Your examination is as always precise and meticulous ... from the first chapter onwards everything is compliant and demonstrates the ineffectiveness of the project ... I do not explain, however, why a 230/170 transformer instead gives 238/165 and that there are only 209v after the bridge ... these data distort the rest of the project and do not even give the possibility to do something similar to the Armstrong ...
Last edited:
At this point I would check the EL84 emission, in a tube tester or by by comparing it with a known good specimen. To get the most out of the output transformers, best course of action would be to replace the power transformer or select another tube. EL84 requires at least 250V. ECL82 would be a better choice at 200V.
My guess is that the power transformer is overloaded. It is unable to supply the current spikes required to charge the filter capacitors due to core saturation and winding resistence.
Don't forget to take voltage regulation into according which is usually 10%, more for smaller VA transformer due to voltage drop caused by resistance of internal wire size. So 100*((238-209)/238)=12% or 29V from 0 to 76mA.
Last edited:
Cathode is 5.27V in my sim. Btw what brand 6au6 and el84 you're using? For completeness, I sim the entire circuit, you can verify against your measurement quite quickly and if you use LTspice and maybe a scope is necessary for making alignments.
Attachments
Tubes seem to be ok on my tube checker. I'm using Siemens 6AU6 and Telam EL84 already tested on another amplifier.
ECL82 were plan A also because of the very compact size of the box but they ask 5k6 OT and, at this point, I don't know if this chinoiserie of transformer can drive them.. wouldn't there be the same problem that Koonw just mentioned?
Last edited:
Are there other taps on the secondary of your OPT's than the 4 Ohm you show in your schematics?
Like pcan pointed out already, 2 x ECL82 would do fine with your power transformer. The only problem is the impedance of the primary of your OPT's. For a supply voltage of about 200 V, you would need the primary to be between 3K9 and 5K6 (see this Philips datasheet for the ECL82: https://frank.pocnet.net/sheets/030/e/ECL82.pdf ).
If your OPT's would also have a tap for 8 Ohm, you could run them with loudspeakers of 4 Ohm. In that way, the impedance at the primary becomes 3K5, which in real life is close enough to the 3K9 in the datasheet.
If your OPT's only have 4 Ohm at the secondary, you could connect two loudspeakers of 4 Ohm paralled per channel (so four speakers in total). In that way you will also get an impedance of 3K5 at the primary.
Addition: And with an 8 Ohm and a 4 Ohm speaker paralled per channel, the primary impedance even comes into the range of 3K9 to 5K6.
Like pcan pointed out already, 2 x ECL82 would do fine with your power transformer. The only problem is the impedance of the primary of your OPT's. For a supply voltage of about 200 V, you would need the primary to be between 3K9 and 5K6 (see this Philips datasheet for the ECL82: https://frank.pocnet.net/sheets/030/e/ECL82.pdf ).
If your OPT's would also have a tap for 8 Ohm, you could run them with loudspeakers of 4 Ohm. In that way, the impedance at the primary becomes 3K5, which in real life is close enough to the 3K9 in the datasheet.
If your OPT's only have 4 Ohm at the secondary, you could connect two loudspeakers of 4 Ohm paralled per channel (so four speakers in total). In that way you will also get an impedance of 3K5 at the primary.
Addition: And with an 8 Ohm and a 4 Ohm speaker paralled per channel, the primary impedance even comes into the range of 3K9 to 5K6.
Last edited:
My friend has very normal 8 ohm speakers ... the problem is that these little transformers are almost all set to operate at 4 ohm or even less. I have another couple that have a turn ratio of 35 ... I could reformulate the project if I can get another power transformer that is able to give me the 250v required by the EL84 ... I still listen to the 6Y6 SE with some 16 ohm speakers connected to the 8 ohm terminal without particular problems.
7K is not a issue. ECL82 Philips datasheet (1960) lists several recomended configurations. The "low distortion" one does have 8K load and 650 ohm cathode resistor at 272V anode supply. Your supply is lower and the output transformer DCR is high, so you will get less power compared to the Philips datasheet. I guess that you may get between 1W and 2W at 5% distortion with 8 ohm speakers; at normal listening volume below 1W it should sound just fine. A little bit of negative feedback will increase the useful power.
With a supply voltage of about 200 V the impedance of the primary of 7K in combination with an ECL82 is an issue. For power pentodes the optimal impedance of the primary = Vak/Ia.
On top of that we now learned that TS and/or his friend, for which TS is making the amplifier, has no 4 Ohm loudspeakers. With 8 Ohm speakers connected to the 4 Ohm taps the primary impedance will be doubled, so will be 14K. That's not good...
On top of that we now learned that TS and/or his friend, for which TS is making the amplifier, has no 4 Ohm loudspeakers. With 8 Ohm speakers connected to the 4 Ohm taps the primary impedance will be doubled, so will be 14K. That's not good...
I agree that the result will be sub-optimal, altough I still believe that it could be somewhat acceptable. Repurposing in a meaningful way reclaimed small output and power transformers is a challenge; I have a box full of them, and I haven't found a good solution yet. They were usually paired with EL84 or ECL86 tubes and small 3 ohm speakers, but today we have 8 ohm speakers with far more bandwith. ECL80, EL95 and 6AK6 seems to be the best tubes for this job. They also require less current and will not overload a small power transformer. The drawback is that they will never be manufactured again, while ECL82 and EL84 are still being produced.
1w is not enough for medium sensitivity speakers ... with this power transformer and 7000 OT the deck is always short ... today they should bring me a 230v transformer (I think of Philips derivation) and a selenium B250C125.
Bandwidth is not an issue because he has small loudspeakers and has no high-end ambitions...
Perhaps the best solution would be a pair of ECL86... or run 2 EL84s at 250v and 35-40mA...
Bandwidth is not an issue because he has small loudspeakers and has no high-end ambitions...
Perhaps the best solution would be a pair of ECL86... or run 2 EL84s at 250v and 35-40mA...
Yes; but this power transformer is in-hand.
We may be seeing different data? I see 204V plate-cathode, 40mA total cathode current, so probably 35mA plate current. 7 Watts dissipation (half of EL84's rating). "Happy Load" about 204V/0.035A or 5.8k. Likely output maybe 2.8 Watts sine RMS at 5%-10% THD.
A large home radio.
The noted 165V AC should make 233V of DC, before sag. I don't think 209V raw DC is suspicious. But there may not be a lot more juice in this barrel.
I normally see ~1.25X the available HV at the DC OP of cap input filters. Some folks load up the first cap to high values, not the best of ideas. It is the RMS current that cooks PS transformers, not the average current. This is a Class A amp, PS regulation should not be a major problem. My 2 Centavos, anyway.🙂
The data are those measured ... on the other hand, after 209v on first cap, the measurements follow quite faithfully what it should be, so the issue is upstream ... if the result is to be 2w and not even of good quality, it is better to abandon ship before it sinks and think of another solution. Yesterday I received the transformer in the photo ... from the size I think it may be enough to run 100 mA or so. In the meantime, I've to decide whether to leave the project similar, however, with HT 250v or change it by thinking of 2xECL86 which also have the advantage of working at 7k load and therefore closer to the 10k / 8ohm required ... the second option, however, would in practice mean starting over almost from zero...
Attachments
Try using a 6bk5 as a power tube with 4 ohm car speakers.
Should be a good match for your transformers. 6bk5's are cheap!
Should be a good match for your transformers. 6bk5's are cheap!
I don't see how a 6BK5 instead of an EL84 would improve things for TS.
The optimal primary load for a power pentode is Vak/Ia. This applies for both the 6BK5 and the EL84. So the problem stays the same. B+ is too low.
For my idea, best would be (like others already pointed out before in this thread) to get an other power transformer (higher voltage).
The optimal primary load for a power pentode is Vak/Ia. This applies for both the 6BK5 and the EL84. So the problem stays the same. B+ is too low.
For my idea, best would be (like others already pointed out before in this thread) to get an other power transformer (higher voltage).
