AC128's from Czech - a review

[jkokura]

This post is in response to the thread found here: http://www.diystompboxes.com/smfforum/index.php?topic=87285.0 but you can read my explanation below and get the gist of it.

So one day, a long time ago, I decided I wanted to build a Fuzz Face clone with real AC128's. Knowing that those transistors are getting a little bit rare and expensive, I started my search. I stumbled across a guy in the Czech Republic on ebay who was selling what seemed to be AC128's at a reasonable price. Two pairs plus shipping to me would be a little less than 14 bucks, making it a 7 dollar per matched pair price. Seems good, but just to double check so I won't get hosed (like half the guys who get MN3005's...) I posted a question here to see if anyone knew of the seller and the quality of AC128's.

Long story short, I learned that there are many era's of AC128's, that these ones are modern ones, not the original coveted AC128's. These ones are a little bit more recent, but still high quality. What's more, the guy who's the seller on ebay pipes up, and then makes me a stupendous offer. Jan, a real standup guy, offers me a chance to try two pairs out without paying for them until I'm satisfied. I agreed, and informed him that while it may take a while, I was going to test the trannies and then post my results!

Today is the first of those posts, and it's been a long time coming. Here's my results...

Jan sent the transistors out promptly, and they came neatly and quicker than I was expecting for them to travel from the Czech Republic to Vancouver Island in Canada. I will note that while english is obviously not Jan's first language, he's easy to communicate with, has great customer service, and the fact that he's willing to allow this sort of public testing (with out payment yet) is pretty awesome. Kudos to him.

The first thing I noticed was that the transistors came in some clearly marked wax paper. One package was marked "Q1" and the other marked "Q2". Opening up the package marked Q1, a pair of smaller wax paper envelopes fell out, each marked with a little label and some readings Jahn took. Each clearly indicated the HFE and some other data. Neat. Each of the Transistors had lots of LONG lead, and they were clearly marked with AC128. Comparing them to the link of the AC128 historical pictures, they seem to be exactly like the Modern ones we expected to them to be in the other thread. Apparently those aren't quite as precious, but still the real thing and close to the original in sound.







After matching the pairs up, I tried to verify the HFE readings using my Multimeter. This is the only complaint I had about the transistors: I could not replicate the HFE readings. I'm not sure if there was something I was doing that was wrong, but no matter how I tried, the HFE readings Jan got eluded me. In fact, I could not get any reliable readings. The reason I couldn't get any readings could be my multimeter being crap for this purpose, or from user error of course, but I've been able to reliably measure the HFE of other transistors in the past, so the fact that I couldn't get a reading off of these transistors seemed strange to me. So from this issue, I would need either someone else to verify that they too couldn't get HFE readings, or that they could for us to really verify that these transistors were suspect.

The first pedal I put the AC128's into was a Fuzz Factory clone. I used Metal film resistors, Box film and typical Aluminum Electro caps on a fabbed board from Madbean. 9mm pots, and the only mod to the project was the suggested 5kC pot for the Stab function. The neat thing about the Fuzz Factory is that it can sound like a fuzz face if you want it to, or it can sound like a glitched out Computer about to explode! In my fuzz factory, the transistors sounded great. Dialed to a Fuzz Face sound, I liked the easy, creamy fuzz I got. Dialing back my volume, the fuzz factory clone cleaned up nicely. Playing with the knobs got me into all sorts of crazy sounds. The best part about it, at least from my point of view, is that I was easily able to replicate the kinds of sounds I have heard from demos online. The only way I could test this better would be to take a real fuzz factory and AB them, but I don't have that capacity right now.



The second pedal I tried the AC128's in was a mojo Fuzz Face clone from Madbean also. I used Carbon Comp resistors and axial film and AE caps. I even tried to nail the look of a fuzz face, using my typical method of decal over a red textured powder coated 125B. This pedal was wired for positive ground (By the way, I hate wiring for positive ground. Wiring this pedal was too much of a headache for me to do it again.). The great thing about this fuzz face is that it's very similar in construction to half a dozen fuzz face clones that are available out there already, including the Hartman Germanium Fuzz that I own. A video comparison of the Hartman Fuzz and this Fuzz Face will be coming, and I'll post that video later. The AC128's performed well in this. After biasing it, I was easily able to replicate a lot of the fuzz tones I'm familiar with, from Hendrix type up to more of an EJ lead tone. I was able to dial back my volume and clean the sound up nicely.



All in all, even though I wasn't able to replicate Jan's HFE readings, I'm very satisfied with the sound I'm getting from these transistors. They add a little 'mojo' because they are, from all apparent sources, authentic AC128's that people covet in Fuzz Faces today. They also sound good, which makes me think they aren't cheap knock offs, and they're worth the money. I could probably find cheaper transistors that would sound good, but that AC128 sound is what I was wanting, and I feel I'm getting it.

The only way I can think I could test these some more would be to find an Authentic Fuzz Face and Fuzz Factory with AC128's in them and AB them. If anyone would like to loan me either of those kinds of pedals (insured of course) to do an AB test, or would like me to loan them my pedal for an AB test for the community that would be appreciated.

As stated, videos will be forthcoming, thanks for reading! Jan, you can send the invoice to jacobkokura@gmail.com

Jacob

[PRR]

> I could not replicate the HFE readings

What readings did you get? Zero? A million? TILT? Different every time you blinked?

Could the leads be tarnished and making poor contact in your hFE socket? IIRC, that wire was thinner than most modern leads. (Trim some legs off a 2N2222 or similar, tack-solder to the end of these leads.)

That said: I accept that you have read other transistors so these are different.

His test-conditions are unusual. They may be specific to the intended use (first transistor in a direct coupled pair). We normally take both Ico and hFE at several volts. However the first transistor in these circuits runs with slightly over Vbe of the second transistor, indeed 0.3V-1.0V, so a 0.7V test is appropriate for the labeled use ("Q1 FuzzFace").

You could try building a test-rig specifically to duplicate his stated conditions. Would you? Do you have two healthy 9V batts and two battery-powered multimeters? (One could be cheap or wall-power, the other should be at least a $3.99 DVM or equivalent/better without wall-power).

OR- in-circuit voltages and resistor values will reveal the hFE in actual operation, though for the classic FF this won't be 10mA.

[tubelectron]

Hi,

I could not replicate the HFE readings

PRR is right : you don't have the appropriate instrument. Measuring Ge Qs implies loss measurements to be taken in count, otherwise your readings are usually inconclusive. I have built some years ago a test rig as suggested by PRR - despite its appearance, it is a simple instrument, with nothing new from the fully available classical test schematics for transistors. You can built something similar, even cheaper and simpler, for your needs.




A+!

[jkokura]

Thanks for the suggestions and encouragement guys! I didn't realize GE trannies needed special equipment to measure their HFE. As for the suggestion of spending the time and money to do so, I think I'm going to trust Jan's readings on the Trannies, and I'll take some time to look up how to read GE HFE for the future. As it is, I won't be buy any GE trannies in bulk so that I need to read them in the near future, so spending the time and money isn't really justified for me right now. If I can find a way to do it, I will, but for now I'll just assume that Jan's readings are good to use because these transistors seem to sound good to my ears.

Jacob

[Brymus]

You dont need special equipment.
Did you try R.G.'s method from GEO ?
Or even the Small Bear version will get you an idea of the Hfe and leakage.

When I ordered some Russian ge's I tried both ways and actually like RG's a little better.
I even found a 9v battery that actually put out 9V,and sorted through resistors to get the exact values he stated.
But when sorting through 100 or so transistors its worth it to know roughly which ones have approx X gain and Y leakage.
I labeled all mine with the results and sorted them by such.

It does sound like you got a nice pair of 128's, It might be worth your effort to compare them with some of higher gain in Q2.

[zombiwoof]

What were the marked gains of the two Q2 transistors you got?.  Only the Q1 trannies are pictured.

Al