John Curl's Blowtorch preamplifier part II

Status
Not open for further replies.
AX tech editor
Joined 2002
Paid Member
For the CE mark in Europe you could do self-certification, I once went that route when I designed an electro-skin-stimulator for a beauty salon :cool:

Self-certification means that you document all your design steps and decisions, especially towards customer safety, into a Design File.
You can then just slap on the CE mark without any 3rd party input.

The idea is that when something ontoward happens and you get sued, the Design File would show that you took all reasonable measures, used all reasonable (over) dimensioned parts etc, so that you can't be blamed.

In practice of course there's always the lawyers, but it gives you a way to use the CE mark, that lots of people trust, without too much cost.

jan
 
Jacco,
What are you referencing with the tail lights on American cars in Europe. I know that all light assemblies here in the states are marked clearly with an SAE marking that they meet specific standards. I have always found that American light assemblies seem to be much more reliable than the European produced products. Mercedes and BMW seem to have much greater light failure than I have ever seen on American cars. Whether that is because of the connectors or simply poor vibration transfer I don't know, but I often see these cars with non functioning lights, mostly the rear lights.

And I think that I was perhaps a dog before, but I don't remember that I like the taste of grass so I don't think I have been a cow yet! :D
 
Last edited:
Totally true. The other thing they do is to demand all sorts of trade secret info as a condition of importation; this is something I'm dealing with in my day job at the moment, not for the first time.

My experience in dealing with the Chinese government (Telecom ministry) was that if you wanted to import a product for sale in China, you would have to include a plan to manufacture that product, including technology transfer, in China. It seemed rather obvious to me that after that, the next step would be for them to export the products you taught them how to manfacture back to your home country, wiping out your domestic production. Hard to fault them because greed motivates corporations to willingly walk in to the Dragon's mouth.
 
Last edited:
Ed thanks for sharing.
Wide spreading of freq. peaks suggests edges on the waveform.
Studying the shape and location of distortion in the waveform may tell a lot.

The spectrum of the unloaded case indicated the magnetic core is saturated by the magnetizing (idle) current. Check the waveform for flat top and bottom peaks. Increasing ~10% the primary turns usually cures it.

3rd, 5th and 6th attached FFTs show quite a difference in loaded condition rel unloaded. I am not sure on this but I suspect close primary/secondary winding coupling and/or load is high for the trafo.

George

PS Test lead shown here form a large loop. This way your set-up is sensitive to other magnetic pick-ups in addition to the intended one (voltage across the chassis). Shielded separate leads may be better for the purpose.

1 Ouch!

George you made me have to think a bit. In my shop I have three phase Y 208 service. As I am in a residential neighborhood there are residences also served by the same mains distribution transformers. So where I should have 120/208 that would not work for them as 208 would be a bit low. So my 120 is actually 126 so they get 220 from 2 of the 3 phases. Now the allowable maximum line voltage is 130 (This keeps increasing as the years go on, to improve distribution efficiency.) or so my memory recalls.

Now the higher voltage is not a problem as my large equipment uses 3 phase motors and 208 is actually the lowest they want to see and are quite happy on 220 Y 3 phase. The electronic lighting ballasts all failed quickly in my office and replacing them with 120 to 220 input units solved that problem.

Now as I am using off the shelf commodity transformers typical of what I think folks use, there is the issue of primary voltage. Most of them actually are rated at 110 volt line input.

When I saw the harmonic spectrum, my OPINION was that I was seeing a flattened AC mains waveform from all of the rectifier loads both local (in my shop) the neighbors on the same distribution transformer and even from the high voltage distribution distortion. I thought the lack of higher harmonics indicated the attenuation with distance and power transformer filtering of the AC line distortion.

What got me thinking was that I would need to be sure that it was not my higher AC line voltage being limited by the transformers primaries. UNTIL I realized the R-core unit is being used on it's 220 volt primary to get secondary voltages as close to the rest as possible.

So in my opinion what we see is AC line distortion being passed by the transformers.

Now what was quite interesting was the abrupt rise in high frequency noise on some of the transformers under load and this is where I need to do a few more measurements.

Now I did do noise pass through and ripple harmonics. I use my relay buzzer on the AC line as a noise source. However I seem to have screwed up saving the files properly so I will need to repeat those.

The noise pass through was as expected. The R-core with a shield was best. The toroid was worst. All of the others were in the middle and very close to each other.

I will post the ripple data when I redo it as it was quite interesting.

As to the loop area by my test wiring, I did do runs moving it about. As I tried to make sure it was the same for all data run tests, my OPINION is that there is some error due to slight difference in loop area, but it is much less than the difference between transformers.

Now as to the leakage depending on orientation, that is why there are so many EI models. Pretty much toroids have the lowest magnetic leakage but the highest noise pass through, The R-core was low magnetic leakage, lowest noise pass through, but has some major problems yet to be explored and was of very poor quality overall and not suitable for use due to the 120 volt AC hot lead being green! (Dick, absolutely no legit UL sticker here!)

The classic EI cores proved to be in the middle, however the one with the extra plastic to meet international standards did seem to have other advantages. The dual bobbin flatpak was the worst for magnetic leakage and about average for noise pass through. One sample had secondary imbalances another did not?

I will try to do a run with the flatpak and try an external magnetic shield or two. But I am really straining to get the article finished!
 
Last edited:
Status
Not open for further replies.