Jump to content

The long awaited "Little Sweetie" mono SETs


tube fanatic

Recommended Posts

 

The Edcor for 17 bucks has a frequency response of 70Hz-18kHz at minus one decibel. Minus 3 decibels would be lower yet, a dumb guess on my part would around 50Hz?

 

The one for 30 bucks has a low end roll-off of 50Hz at 1 decibel, lower yet at -3dB.

I know these are good options on the cheap, but may be lacking in lower bass once the circuit is lashed up. Then again, maybe not...

 

The big 87 dollar 25 watt single-ended Edcor OPTs advertise 20Hz-20kHz frequency response. I know that 200 bucks is a bit of cash, but still not a bad deal at all considering the price of output transformers nowdays.

 

Another brand, (Transcendar?) is reasonably priced option from what I understand...

 

 

 

I saw that and was going to link the $87 model too, which I would consider if I didn't have the Electra-Prints, but also noticed that the Hammond 125CSE data sheet shows 100 Hz. - 15 Khz (+/- 1db max. ref. 1 KHz). 

 

I don't have any experience either with the Transcendar; however, here is the website link. 

 

http://www.transcendar.com/2.html

 

Transcendar 5 watt option, 5K to 8 ohms, Frequency Response: 20 Hz to 80 kHz, Maximum DC Bias Current: 65 mA, Primary DC Resistance: 350 ohm, Primary Inductance: 20 H,  Part Number TT-005-OT , $45.00 each

 

Transcendar 5 watt option, 5K to 4 & 8 ohms, Frequency Response: 20 Hz to 100 kHz, Maximum DC Bias Current: 65 mA, Primary DC Resistance 370 ohm, Primary Inductance: 20 H, Part Number TT-104-OT, $50.00 each

 

 

 

 

.

 

When comparing the bandwidth of audio output transformers you need to be very careful that they are measured at the same power level.  For example, Hammond always presents the spec at full rated output.  Transcendar, on the other hand, measures at 1 watt!  At 1 watt, the Hammond 125CSE has phenomenal bandwidth (see George Anderson's review below).  It also pays to look at the transformer's distortion if such specs are available- usually they are not!  In terms of price vs. value, I've always found that Hammond offers a great deal for the money.  This isn't to say that the other companies don't do the same, of course.  Just wanted to point this out.

 

http://www.transcendar.com/

 

https://www.hammfg.com/electronics/transformers/audio/125se

 

http://tubelab.com/articles/component-testing/budget-output-transformers/

 

Maynard

Link to comment
Share on other sites

Maynard brings up a good point about the correlation of bandwidth and power tested. I like to look at the primary inductance as a good indication. For example the Edcor 10 watt 3.5k:8, GXSE10-8-3.5k only has a primary inductance of 6H, not that great. You will see the Transcendar is 20H for a 3k:8. So the Transcendar has less turns but way more inductance :)   The similar impedance Hammond transformer has a primary inductance of 20H as well.  

 

I spoke with George about the Transcendars, he bought a bunch of the older designs, they have improved since then.

  • Like 1
Link to comment
Share on other sites

 

When comparing the bandwidth of audio output transformers you need to be very careful that they are measured at the same power level.  For example, Hammond always presents the spec at full rated output.  Transcendar, on the other hand, measures at 1 watt!  At 1 watt, the Hammond 125CSE has phenomenal bandwidth (see George Anderson's review below).  It also pays to look at the transformer's distortion if such specs are available- usually they are not!  In terms of price vs. value, I've always found that Hammond offers a great deal for the money.  This isn't to say that the other companies don't do the same, of course.  Just wanted to point this out.

 

 

http://www.transcendar.com/

 

https://www.hammfg.com/electronics/transformers/audio/125se

 

http://tubelab.com/articles/component-testing/budget-output-transformers/

 

Maynard

 

 

 

 

Thanks for providing this insight.  Although, I may not be able to build this for a couple of months yet, I believe posting various potential alternative parts and observations will help pull out various nuances not considered like you noticed and serve to help us less experienced. One of the things I considered when posting the links to Edcor budget transformers was George's (tubelab) evaluation of the Edcor XSE 15-8-5K, along with the Hammond 125CSE and figured that for the budget build, either an Edcor or Hammond selection would work fine.

Link to comment
Share on other sites

Maynard has went to the trouble of designing this amp for our members and we should reciprocate by building some of them.

 

 

I'm ordering parts for the Parallel 6AQ5 SEP...however, I'll be doing the circuit on a breadboard with a DC tube power supply.

If I like what I hear, I'll build a complete amp...or a pair of mono ones...

 

I'm curious to see how this circuit can compare to my directly heated 2A3 amplifier. My 2A3 ain't no slouch. A audio friend, (Dan Marshall) designed and made the circuit, scrounged all the parts for a complete amplifier, and then eventually sold it to me as a kit...not a complete kit, but close enough.

The driver is a 6EA7, the smaller side is the AF stage, the large side is coupled to a cheapie Allied 10K 1.1 inter-stage transformer. It's a push-pull type of IT, so one side is bridged with a .1uF cap. 

 

I've bought a pair of these Allied IT's and through emails with Dan, tested a pair on the bench, mostly to determine proper polarity and how wire in the circuit proper.

I can't recall off hand how I set these up for testing, but I wrote everything down.

Un-bridged with square waves, the IT rings terrible. By 30kHz, it looks akin to heart rate from a meth-head with a murmur. Once bridged with the .1uF cap, square wave response is nice up to 30kHz, where the upper harmonics are starting to distort the square wave itself. 

 

The 10K 1.1 IT is coupled to the 2A3. It's a three-stage circuit, so the speaker taps need to be inverted for proper speaker polarity. He used the Duncan PS program to come with a power supply design. Found a old Gramer television PS transformer with a rather chunky 6.3 heater winding. This winding is used for the 6AU4 half wave damper tubes. The TV PS xformer has lighter 6.3 winding used for the 6EA7.

The PS is CLC, with 4 henrie Raytheon choke...Mallory 36uF oil caps, and 100uF Mallory electrolytics.

 

The 2.5 filament transformers were placed in the chassis using a scope with power applied to them, for best placement with regard to magnetic interference.

The filament/cathodes are connected to the center-taps of the filament transformers, no hum/null pots. I think a Mills resistor is in the cathode/filament, with a oil bathtub cap of around (50uf?) for the bypass.

The AF side of 6EA7 has a polyprop of 12uF for the cathode bypass. Mills and Caddock are used for resistors.

 

The OPTs are Scholl...they do have a 3.6K primary. The project was intended for 300B, and the kit came with 2.5 volt TJ 300B solid plate globes. (which lost most emission years ago...)

However, 2A3 works just dandy. Sylvania 5930 is a ugly 2A3, but they are military regulator tubes and won't burn up any time soon...I've used Sovtek with good results.

 

I never understand why some complain of no lower bass with a 2A3 amp. Maybe it's because Dan had bass in mind when he designed the circuit...something with regard to how the inverted three stage design works with the spacing of triode curves and how it relates to bass transients such as synth notes.

It's really way over my head to explain in proper manner.

 

If the 6AQ5 parallel SEP can sound as big and open as the 2A3 amplifier, I'll be very happy. I'm sure I'll get deep strong bass with the James 20 watt OPT.

 

After time, an abundant pile of amplifiers just becomes a hoarding issue.

 

post-4542-0-93440000-1446398363_thumb.jp

Link to comment
Share on other sites

I also found this transformer at Antek I fell would work if Maynard can design a circuit with a ground using a transformer without a center tap. I do think this one is overkill though.

 

http://www.antekinc.com/as-4t360-400va-360v-transformer/

 

That transformer with an EY88 diode half wave rectified with a choke input (not capacitor input) would give around 200V.  http://www.ebay.com/itm/6AL3-EY88-NOS-NIB-Amperex-VACUUM-TUBE-GUARANTEED-/131639670778?hash=item1ea655f7fa:g:6zUAAOSw~bFWMQ0E

Link to comment
Share on other sites

William asked for a possible alternative power transformer for use in a stereo (dual mono, single chassis) "Little Sweetie."  This offering by Antek should work just fine but would require a bridge rectifier, and a probable change in the value of the filter resistor:

http://www.antekinc.com/as-1t200-100va-200v-transformer/

Since I don't know the exact voltage this transformer will provide under load, the filter resistor value would have to be determined by experimentation.  I can draw a simple schematic if anyone wants it.  There are many ways to work around getting the voltage we need from transformers which may not be an exact match.  As JP suggested, we sometimes use a choke input filter which provides a much lower output voltage than a capacitor input filter.  Or, we may have to use the internal voltage drop of a tube rectifier to get rid of excesses.  It's rarely a problem unless the transformer chosen is just too far off from the needed voltage.  

Mike, I can't wait to read about your results with the parallel SEP!  With the Hammond opts, the bass is unbelievable.  I'm sure those James xfmrs will do as well.  It won't sound like a SET, but its midrange and top end can get you close while retaining the typical bass characteristics of SEPs.  

Maynard

Link to comment
Share on other sites

Since I don't know the exact voltage this transformer will provide under load, the filter resistor value would have to be determined by experimentation.

 

In the spec sheet they give load test data for different loads, I calculated each high tension winding to 50 ohms source resistance (they tested them in parallel and I calculated 25 ohms from the data). With a bridge rectifier and the same value input capacitance from Maynards schematic you would need to adjust R10 to 1k to drop the 263 volts down to the intended 210.

 

Maynard feel free to check my numbers and correct me if I am wrong.

 

EDIT: You would need to use a 5 watt resistor for R10.

Edited by xxJPMxx
Link to comment
Share on other sites

The 175V version of that power xfmr is a bit too low.  William, please post the dc resistance of the primary of the output xfmr which you want to use so I can see how much voltage will be dropped there.  I'll get a p/s schematic drawn within the next few days.  What we're looking for is approx. 200V between the plate and cathode of the 6Y6.  We need to add the 30V of bias to that, and also the dc voltage drop across the opt primary.  So, we're going to need around 250V minimum out of the p/s filter for the output stage.  Regarding using the Triad chokes, it's really overkill.  Why use a $15 choke, when a 50 cent resistor will work just fine?

 

Maynard

Link to comment
Share on other sites

The 175V version of that power xfmr is a bit too low.  William, please post the dc resistance of the primary of the output xfmr which you want to use so I can see how much voltage will be dropped there.  I'll get a p/s schematic drawn within the next few days.  What we're looking for is approx. 200V between the plate and cathode of the 6Y6.  We need to add the 30V of bias to that, and also the dc voltage drop across the opt primary.  So, we're going to need around 250V minimum out of the p/s filter for the output stage.  Regarding using the Triad chokes, it's really overkill.  Why use a $15 choke, when a 50 cent resistor will work just fine?

 

Maynard

 

 

You are correct! For some reason I was thinking there was only 15v for bias and 15v (300DCR?) across the primary output transformer so I was shooting too low. The 200V transformer is better but you still have to increase R10 to 470 to get to 250 at that node.

Link to comment
Share on other sites

As I see it the dc reisistance of the primary is 187 ohms.

 

 

Good info!  So That's 10 volts there across the primary of the output transformer and then the 30v for bias...................so we want roughly 240v B+ correct? 

 

Here is what I got. Maynard please double check and approve :D

 

FIdbWCo.png

 

 

http://www.antekinc.com/content/AS-1T200.pdf

Edited by xxJPMxx
Link to comment
Share on other sites

Mike, I can't wait to read about your results with the parallel SEP! With the Hammond opts, the bass is unbelievable. I'm sure those James xfmrs will do as well. It won't sound like a SET, but its midrange and top end can get you close while retaining the typical bass characteristics of SEPs.

 

I ordered the parts from Mouser, except I completely forgot about the 1 meg pots for RC HP feedback filter. I guess I'll just wing it with resistors...

Link to comment
Share on other sites

You guys have been busy since last night!!!  To summarize a bit, William asked about using an Antek power transformer to simplify building a stereo version of the amp on a single chassis while saving some money.  The AS-1T200 is a perfect choice as it is basically two transformers wound on a single toroid.  Since the high voltage secondaries are not center tapped, a full wave rectifier as I used with the Hammond power xfmrs, can't be used. The 2 choices are to use either a half wave, or a bridge rectifier.  The advantage of the latter is that the ripple will have a frequency of 120 Hz (instead of 60 Hz for the half wave) which is easier for the caps to filter.  Having 2 primaries even allows each channel to have its own on-off switch if desired.  If a single switch is used, it may be necessary to increase the value of the fuse a bit to avoid popping it from the turn-on surge.  One concern about unshielded power xfmrs is magnetic field radiation.  However, toroidal types are usually much better in that regard.  If using unshielded opts like the Hammond 125CSE, there is still the possibility of some field pickup by the latter, so using William's "double decker" construction idea may be a good way to go.  On the other hand, if going with the enclosed Edcors, I don't think it will be an issue.  One last point on that is that the Antek has an internal Faraday shield (the purple wire shown in the diagram on their web site) which should be grounded.  It can help keep some of the AC powerline grunge from getting into the secondaries.    

Many thanks to JP for drawing out a schematic and modeling the supply!!!  I agree with the information completely.  To clarify, R2, the 4800 ohm resistor shown in JP's schematic, represents the effective load resistance of the output stage (using Ohm's Law, R=E/I, we have 240/.05= 4800 where 240 is the supply voltage, and 50 ma the current draw of the 6Y6).  DO NOT INCLUDE THAT RESISTOR IN AN ACTUAL BUILD!  In JP's schematic, the lead which connects the 4.8k load to the 560 ohm filter resistor (which should be in the ballpark of what will actually be needed- simply adjust its value for approx. 200V between the plate and cathode of the 6Y6) corresponds to node "C" in my original schematic above.  The output xfmr gets connected there.  Be sure to add the additional filter section for the 6SJ7 as well as the bleeder.  Regarding the bridge rectifier, there are 2 ways to do it.  One is to make your own using 4 diodes, like the 1N4007s  shown above.  The other is to simply buy a ready-made bridge rectifier (a 1000V/4 amp unit like the RS-407L works great).  I like the latter approach as it's easier to work with.  Be aware of the lead markings on the housing of the bridge.  You will see 2 leads marked AC, or with sine waves, which go to the power xfmr secondary, and leads marked plus and minus.

So, I guess that summarizes things for now.  Keep posting questions as you think of them. 

Maynard
 

Link to comment
Share on other sites

Good morning everyone!

 

 

Yes do not include the 4k8 (R2) resistor in the actual build, as Maynard pointed out it's just the "load". The 6SJ7 is also part of this load but only 1mA of it.

 

R1's value of 560 should be fairly accurate. One thing that might cause this to be different is if someones line voltage is really low or really high. Or I calculated the winding resitance wrong. Maybe someone could double check my work.

 

This is from the datasheet of that transformer:

 

 

LOAD TEST: Input 120V 60Hz into the primary coils (in parallel), output 1 and output 2 in parallel to the load, and measure voltage and current at different load levels.

 

0 current 206v

 

.46A  194V

 

.7A    188V

 

 

To figure out the winding resistance I divided the change of voltage by the change in current: (194-188) / (.7-.46) = 25

 

Since the test was with the two secondaries in parallel I doubled that figure to 50 because when paralleled the resistance would halve. So I plugged the data into Duncan's PSU and voila. Since those resistors are not much money it would be wise to still buy some values higher and lower just in case.

 

 

I will get to drawing a full schematic for the power supply and the Antek transformer later this evening.

Edited by xxJPMxx
Link to comment
Share on other sites

JP, you're off the hook for drawing the updated schematic!  I had some free time to draw one for William and anyone else interested, but please look it over for any mistakes!  C6, C7, C8, R11, and R15 are the same values as in the original parts list.   Values for the bridge, and new value for R10, are shown on the schematic.

 

I'll take this opportunity to say how fortunate we are that Klipsch speakers are so amazing with these low power tube amps.  I keep trying to find a justification to do some high power designs for this forum and can't come up with any reasons to do so.  May the company continue to bring out super high-efficiency models in the future!  Even after all these decades it's still fun for me to see the look on people's faces when they experience their speakers being easily driven to louder than needed levels with these small, inexpensive, low power devices which allow the speakers to provide sound quality which surpasses anything they have previously heard!  To say that Klipsch speakers are "made to be used with tubes" is an understatement in my opinion.  There's nothing like synergy when it comes to high quality audio reproduction.

Maynard

alternative-ps-10.pdf

Edited by tube fanatic
Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...