Ballast Testing: What NOT to buy – $20 eBay Kit

This is a post about things you should NOT buy. Please DO NOT BUY things on this page.

I’ve noted that the SKBOWE will work with just about any HID kit, but when it comes to choosing a kit, there is a difference between “can” and “should”. The old adage “you get what you pay for” is very relevant, so avoid listings like this one:

eBay auction listing of what to avoid
An example of what *not* to buy for your Volvo

Of course, I couldn’t help myself in seeing how they could possibly sell an HID kit for less than a single D2S bulb, so I put in an order and the kit arrived 7 days later:

Warning Signs to look for before installing HIDs

If you observe any of these warning signs, you should not install the kit. They are easy-to-recognize:

Warning sign #1 – It was cheap. Obviously, at under $20 for 2 ballasts and 2 bulbs, this was not going to be top quality. The Kensun (Amazon) brand kits are $60, and are significantly better than this.

Warning sign #2 – It feels cheap. The first thing I noticed was the weight of the ballast – it feels like an empty case.  My postal scale says 3 Oz (85g), including the internal ignitor! Compare to the Morimoto XB35 at 10 Oz (295g). This is a DC ballast design, so even if it did work the bulbs would wear out very quickly.

Warning sign #3 – It looks cheap. Crummy sticker placed off center with no brand name, “input corrent [sic] : 10A” (this would be 135W, BTW). Back side has wood screws holding an aluminum plate onto the plastic case, didn’t bother to countersink so the heads are proud of the surface. No weather sealing whatsoever. No strain relief on the output cord (it looks like there is a nut, but it is cast and part of the case, the wire easily slides in and out).

Warning sign #4 – It’s built cheap. The internal construction of this ballast is actually scary, reminds me of the counterfeit usb chargers that routinely cause fires – but this is running at 5000 times more volts! No potting or through holes, which will shred this in an automotive environment. Integrated ignitor with no insulation. Random transformer rather than an actual boost converter. This is a disaster waiting to happen.

Guts of a super cheap HID kit.

Please do NOT use this type of kit in your Volvo (even with the SKBOWE)! When there is talk about “EMI” issues, it is these sorts of ballasts that are to blame. Any decent company producing a reputable ballast will be 100x better than this garbage.

Honestly, I am not even sure I want to hook it up to the testbench. But if the SKBOWE can run this, it really can run anything! Stay posted for an update with electrical performance.

Ballast Testing: Pro Tuning Lab “CANBUS HID” 35W kit

Updated 7/5/17 – See temp data below!

This post is part of the Ballast Testing series, where kits are evaluated against the SKBOWE TestBench.

H11 35W DC CANBUS HID Xenon Headlights Conversion Kit


protuninglab 35w kit in box

The Protuninglab 35w kit in box

This kit is being peddled on C30crew as direct-fit compatible, allegedly no SKBOWE or relay harness required. Well, we shall see about that… Picked one up for testing, note that the ballast case is not as pictured online.

The kit does indeed run without a SKBOWE, even at DRL duty cycles. There’s a catch though!

protuninglab kit scope trace with no skbowe
Voltage and current for the ProTuningLab 35W kit when run without an SKBOWE

That current (red line) is off the chart – the peak is over 15A in every cycle! The thin purple line is where the red line should be. This current distribution is characteristic of a “warm start” strategy where the arc is extinguished and re-initiated every cycle, but at low voltage (without using the ignitor) rather than ignition voltage (30kV).

This is not good for the CEM  or the ballast, indeed the integrated “error canceler” gets VERY hot after a short time. Notice the noise on the input voltage – this will only get worse as the internal components age – and is what can cause WMM failure without an additional ground point.

Update 7/5/17 – Upgraded test bench data

On the upgraded test bench, after about an hour the relative output dropped 10%. The whole time this 35W kit is pulling 55W (i.e. wasting 20W somewhere). But the real question is how hot is really hot?

ProTuningLab 35W CANBUS after 1hr of operation
ProTuningLab 35W CANBUS after 1hr of operation showing 145-150°F, i.e. a 80° rise in temperature from 70° ambient.

The “error canceler” on this thing hit 150°F, which is a +80° rise over ambient. In the car’s engine bay that starts around 140°F, this thing would be running near 220°F or 105°c – literally the top of the temp range for the cap(s) inside. I suspect that there is a resistor in there causing some of that heating, but it runs cooler at higher duty cycles so there is something else going on.

Stay tuned, when production is complete this bad boy will get dissected!