I notice another problem in the heater smps power supply.
the heater smps takes about 15 to 30 sec to power up and reach 6.2 v after switching on the main power .
The HT dc is tranformer based with CRCR filtering. it is about 370 v dc.
In other words, the HT is applied to the tubes much earlier than heater power.
Is this an issue that need to be rectified ??
the heater smps takes about 15 to 30 sec to power up and reach 6.2 v after switching on the main power .
The HT dc is tranformer based with CRCR filtering. it is about 370 v dc.
In other words, the HT is applied to the tubes much earlier than heater power.
Is this an issue that need to be rectified ??
Output Trannies: For EL84 (Not EL34)
Any supplier of Dynaco ST-35 replacement transformers.
The Output Tranny was the Dynaco A-410.
Triode Electronics or Parks Audio (DIYTube) would be good places to look for equivalents.
Else
Hammond 1608 or 1609.
Also check Edcor - they have a number of options of various quality from "OK for a Git Amp" to high quality HiFi.
Edcor, Hammond and Triode Electronics would no doubt have power tyrannies to suit as well.
Cheers,
Ian
Any supplier of Dynaco ST-35 replacement transformers.
The Output Tranny was the Dynaco A-410.
Triode Electronics or Parks Audio (DIYTube) would be good places to look for equivalents.
Else
Hammond 1608 or 1609.
Also check Edcor - they have a number of options of various quality from "OK for a Git Amp" to high quality HiFi.
Edcor, Hammond and Triode Electronics would no doubt have power tyrannies to suit as well.
Cheers,
Ian
Last edited:
An update on my BH build. I’ve been busy and have not had time to finish it. Today I did some more testing with some odd results. I’ve attached a couple of photos.
I had 2 pairs of tungsol el34 for the BH. With one pair I get .1% THD at 1w 4ohms and the other pair is .38%. By the time you get to 15 watts the distortion is the same for both pairs at about .5%.
I was running the el34s at 40ma and 450volts into a 6600ohm Toroidy transformer.
I decided to try KT88s. I built in a voltage adjust switch in the regulated HV supply to allow for 400, 425 and 450v settings. I reduced B+ to 425 and biased up 2 pairs of Sovtek KT88 ( the coke bottle shaped ones)
Much better distortion performance than the EL34 and consistent for both pairs. 1w 4ohms .025% distortion. .09% at 4watts and .38% at 16watts. Tested the LF bandwidth and it was flat at 20hz. At 1w into 4ohms 20hz distortion was only .11%. Max power looks like just over 40w. I am using the 4 ohm tap on the transformer to match up with my current speakers which have a minimum impedance of 3.6 ohms.
It seems the KT88 likes this circuit better than the EL34.
Output impedance was about .8 ohms.
My el34s were ordered as a matched pair from parts Connexion. I see they also offer a super matching service at an additional charge. I think what I am seeing with the differences between the distortion for the el34 pairs is that the matching is not as good as it is with the kt88s, which I purchased a long time ago from the tubestore.com.
I have not listened to it yet. I still want to address some peaking in the HF response. About +1db at 200khz. Yes, 200khz! These Toroidy transformers offer impressive performance for the $ IMO.
I also need to build a cover for the regulators and a wood frame for the chassis.
I built it with a pivoting top so I could access it easily. The 2 transformers you see in the front of the open chassis are there temporarily for weight. It is a bit light in the front with the cover all the way up.
I had 2 pairs of tungsol el34 for the BH. With one pair I get .1% THD at 1w 4ohms and the other pair is .38%. By the time you get to 15 watts the distortion is the same for both pairs at about .5%.
I was running the el34s at 40ma and 450volts into a 6600ohm Toroidy transformer.
I decided to try KT88s. I built in a voltage adjust switch in the regulated HV supply to allow for 400, 425 and 450v settings. I reduced B+ to 425 and biased up 2 pairs of Sovtek KT88 ( the coke bottle shaped ones)
Much better distortion performance than the EL34 and consistent for both pairs. 1w 4ohms .025% distortion. .09% at 4watts and .38% at 16watts. Tested the LF bandwidth and it was flat at 20hz. At 1w into 4ohms 20hz distortion was only .11%. Max power looks like just over 40w. I am using the 4 ohm tap on the transformer to match up with my current speakers which have a minimum impedance of 3.6 ohms.
It seems the KT88 likes this circuit better than the EL34.
Output impedance was about .8 ohms.
My el34s were ordered as a matched pair from parts Connexion. I see they also offer a super matching service at an additional charge. I think what I am seeing with the differences between the distortion for the el34 pairs is that the matching is not as good as it is with the kt88s, which I purchased a long time ago from the tubestore.com.
I have not listened to it yet. I still want to address some peaking in the HF response. About +1db at 200khz. Yes, 200khz! These Toroidy transformers offer impressive performance for the $ IMO.
I also need to build a cover for the regulators and a wood frame for the chassis.
I built it with a pivoting top so I could access it easily. The 2 transformers you see in the front of the open chassis are there temporarily for weight. It is a bit light in the front with the cover all the way up.
Attachments
Last edited:
Bfpca,
Thanks for sharing your observations and measurements. Great build and arrangement. I shall appreciate a few clarifications:
1. I assume your distortion measurements are for one PCB switching the output tubes. Did you get similar results for the other PCB, i.e. you did not measure differences originating in the components of the PCBs?
2. The EL34s were biased to 40ma with 450 VDC to output transformer; what was the current for the KT88s at 425 VDC? Would you kindly measure what the negative bias voltage is at this operating point? (I’m concerned my negative supply might not be negative enough). It seems to be a good operating point for KT88s with 6.6k OT and I might target it for my own build, since I don’t need more power from a 4k OT. The KT88s are loafing along and should last a long time here.
3. Did you apply any GNF when you measured the output impedance?
4. Wow, +1dB at 200k! If that tiny bump is your only concern those Toroidys are really good. Hoop you like the way all will sound. Let us know when you had time to listen.
Just to quickly answer questions. The adjustments are available through the top of the chassis. The meter on the left front reads the cathode currents of all the power tubes and has 2 ranges 50ma and 100ma
1. I found the differences in el34 distortion followed the power tube pair from channel to channel, so it's tube related.
2. Running at 60ma or slightly higher on KT88. Bias is 46-47 volts. In order to get undistorted full output I increased - bias voltage to -98.5. If your bias supply is not high enough you will see a small glitch in the sine wave which occurs when the - side of the grid waveform clips on the power tube that is being cutoff.
3. About 6db of global feedback. It is applied to the 8 ohm winding which i also have terminated with a 100 ohm resistor as a minimum load. Your overall voltage gain is very low in this configuration. Less than 4 to the 4 ohm winding. If I were doing it again I would order a lower impedance transformer which would allow 40w out with lower b+ and bias voltages.
4. They are very good IMO. Also flat to 20hz with low distortion. High inductance compared to most ei output transformers supports bass with less distortion.
Thanks for your answers - good information. Since I have not yet ordered my Toroidy OTs I will now plan to get the 4k/80VA ones.
I had planned to get the 4K but made a mistake when ordering. I ordered a set of the 8k for EL84 and a set of 6.6 by mistake. I didn’t realize it until after they had shipped. So I have had to push B+ and bias voltage levels higher to get good power. I imagine that KT88 will put out more power into lower impedances than the nominal tap rating. The next time I power up, I’ll check that and report back.
You wil also get more gain using the 4K vs 6600. For instance with the 6600 OT and 6db of feedback it takes about 3.5v rms for full power. It should be somewhere around 2.7v for 4K with the same amount of Feedback. I am using the stock 47k/39k resistors in the internal feedback between the input and output stages.
Good luck on your build! If you have any more questions, just ask.
Last edited:
I couldn’t resist having a listen. Initial impressions are .....very nice! It sounded a bit thin in the upper mids and a little bit loose in the bass compared to the Aleph J I have been using lately. However, it seemed more 3 dimensional and real in some ways. Vocals were really well rendered. This is a strength of the Aleph J as well. Too early to make any conclusions for sure. I need to get that chassis finished and make a few tweaks to the operating point etc.
My system is in a state of flux lately after getting a new pair of speakers. Monitor Audio Gold 300 (4th generation) which are much better than anything I have built, but are also very revealing of any shortcomings in your system. My Aikido preamp laid an egg on me and needs repair. I think a bad connection at a tube socket. So I am using an Emotiva DC1 dac as a source and preamp. It may be a while before I can get a stable system for testing new components.
My system is in a state of flux lately after getting a new pair of speakers. Monitor Audio Gold 300 (4th generation) which are much better than anything I have built, but are also very revealing of any shortcomings in your system. My Aikido preamp laid an egg on me and needs repair. I think a bad connection at a tube socket. So I am using an Emotiva DC1 dac as a source and preamp. It may be a while before I can get a stable system for testing new components.
Most Beam pentode (or tétrode) like the 6550 by exemple ( JJ Electronic - 6550 ) have no connection on pin 1 
In fact only the EL34 or 6CA7 which are pentode have a separate connection for G3 and it is for that reason that pin 1 and 8 are connected on the PCB
But check the voltage, only 250 V !
Marc
In fact only the EL34 or 6CA7 which are pentode have a separate connection for G3 and it is for that reason that pin 1 and 8 are connected on the PCB
But check the voltage, only 250 V !
Marc
Cathode Feedback output transformers for Baby Huey (BH)
I was looking at the Toroidy website/catalog of output transformers and noticed their new Cathode Feedback (CFB) offerings. Among others they are available in 8k and 5k impedances (80 watt ouput) and I’m tempted to get a pair each for my build of the EL84 and EL34/KT88 Babies. I have never used (or studied) CFB output transformers and have a lot of questions.
Has anyone tried these CF Toroidy transformers? Or CFB on BH amps? Any reason why one should not use CFB on the BH? Could there be undesirable interaction between the CFB and the anode feedback scheme used on the BH design?
The 8k CFB transformer is capable of 80 watt output, which is overkill for the EL84, but would it be damaging to the performance of the BH to use such an oversized transformer on the EL84 BH?
I was looking at the Toroidy website/catalog of output transformers and noticed their new Cathode Feedback (CFB) offerings. Among others they are available in 8k and 5k impedances (80 watt ouput) and I’m tempted to get a pair each for my build of the EL84 and EL34/KT88 Babies. I have never used (or studied) CFB output transformers and have a lot of questions.
Has anyone tried these CF Toroidy transformers? Or CFB on BH amps? Any reason why one should not use CFB on the BH? Could there be undesirable interaction between the CFB and the anode feedback scheme used on the BH design?
The 8k CFB transformer is capable of 80 watt output, which is overkill for the EL84, but would it be damaging to the performance of the BH to use such an oversized transformer on the EL84 BH?
Francois- I am not familiar with the details of cathode feedback. I believe it was used, and perhaps still is used, by McIntosh.
The first thing that comes to mind is that with any type of feedback you are reducing the overall gain. IME the BH being a 2 stage amplifier does not have a lot of extra gain, at least in the EL34 version. The situation will be better for EL84 but it is something to consider.
How will the feedback loops interact? That is something that could only be answered by someone with experience using the 2 types of feedback. Whenever you get more than one feedback path the resulting response can be more difficult to compensate. You could always leave out the crossfed feedback between the driver and the output stage in the BH. Simply leave out the cross feed resistor and series resistors to the output transformer and bring B+ to the top of the driver tube plate resistor. If you are going to do this you need to study the drive requirements for EL84 when driven in cathode feedback mode. Since the existing driver can swing a lot of voltage when driving el34 or KT88 you may have enough drive voltage. Again, you will then need to look at the overall gain to see if it is reasonable for your intended purpose.
For instance in my situation with KT88, a 6600ohm OT, and 6db of overall feedback I have an input sensitivity of 3.5vrms for roughly 40 watts out. The gain is less than 4 at the 4ohm tap. If I had wanted to use the BH as an integrated amp, that is not really practical and even with a moderate gain preamp the volume control is up past half way for even moderate levels.
When I build the EL84 version I am hoping for a more reasonable amount of gain.
The first thing that comes to mind is that with any type of feedback you are reducing the overall gain. IME the BH being a 2 stage amplifier does not have a lot of extra gain, at least in the EL34 version. The situation will be better for EL84 but it is something to consider.
How will the feedback loops interact? That is something that could only be answered by someone with experience using the 2 types of feedback. Whenever you get more than one feedback path the resulting response can be more difficult to compensate. You could always leave out the crossfed feedback between the driver and the output stage in the BH. Simply leave out the cross feed resistor and series resistors to the output transformer and bring B+ to the top of the driver tube plate resistor. If you are going to do this you need to study the drive requirements for EL84 when driven in cathode feedback mode. Since the existing driver can swing a lot of voltage when driving el34 or KT88 you may have enough drive voltage. Again, you will then need to look at the overall gain to see if it is reasonable for your intended purpose.
For instance in my situation with KT88, a 6600ohm OT, and 6db of overall feedback I have an input sensitivity of 3.5vrms for roughly 40 watts out. The gain is less than 4 at the 4ohm tap. If I had wanted to use the BH as an integrated amp, that is not really practical and even with a moderate gain preamp the volume control is up past half way for even moderate levels.
When I build the EL84 version I am hoping for a more reasonable amount of gain.
Hi Francois,
I agree with Bfpca, I think that using a CFB output transformer for the Baby Huey would an overkill... And as I have been very interested since many years by the very nice design of Norman L. Koren of a quad 6550 amplifier with a CFB toroidal output transformer : The Emperor's New Amplifier I wanted to make one...
After my good experience with the Baby Huey, I wanted to build somethings more powerful with two mono blocks chassis, a little like what Norman did in 1997, but including some of the features of the Baby Huey like the shunt feedback, a differential input stage and the replacement of the driver tubes by a MOSFET stage
I started the project last year, I have already bought the Toroidy transformers ( TTG-CFB2000PP - Tube output CFB transformer [2kOhm] Cathode Feedback Push-pull - Shop Toroidy.pl ) with the very nice SUPREME finition ( see photo ) and the power supply transformer with the same finish
However, I have been busy with many other things like building solid state amplifiers to compare them with the Baby Huey, and I wanted to have more experience before to start such a high end project... As Norman said : "He was so satisfied with his amplifier that he never designed an other one..."
I am planing to use KT120 output tubes since they have very good reviews and because it is difficult to find good 6550 today, I had very good ones from Svetlana but they are no more produced
I put the schematic and the PCB in this message but it is still a work in progress therefor all remarks and/or suggestions are welcome... For info I made the PCB with the serigraph on the bottom side to facilitate the building
Rgds,
Marc
I agree with Bfpca, I think that using a CFB output transformer for the Baby Huey would an overkill... And as I have been very interested since many years by the very nice design of Norman L. Koren of a quad 6550 amplifier with a CFB toroidal output transformer : The Emperor's New Amplifier I wanted to make one...
After my good experience with the Baby Huey, I wanted to build somethings more powerful with two mono blocks chassis, a little like what Norman did in 1997, but including some of the features of the Baby Huey like the shunt feedback, a differential input stage and the replacement of the driver tubes by a MOSFET stage
I started the project last year, I have already bought the Toroidy transformers ( TTG-CFB2000PP - Tube output CFB transformer [2kOhm] Cathode Feedback Push-pull - Shop Toroidy.pl ) with the very nice SUPREME finition ( see photo ) and the power supply transformer with the same finish
However, I have been busy with many other things like building solid state amplifiers to compare them with the Baby Huey, and I wanted to have more experience before to start such a high end project... As Norman said : "He was so satisfied with his amplifier that he never designed an other one..."
I am planing to use KT120 output tubes since they have very good reviews and because it is difficult to find good 6550 today, I had very good ones from Svetlana but they are no more produced
I put the schematic and the PCB in this message but it is still a work in progress therefor all remarks and/or suggestions are welcome... For info I made the PCB with the serigraph on the bottom side to facilitate the building
Rgds,
Marc
Attachments
Thank you Brian and Marc, for your comments.
Brian, your observations about the relatively low gain of your BHs were especially helpful and persuasive. I will get on with the regular Toroidy transformers and delay projects with CFB output transformers until after my Baby Huey amplifiers were done and enjoyed for a while.
Thank you Marc, for posting the information about your future project with the “ultimate Baby Huey TENA”. I look forward to further postings on your project. Gorgeous looking transformers!
Brian, your observations about the relatively low gain of your BHs were especially helpful and persuasive. I will get on with the regular Toroidy transformers and delay projects with CFB output transformers until after my Baby Huey amplifiers were done and enjoyed for a while.
Thank you Marc, for posting the information about your future project with the “ultimate Baby Huey TENA”. I look forward to further postings on your project. Gorgeous looking transformers!