Pumpkin Problems

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
This new thread is NOT INTENDED as a slanging match against Pre-Amps or ZenMod - who produced the Pumpkin.

Alexander has been very helpful in trying to resolve certain issues with the Pumpkin and the Shuntky - Particularly poor PCB quality which I believe he has now addressed.

This thread is intended to get these beauties working az ZenMod intended.
 
My build was unusual. As I couldn't find a single 42-0-42V transformer, unless I wanted to pay for a custome build - EXPENSIVE. I used two 20-0-20V s connected in series.

The finished result was diassapointing.

Both Shuntkys failed due to shorts caused by incomplete etching. Luckilly nothing blew up and it was only a case of finding the shorts and running a craft knife between the tracks.

Finally I had 36-0-36V

I've still got a problem though. The Pumkins are NOISY. Both of them have "Whooshing" noises like being at the seaside.

ZenMod has offerred that the transformers are the problem.

I'm now in the process of testing the Pumpkins with a simple regulated supply.

This will at least prove if the Shuntkys are at fault of the Pumkins.

ZenMod also suggested that it might be down to my choice of components. I have never heard of something sounding completely HORRIBLE with the recommended components - Things might sound BETTER with AUDIOPHILE components.

First question. Is it OK to test Pumpkins at +/- 25V ? The input to the regulators is +/- 42V off-load which may be too high for the Pumpkins.
 

Attachments

  • DSIR9286.JPG
    DSIR9286.JPG
    203.3 KB · Views: 387
  • DSIR9287.JPG
    DSIR9287.JPG
    189.3 KB · Views: 383
Can I just add.

Except for the "Whooshing" and minor "Hum", the Shutkys and the Pumkins do produce lovely music. There cant be too much wrong here.

ZenMod himself says that these are not easy to use. They might be VERY CRITICAL about layout and component choice.

This time I am building them into a much larger reclaimed VCR chassis.

Alexander - I will be dropping the gain from 11X to about 3X as I don't need the extra gain.

Since our last contact I have built a B1 which proves that the Aleph 4 is working faultlessly and silently.
 
I'm not the only one who has had problems with the Pumpkins.

Working correctly they are brilliant. Generg and I have been comparing them with the Nelson Pass B1. The B1 is lovely but the Pumpkin has so much more LIFE.

This is one of the qualities that is difficult to quantify or measure. I think a lot of it comes down to taste and the ancilliaries that you are using.

The Pumkin certainly has more energy and sounds alive. Whereas the B1 does its job and - what can I say - its uninteresting.

I bought the B&W DM683s to bring to life the midrange compared with my DM602s3's
Other speakers may not like the enlivened mid-range, and may sound too bright with the Pumpkin.

All I can say is that the Pumpkin imparts its part into the music chain which is both positive and pleasing to listen to.
 
Last edited:
Once I've got the Pumpkins working on a simple regulator I'm going to try and find out what is wrong with the Shuntkys.

This is unusual. Both are supplying +/- 36V but with horrendous noise.

If I remove the two fuses the RAW 60V DC at the fuse input is CLEAN.

Put the fuses in and I get the most horrendous 100mV 100Hz ripple. This is not down to the preceding circuitry as I can load that up quite happily to 100mA and it remains CLEAN.

There is most definitely something amiss in the Shuntkys and I cannot see any component or soldering problems.
 
Having read a few threads about the LM317/LM337 regulator family.

Here's a question about using it close to its limits.

In the ongoing problems with the Pumpkins I'm building a basic regulator test bed just to find out where the problem lies.

I've attached the schematic of the +ve regulator, the -ve it its opposite.

The design is BASIC and does not protect against over zealous operation of R2 - Basically if R2 is set to minimum then the Zener Diode will Pop. R2 is adjusted to half resistance during assembly and then carefully adjusted when testing the circuit.

As it stands it works perfectly into a resistive load.

My concern is what will happen with Vin at 42V using the Pumpkin as a load ?

Initially the Pumpkin supply bypass caps will be apparent short circuits to ground. They are only 2 x 220uF but might that be sufficient to exceed the Vi-Vout limit of the LM317 ?

Anyone wanting to use this circuit as a working 36V regulator please place a 1K resistor is series with R2.
 

Attachments

  • 002.jpg
    002.jpg
    308.5 KB · Views: 307
  • DSIR9295.JPG
    DSIR9295.JPG
    168.2 KB · Views: 297
Last edited:
Reading for the testing of the Pumpkins

I've now got both regulator boards built.

Just waiting for some intuitive answers about using the LM317/337 at its limits.

For those with inquisitive eyes the two PBC layouts are marginally different, that is how the two layouts look similar - We all know that the LM317 has different pin-outs to the LM337.
 

Attachments

  • DSIR9296.JPG
    DSIR9296.JPG
    196.8 KB · Views: 280
Replace the IN to OUT diode with a 30V Zener.
It still acts as reverse current protection but when the IN to OUT voltage exceeds 29V, the Zener starts to conduct. The in out voltage should never exceed the 317's voltage limit.

The Zener junction temperature is going to go astronomic during start up.
I have no idea how many restarts will cause it to fail.

You can add a CCS pre-reg to limit the temperature rise of the Zener.
 
Replace the IN to OUT diode with a 30V Zener.
It still acts as reverse current protection but when the IN to OUT voltage exceeds 29V, the Zener starts to conduct. The in out voltage should never exceed the 317's voltage limit.

The Zener junction temperature is going to go astronomic during start up.
I have no idea how many restarts will cause it to fail.

You can add a CCS pre-reg to limit the temperature rise of the Zener.

WHY.

Even if the full 42V is applied at the output of the LM317, the zener has nearly 4K in series with it.

If R2 was set to minimum, which it isn't. its set to nearly maximum, then I would agree.
 
The limiting IN to OUT voltage for the 317 is 40V absolute maximum.

At start up the OUT is at Zero Volts.
The IN pin goes to +Vdc of the PSU.

What is the maximum PSU output at start up?
If you have a nominal 42Vdc when drawing continuous operatimg current and when mains is at nominal UK voltage then I would suggest that you allow for 42Vdc+10% as the worst case start up IN voltage.

If you replace the IN to OUT diode with a 30V Zener, then the Zener will pass current when the IN to OUT voltage exceeds ~29V and the higher the applied Zener voltage the higher the current.
The Zener acts as a temporary bypass to help prevent the 317 failing due to excessive IN to OUT voltage at start up.
a 500mW 30V Zener can safely pass 17mA continuously. In temporary (transient) bypass mode it will pass morem but I don't know how many transients it can pass until the Zener starts deteriorating.
That's why I suggested adding on a CCS pre-reg. This will limiting the Zener peak current and thus extend the life of the Zener.

You could use a 1W3 20V Zener for 65mA continuous duty current capability.

That is "WHY."
 
OK.

Lets get back to the problem of oscillation.

Initially I'm going to reduce the gain of the Pumpkin from 10X to a more realistic (in my application) 3X.

As components on these double sided boards are difficult ro move I intend to simply solder a 47K across R5 and R6.
 

Attachments

  • 001.jpg
    001.jpg
    818 KB · Views: 162
OK. Gain has been dropped to 2.7X I managed to replace R5 and R6 with 27K.

The Pumpkin still operates like a perfect 20MHz oscillator with over 20mV pk-pk of pure 20MHz sine wave.

I've checked all the components, my soldering and for obvious copper bridges. I can't see anything amiss.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.