TEPPAZ 730S

[chocopower]

Hi.

I saw this baby time ago in ebay Spain and just for fun make a bid.
I totally forgott about it till i recive a message about a winning auction!

http://cgi.ebay.es/Amplificador-Valvulas-TEPPAZ-730S-EL34-Tube-AMP-Power-/200610444848?pt=LH_DefaultDomain_186&hash=item2eb5507630

Pretty nice construction and my favority configuration. EF86 and EL34!! for 60€!!

Sorry for the "ego trip", but i needed to tell somebody!!!

edit:
Schematic and description. Any French in the room?
radio-piffret.pagesperso-orange.fr/Audio_Teppaz_730.pdf

[Willabe]

Hey that looks realy clean. Nice score chocopower!


       Brad     

[overtone]

A very nice sore! That is a real head turner.

I would really like to comprehend why they chose the pentode section of a ECL82 for the phase inverter position. If I understand correctly it is set up to run as a triode cathodyne PI. My french is useless, but there seems to me to be something in the text about this choice of PI not needing an extra driver stage before the PA. I always thought that the cathodyne PI only ever has a gain of 1, so I am confused... again!

Best, tony

[PRR]

> not needing an extra driver stage

All audio BS is relative BS.

In 1957, the leading design for power-amps was the Williamson. Gain, cathodyne, two gain/drivers, then the power tube grids.

Normal skimpy triodes as cathodyne have a hard time making BIG swings into LOW impedance grids. The added drivers do the hard work.

Especially hard in the 730S because it uses very conservative 47K EL34 grid resistors.

> they chose the pentode section of a ECL82

Not (just) a pentode, a POWER pentode. Or as it is used, a BIG FAT triode. At 10mA (as used here) the triode plate impedance is under 2K. We find 1:5 relation between the 2K tube and the 10K cathodyne resistors, another 1:5 from 10K catho resistors to EL34 grid resistors. This relationship will allow LARGE voltage swings. Far in excess of the 30V apparently needed to flog the tubes de puissance grids.

> the cathodyne PI only ever has a gain of 1

Here about 0.97, yes. But there is a gain stage -ahead- of it, gain of 40 or 50. Say gain of 40, catho gain about 1, EL34+OT gain about 0.5 to the 4 ohm tap. Total forward gain about 20. Feedback path 2K2:100 is loss of 23:1.

In fact it has very little NFB on nominal load. Very much like the older Fenders. This also explains why the THD numbers are not so low. A Williamson has 10X the NFB and shows ~~1% THD at medium power, this has 2% at 2/3rd power (but twice the total power of early triode-KT66 Wiliamsons).

It is a very good PA amp. Not a hi-fi amp (though there were much worse "hi-fis", they sold for less than this amp). It's not a guitar amp but triode-strapping V1 or V2 (or both) might get the gain down enough to be useful. Or chop everything to the left of V3 and use standard guitar preamp ideas.

[chocopower]

>It's not a guitar amp but triode-strapping V1 or V2 (or both) might get the gain down enough to be useful. Or chop everything to the left of V3 and use standard guitar preamp ideas.

I wanna keep this one as stock as posible, because i really like the construction, look, etc, so i going to try to use the "PREAMPLI" conecction in the anode of the "real" triode side of the ECL82.
I still not have the amp in front of me, but can´t wait to know if is a PREAMP IN or a PREAMP OUT.
My low tech skills level tell that is an OUT, but , what do you think?

some more questions:

.-Just after the GZ32 (a really BIG tube), there are a light bulb marked as "TEMOIN SURCHARGES" and, seems to me, wired in series. Just like a bulb limiter, but in the HT line.
In the fron panel of the amp there are a green plastic/cristal who allow to see that light. I supoose is an "overlord" indicator.
Maybe is acting like a standard bulb limiter saying us when there are a shortcircuit in the amp?

.-You can see in the schemo and pictures, that there are a internal monitor speaker. No information about the impedance.
Is connected to a dedicate sec. on the OT. My experience tell me that i have to disconet it when a external speaker is working, but maybe the design allow a combinated use.
¿Any idea?

.-The bias resistor (R32) is 12k. I would like to a add a pot.
25k pot with a 1k5 to 4k7 resistor to ground, is a good starting point?

Thanks in advance!

[PRR]

> the "PREAMPLI" conecction

This is an OUTput.

> I supoose is an "overlord" indicator

OverloAd, yes. The idle current is 60mA, 0.080A. The full-power current is 290mA, 0.0290A. The lamp is full-bright at 0.3A. At idle and for most un-clipped speech/music, the lamp will be too dim to see. In full-power testing, or heavy overload, it will glow pretty bright.

> internal monitor speaker. No information about the impedance.

It appears to be tapped "below" the 2 ohm tap. I am sure it is connected for negligible power (less than 1 Watt). It is useful when the amplifier is in another room, far from the main speakers.

> The bias resistor (R32) is 12k. I would like to a add a pot.
25k pot with a 1k5 to 4k7 resistor to ground, is a good starting point?

No. It would allow dialing the bias to -7V, which is much too small. 10K stopper (minimum -25V bias, which is still too-hot) with 10K or 25K pot.

[chocopower]

> internal monitor speaker. No information about the impedance.

It appears to be tapped "below" the 2 ohm tap. I am sure it is connected for negligible power (less than 1 Watt). It is useful when the amplifier is in another room, far from the main speakers.

I´ll test it when arrives. The two questions are:

.- Can i use the amp just with the internal speaker? (With 1watt, could be a funny living-room amp)

.- Should i disconect it when a external speaker is used?

as usual, thanks!

[PRR]

> full-power current is 290mA, 0.0290A. The lamp

I am wrong. The lamp does NOT pass plate current. Driver current should be ~~11mA, and EL34 screens should be 10mA idle 50mA loud. Normal current will not light the lamp at all. Any glow in the lamp means extreme tube abuse. Fix the load (it may be disconnected) or turn it down.

R34 must really be 4_._7K, not 47K.

You must connect external load (speaker).

You can keep or disconnect the internal monitor; does not matter to the amp.

[chocopower]

Thanks!

i´ll post some pics when amp arrives.

[birt]

these old french amps look nice. you should also google the brand Bouyer. their designs are really cool.

[chocopower]

Yes, is a really nice amp.

i found this pic. Here is inserted in a more elavorated chasis:

[birt]

I'm not french or english but i can speak both languages. There might be some mistakes but here's what i can come up with when i translate the PDF:


Teppaz is mostly known for it's record players, but is also a maker of PA equipment that has amplifiers and soundsystems. But they don't have a range as big as Bouyer.

We think it would be interesting to see what goes on inside a Teppaz amplifier. We chose the 730, a 30W amp that is relatively new (1957), that over time had a number of improvements. The 730 also exists with an integrated record player.

The philosophy that preceded the design of this device is different to what we saw in the Stentor range made by Bouyer. The increased value of the plate resistors in the preamp stages, R11 for V1 and R16 for V2, permits to get the maximum gain of these tubes. The decreased bandwidth is compensated to a certain point by the network R17-C11 that brings out the high notes. It precedes the passive tone control that is made by the network around P7 and P8.

The use of the ECL82, V3, is a bit special as it is a cathodyne phase splitter stage. This stage adds no gain, but has the advantage that a driver tube or interstage transformer is not needed. All this contributes to a reduced number of tubes needed. The low value of the bleeder resistors (R29-R30) avoids the use of grid stoppers and provides a certain stability for the power tubes (V4-V5) that could be touchy with their 11mA/V.

The use of a switch to change the output impedance means that this device has to be very good quality. But it also has an original way of protecting the OT with the bulb V7.

It is worth mentioning that the CT of the OT primary is connected to the cathode of the rectifier, but the first filter cap (C26) is connected after V7.

In case of severe overload, the energy stored in C26, is released only through V7 and it can be assumed that the filament will break. This really is the weak point of using a bulb as protection, but the stress this will put on the OT should be small. The current available at this point is not continuous, but a pulse at 100Hz and its the filter capacitor C26 that reduces the apparent internal resistance. (the note on the left side sais that you can provoke arcs in transformers, as long as they are weak enough. This way you can also find the breakdown voltage in a non destructive way.)

the 730 also has an optional upgrade to amplify the sound of film tapes. Teppaz has also made poweramps powered by batteries where the B+ is generated by a converter.

the characteristics:
bandwidth, output power, working power, distortion, mic input sensitivity, PU input sensitivity, range of controls (bass and treble)

[RicharD]

That thang is beautiful.  I'd love to see what makes that control surface tick, mechanically that is.  I assume it's a sting and pulley deal.  It's in too good of shape to attack with a pair of dykes.  I'd keep it as original as possible.  If you wanna do the internal speaker only thang, get a 100W 8 ohm resistor and hang it in place of the main speaker.  If you do jalopy it, please please please don't poke any holes the front.  Sweet find.

[chocopower]

Now worry!

I going to preserve even the banana imputs/outputs. I had built some banana-to-jack adaptors time ago for another vintage amp.
For the main power cord i was planing do the same thing. 3 cord wire with a splited banana conector for the "terre" (ground) hole.

if is everithing ok, i just take out the C27 and C28 caps in the AC input, check the polarized caps and thats all.

As PRR said is to much gain for guitar, but the PU-RADIO input betwen the EF86 worth a try. (that way will be something similar to a DrZ Rout 66...)
If that dont work, i´ll try a not invasive preamp input before the ECL82.

I found that link with some nice pics:
http://www.hiboox.com/go/albums/miscellaneous/teppaz-730td-eco-60,3e7581af7d0b6f988e7c8cd2c9dd9162

[chocopower]

Amp arrives!

woa!! thas a beautifull piece of ingeniery!

There some caps with evident signs of damage, and the frontal pannel needed some arragement, but is a really nice "tiny terror" size amp!
Sorry for the phone quality pics.....

[RicharD]

Thanks for the pictures.  It looks like 2 of the pots are on string and pulley and the other 2 are on "spirograph" gears.  Really cool.

[chocopower]

PRR
"R34 must really be 4_._7K, not 47K."
Checked. You are right. Is a big 2w c.c. resistor.

Butterylicious
You are on right too.
The 2 gain controls in the front panel move a string, who move a wheel, who moves the pot....
The "spirograph" part is for the EQ.
I´m really surprise. because the mechanical is working perfect after a lot of years (1955)

[chocopower]

> full-power current is 290mA, 0.0290A. The lamp
I am wrong. The lamp does NOT pass plate current. Driver current should be ~~11mA, and EL34 screens should be 10mA idle 50mA loud. Normal current will not light the lamp at all. Any glow in the lamp means extreme tube abuse. Fix the load (it may be disconnected) or turn it down.

Yesterday, after the safety works, i made a light bulb test and everything seems ok.
Them, i made a test at full AC voltage and the internal lamp glows after 15-20 seconds.
It glows with or without tubes (righ now i have a diode plug into the rectifier shocket)

My first suspect are the power filter caps, but maybe you have other culprit.

thanks!

[chocopower]

Hi.
Had been a while, but finally is time to finish this one.


Basic stuff is done (3 prong, e-caps, power resistors, ...)


Problem is voltages


Im getting getting 500vdc at plates with rectifier and power tubes installed.


400vac before rectifier.
7.4vac in heaters.


220v primary with 225vac at wall.


I remove the "temoin surchages" bulb.
I still not have a ECL82 in hand.  Maybe is important to tame down voltages. Its a power pentode and for sure is gonna demand a lot of power....
I installed too, safety diodes in rectifier shocket.




Any idea???

[sluckey]

Im getting getting 500vdc at plates with rectifier and power tubes installed.

That sounds about right with 400vac before rectifier.

EDIT... Just saw the schematic and noticed the voltage selector. I suggest you set it to 250V.

[chocopower]

 Yes...




Schematic call for 400vdc at first node, thats why asume 300vac before rectifier....


And in a old version of the schematic they wrote 150v at secondary center tap.....


At first I suspect about a wiring error when I disconnect 110 - 125 - 185  primary wires, but I recheck every thing and even measure primary winding resistance. Everything is correct there. Just in case, I test with the built in limiter bulb without success...


ECL82 is in the way, so I im gonna recheck every connection, but with p.t. isolated I get same numbers, so there is few options.....

[chocopower]

Silly me... I just realized a "360" inscription in the secondary outs....

[chocopower]

EDIT... Just saw the schematic and noticed the voltage selector. I suggest you set it to 250V.

Yes. I used it in the first steps of the test in conjunction with a bulb limiter. I was getting more "expected" voltages. I have to check heater voltages if they are in the ballpark could be an easy solution

[sluckey]

I mentioned the voltage selector because of the 7.4V filaments. Selecting the 250v position will give you the lowest filament voltage possible (lowest B+ also). But if you are already set to 250v, there's not much to do inside the amp.

[chocopower]

With 250 primary i get  460vdc on plates and 6.00vac on heaters..


I'm gonna wait for the ECL82 to go further...

[chocopower]

ECL82 arrived.


Amp working and sounding quite god at home levels.


Cap job and all drifted resistors upgrade is done.
I split too, the 2 EF86 in two different channels.
Waiting for 1k/5w for the EL34 screens.

[chocopower]

Some pictures.

[chocopower]

More

[chocopower]

And updated schematic with voltages.

[chocopower]

hi again.


Amp running and sounding quite well.
 I have an issue.


With volume (both channels), below 20% i have a hum noise.
In both volume pots, there are a spot where the noise just disappear.
Noise disappear too with NFB loop disconnected.
I cheek every solder joint, tube socket...
My main suspicious is the triode side of the ECL82.
Voltages in all the amp match the schematic quite well, but that triode is way off specs.




Any idea?

[shooter]

Any idea?

your idea of the triode seems good, looks to be easy enough to "jumper past" it for testing

[PRR]

> below 20% i have a hum noise.
> In both volume pots, there are a spot where the noise just disappear.

Then you have *two* hum sources. Out of phase. One bfore the pot. One after the pot. At some magic mix, they cancel out.

Turn pots full-down and sort out the post-pot hum. Then find the pre-pot hum.

[chocopower]

hi,


i,ve been playing with this one for a while, and really happy with the sound.
The hum problem was mains center tap sharing ground with bias one (original design).


Right know, i just need to modify the bass response. In both channels, but mainly in the James TS one, with some speakers cabs, even with the bass control set to cero, there are to much bass.


I have some test to do, but i would like to know you opinion.


Where is the best place to cut the extra bass in this amp and why?


I attached a schematic with circle in green where i think i can start.


thanks in advance

[shooter]

couple easy things to try before TS stuff
drop the cathode bypass cap from 22UF to 5uF, maybe 10uF
change the coupling cap from .022uF to .01uF
see if those move you in the right direction.

[PRR]

That MEDIOS control is unfamiliar to me. It is a significant load in the middle of a James tonestack which is all about frequency-selective impedances.

[PRR]

Oh! You can't drive a James (most tonestacks) from a pentode. The tonestack looks like a capacitor. The pentode is a very high impedance. The tendency is for response to fall and fall.

[chocopower]

That MEDIOS control is unfamiliar to me. It is a significant load in the middle of a James tonestack which is all about frequency-selective impedances.

Sorry, the mids control is wrong in the schematic. Output from James TS goes to the top of the 1M pot.
I edit it in the previous reply.


It's a Framus mid control. I think, Gibson used it too in the 50-60s

[chocopower]

Oh! You can't drive a James (most tonestacks) from a pentode. The tonestack looks like a capacitor. The pentode is a very high impedance. The tendency is for response to fall and fall.

I know, but that channel is almost stock.
That was the P.U./Radio input, and only changes are the addition of the vol. pot, a cap change in the ground of the treble pot and the elimination of the resistor/cap at the input of the james T.S.


When i did it, i was not sure if its gonna work, but was easy and just give a try..  It works.   
I like the tone, and i can tune it really bright with the treble pot.. The "too much bass" thing is easy to fix.

As you you said, it should sound very dark, and in fact, i had that issue in the pass with pentodes, but here, is working...