Kalibr L1-3 Test Parameters

Brand name believed to be KALIBR.
Post Reply
User avatar
Allan
Posts: 17
Joined: Wed Apr 04, 2012 3:54 am
Location: Boston

Kalibr L1-3 Test Parameters

Post by Allan »

Hello Everyone.

My amplifier circuits that I design with and play guitar through use EL34 pentodes. Like many of us, I have experimented with different brands and series of tubes to find what works well in the circuits and sounds good. For myself,this started with using tubes from a very popular company that does testing, distortion matching and even designing tubes. I had found that their GT-EL34 types with a performance number of 6 or 7 work well in my amps.

Although their distortion matching process helps, it is the current draw of these that I have used to purchase other brands of EL34. The EL34s that I like to use in my amplifiers draw between 100mA to 106mA at the plate when tested in pentode mode by the old Mullard/Philips datasheet. This is similar to the current draw of my number 6 or 7 GTs. So it is the same or slightly higher than the Mullard bogey tube.

I can also tell you that these favorites draw between 32mA and 37mA (some as high as 42mA) on the Maxi-Matcher (400V triode mode, -36V bias). But not too many use these in Europe and Russia. With all the beautiful testers there, I understand why.

My question to you is, will these Russian testers use the old Mullard test method? Do I simply ask for a quad matched at 104mA? Or do these test at other voltage parameters or in triode mode?

Thanks.
-Allan

User avatar
Admin19
Site Admin
Posts: 75
Joined: Sun Dec 11, 2011 1:55 pm

Re: Kalibr L1-3 Test Parameters

Post by Admin19 »

Hello Allen,

the best tube matching is when the curves of two tubes are identical. The Sofia tube tester has an algorithm for that. Alternatively you can just look at the curves, as you do with a Tektronix curve tracer, and switch from one tube to the other with a self made fixture. A curve tracer can verify the matching, but is very impractical to get 50 pairs out of 100 tubes. That drives you crazy, and no normal person can do a good job that way. The Amplitrex AT1000 has all the hardware for this, but the machine has primitive software, and this feature is painfully missing. So apart from Sofia, which almost nobody has, you always will be getting something near, but never exact.

However there is this practical approach that a great part of the tube curves "DNA", is present in one tube curve only. It turns out if a representative curve of a good and new tube is the same with two tubes, very likely the other curves will be the same also. In theory it doesn't have to be that way, but in practice it is, IF the tubes are new, and good quality to begin with.

Now most people won't have a curve tracer, but you can even simplify this method further. For this you measure the tube at one representive bias point only. Since now screen and plate voltage are a fixed voltage, you are measuring similar to triode mode. (That was your question...) This point for EL34 is useually like 100mA / 250V. If, at that bias point, Vg1 an Gm are the same, quite likely the whole curve at that particular Vg1 is the same, because Vg1 fixes the curve position and Gm is the angle. This will only go wrong if one of the two has not enough emission (problems at high current), or has no clear cut off (problems at low current). So this works only for tubes that are "good" to begin with. If all of that applies, we can assume all curves are the same. The practical situation is, that this applies quite nicely for tubes that are NEW, and GOOD. Yet we must say, this can not replace real curve matching, but in real life this method is the most often used.

The L1-3 or L3-3 are very often used in Russia, they are truely excellent testers. Most amazing is, there can be almost nothing wrong with them. If such a tester appears to work good, that is useually what it is. I would recommend set the Plate and Screen both for 250V, and the plate current for 100mA. Two identical tubes must now have identical grid voltage, and identical Gm. Given Vg1 and Gm are not further off than +/-30% from a bogey tube, you are looking at "good" tubes, and so you have a good match. This is in a nutshell how it works, but doing a good job means having LOTS of experience, having good equipment, and be prepared to remove "not so good" tubes from the lot, and that can hurt. If your Russian supplier can handle the L3-3 well, and is honest with you, matching can be very reliable.

Since this assures the whole curve chart to be the same, it doesn't matter much if you do this at 100mA or 106mA. Just ask for 100mA. (That was your question...)

Hope this answers your questions!

Best regards,
Jac

User avatar
Allan
Posts: 17
Joined: Wed Apr 04, 2012 3:54 am
Location: Boston

Re: Kalibr L1-3 Test Parameters

Post by Allan »

Jac, thank you for the information. Yes you did answer my question.

New question: Suppose the plate and screen are both set for 250V and the bias is set for -13.5V. When the L1-3 tests for current in this manner, is the plate current displayed on the meter? Or is the current measurement taken from the cathode (sum of the plate current + screen grid current)?

Thank you again.

-Allan

User avatar
Admin19
Site Admin
Posts: 75
Joined: Sun Dec 11, 2011 1:55 pm

Re: Kalibr L1-3 Test Parameters

Post by Admin19 »

the L1-3 is a highly professional product. When it tests plate current, it is for plate current only. This tester works totally straight forward. There is one stabilsed power supply for the Anode, and one for the screen. You can measure current draw on each of them.

User avatar
Allan
Posts: 17
Joined: Wed Apr 04, 2012 3:54 am
Location: Boston

Re: Kalibr L1-3 Test Parameters

Post by Allan »

This is good information to know. I would love to purchase an L3-3 that you have described. With shipping to the States, they cost $1400 USD or half the price of an Amplitrex. Maybe someday.

Three months ago I purchased a new MaxiMatcher. Because I am not a professional, it took me years to make the decision to finally buy it. Although it will not test according to the tube manuals, it should allow very good matching at amplifier voltages. At least it should be much better than my DIY mA and voltage tester. Once the holidays are over I am going to get my workshop in order and test all of my tubes. Then I will calibrate an old Hickok 600A that sits in my basement.

It seems that a professional or serious hobbiest should have a tester that works at tube manual specifications and a tester that works at real amplfier higher voltage parameters... ...or get an Amplitrex! Having the MaxiMatcher for high voltages, I suppose the L3-3 would be the next logical tester for me.

I cannot believe that in this century we have a very good selection of equipment to test our tubes. Life is good.

User avatar
Admin19
Site Admin
Posts: 75
Joined: Sun Dec 11, 2011 1:55 pm

Re: Kalibr L1-3 Test Parameters

Post by Admin19 »

The L3-3 was around 300...450 Euro for years. They went up lately as more poeple start to realise what this is. I see them go for 500...600 Euro now. They're well worth it. It can do maximum 300V. There is almost no audio tube that "MUST" have more. They can do, but there is no "must" for a good test. Even so, low voltage and high current is more difficult than the other way around. That is why KT88 and EL34 are factory tested at 250V plate and screen, whereas you use them often at 450V inm equipment.

The Amplitrex is not as good as the L3-3. It's just digital, and it makes curves. However precision of the L3-3 is at the most 2% and you can't say that from the Amplitrex. If you test 6SN7 on an Amplitrex, it gives you gm in steps of 100. So 2500, 2600, 2700. This is 4% read error to which adds the measurement error which is similar. Also the Amplitrex will loose it's value as one day you won't be able to install the program on newer future windows versions. So in they year 2025, perhaps Windows "25" will no longer install this old software, or whatever problems they make up for us at Microsoft. They have a great tradition for refusing to install old software. Frankly, I would say any software updates from Amplitrex... forget it. Incompatibility is the reason why the 1995 digital SOFIA tester is so hard to use now, it runs only on an older computers. Then you take your L3-3, and it still works, and it's price has gone up.

User avatar
Allan
Posts: 17
Joined: Wed Apr 04, 2012 3:54 am
Location: Boston

Re: Kalibr L1-3 Test Parameters

Post by Allan »

500-600 Euro, is that for a used L3-3? The $1400.00 USD price (1070,00 Euro) I mentioned is for an NOS L3-3 tester shipped from Ukraine to the US. The shipping alone is $380.00 USD (290,00 Euro), ouch! It is amazing though that there are NOS old production testers still available.

I do not think I will ever get an Amplitrex. If my business ever takes off, then maybe. Instead I would like to build a preamp tube tester eventually, a stand alone unit. This would probably be my next necessity.

Old software can be a problem. I will admit that one of my computers still runs a highly optimized version of Windows ME! I removed all of the bloat-ware and it runs faster than my XP and Win 7 computers... ...at least for what I use it for. All of my electronics files (tubes, testers, hi-fi, guitar amps, etc.) and automotive stuff is on it. It's my stripped-down technical information computer. It does well with the Internet too.

User avatar
Admin19
Site Admin
Posts: 75
Joined: Sun Dec 11, 2011 1:55 pm

Re: Kalibr L1-3 Test Parameters

Post by Admin19 »

Hello Allen,

for an L3-3 you could try with Roger Kenney from alltubetesters.com. I know for sure he has one, and also he did not do much with it as he couldn't get it going.

Best regards,
Jac

User avatar
Allan
Posts: 17
Joined: Wed Apr 04, 2012 3:54 am
Location: Boston

Re: Kalibr L1-3 Test Parameters

Post by Allan »

Hi Jac.

If Roger Kennedy can't fix it, not sure how successful I would be with it. But it is worth a try. Thank you.

-Allan

Post Reply