync. Your Graflite manual at cameramanuals.org/flashes_meters/graflite.pdf page 11 says "Do not attach a x-type shutter". Times change. You need a modern flash.
    – WayneF
    Oct 25, 2017 at 15:31
  • Halfway joking, but you could use DSLR shutter set to Bulb, then open the shutter, then short the end of the PC cord with a paper clip to manually flash the flash bulb, and then close the shutter. This should work in indoor ambient light (at say f/8), and the camera would be safe from the obsolete flash.
    – WayneF
    Oct 25, 2017 at 16:10
  • @B.Varner the equipment needed to reliably measure steep, low current, current pulses with an oscilloscope tends to be more expensive than a simple oscilloscope, and most oscilloscope owners will not have it.... Oct 17, 2018 at 21:44

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.